id
stringlengths 3
9
| source
stringclasses 1
value | version
stringclasses 1
value | text
stringlengths 1.54k
298k
| added
stringdate 1993-11-25 05:05:38
2024-09-20 15:30:25
| created
stringdate 1-01-01 00:00:00
2024-07-31 00:00:00
| metadata
dict |
|---|---|---|---|---|---|---|
236095472
|
pes2o/s2orc
|
v3-fos-license
|
Annexin A3, a Calcium-Dependent Phospholipid-Binding Protein: Implication in Cancer
Annexin A3 (ANXA3), also known as lipocortin III and placental anticoagulant protein III, has been reported to be dysregulated in tumor tissues and cancer cell lines, and harbors pronounced diagnostic and prognostic value for certain malignancies, such as breast, prostate, colorectal, lung and liver cancer. Aberrant expression of ANXA3 promotes tumor cell proliferation, invasion, metastasis, angiogenesis, and therapy resistance to multiple chemotherapeutic drugs including platinum-based agents, fluoropyrimidines, cyclophosphamide, doxorubicin, and docetaxel. Genetic alterations on the ANXA3 gene have also been reported to be associated with the propensity to form certain inherited, familial tumors. These diverse functions of ANXA3 in tumors collectively indicate that ANXA3 may serve as an attractive target for novel anticancer therapies and a powerful diagnostic and prognostic biomarker for early tumor detection and population risk screening. In this review, we dissect the role of ANXA3 in cancer in detail.
INTRODUCTION
Annexin A3 (ANXA3), a water-soluble protein consisting of 323 amino acid residues, is encoded by the ANXA3 gene located on human chromosome 4q13-q22 (Mussunoor and Murray, 2008). Also known as lipocortin III and placental anticoagulant protein II, ANXA3 subordinate to the annexin family, which is a well-characterized multigene family of structurally homologous, but functionally different calcium-dependent membrane phospholipid-binding proteins ubiquitously distributed in a wide array of cell types (Moss and Morgan, 2004).
The annexin family can be divided into five classes (A-E) based on their biological origins; among them, 12 subtypes of class A annexins, annexins A1 to A11 and A13, are derived from human and vertebrate orthologues, while class B till E are originated from non-vertebrate metazoans, fungi and molds, plants and protists respectively (Gerke and Moss, 2002). ANXA3 has been demonstrated to be virtually solely expressed in differentiated myeloid cell lines and accounts for about 1% of the cytosolic protein of human neutrophils (Sopkova et al., 2002). To date, two isoforms of ANXA3 with the molecular mass of 33 and 36 kDa have been documented. The 36 kDa ANXA3 isoform is mainly expressed in monocytes, while the 33 kDa ANXA3 isoform is more abundantly observed in neutrophils (Le Cabec and Maridonneau-Parini, 1994).
Four conserved annexin repeats structural domains (I-IV) constitute the C-terminal protein core of ANXA3 (Sopkova et al., 2002) (Figure 1). Each structural domain encompasses five a-helices (A-E) consisting of 70 amino acid residues (Favier- Perron et al., 1996;Gerke and Moss, 2002). In addition, one principal calcium-binding site is present on the convex face formed by a-helices A and B in each structural domain (Raynal and Pollard, 1994;Mussunoor and Murray, 2008). The highly variable N-terminus of 20 amino acids in length contains two tryptophan residues, which are W5 in the N-terminal segments and W190 at the end of the IIIA-IIIB loop (Sopkova et al., 2002). These two tryptophan residues are essential for the protein stability as well as the interaction of ANXA3 with intracellular calcium ions and negatively charged phospholipids, thereby regulating a diverse range of biological functions of ANXA3 (Hofmann et al., 2000;Sopkova et al., 2002). Due to its phospholipid-binding capacity and calmodulindependent nature, ANXA3 primarily participates in membrane-associated activities, such as intracellular and extracellular signal transduction, vesicular transport, membrane fusion and endocytosis, formation and transport of ion channels, and interactions of cytoskeleton proteins (Swairjo and Seaton, 1994;Perron et al., 1997;Mussunoor and Murray, 2008). The involvement of ANXA3 in cellular signal transduction facilitates its multifaceted regulatory roles in various physiological activities, including cell division, differentiation, motility and apoptosis as well as anti-inflammation, anticoagulation and angiogenesis (Moss and Morgan, 2004;Mussunoor and Murray, 2008). Meanwhile, dysregulation of ANXA3 has been reported to play a pivotal role in cancer development and progression (Mussunoor and Murray, 2008) ( Table 1 and Figure 4). However, data published so far about its expression in different malignancies are inconsistent. To the best of our knowledge, ANXA3 has been reported to be overexpressed in a majority cancer types including breast (Pendharkar et al., 2016;Zhou et al., 2017a;Zhou et al., 2017b;Guo et al., 2017;Aravind Kumar et al., 2018;Du et al., 2018;Kim et al., 2018;Li et al., 2018;Zhou et al., 2018), colorectal (Madoz-Gurpide et al., 2006;Marshall et al., 2010;Yip et al., 2010;Yang L. et al., 2018;Yang Q. et al., 2018;Xu et al., 2019), bladder (Tsai et al., 2018), ovarian (Jiang et al., 2019), gastric (Takahashi et al., 2015;Wang and Li, 2016) and pancreatic cancer (Baine et al., 2011a;Baine et al., 2011b;Wan et al., 2020) as well as hepatocellular (Pan et al., 2015a;Pan et al., 2015b;Tong et al., 2018) and nasopharyngeal carcinoma (Ruan et al., 2010), while downregulated in renal (Bianchi et al., 2010), prostate (Wozny et al., 2007;Köllermann et al., 2008;Peraldo-Neia et al., 2011) and papillary thyroid cancer (Jung et al., 2010). Furthermore, the expression of ANXA3 in lung cancer remains controversial, with the evidence of both upregulated (Liu et al., 2009;Gy} orffy et al., 2013;Wang et al., 2019;Jin et al., 2020;Liu et al., 2021) and downregulated (Rho et al., 2009;Wu et al., 2018;Lohinai et al., 2019) expression patterns documented in the literature. In an immunohistochemistry-based study of organotypic ex vivo human HCC clinical samples and HCC patient-derived xenografts, Tong et al. found that overexpression of ANXA3 was associated with enhanced resistance to sorafenib and led to poor survival of HCC patients receiving sorafenib treatment. Their data further indicates that targeting ANXA3 could effectively inhibit tumor growth and sensitize the response of tumor cells to sorafenib treatment (Tong et al., 2018). In addition, ANXA3 mRNAs and proteins were overexpressed in gastric cancer tissues and various gastric cancer cell lines, as detected by RT-PCR and Western blot analyses (Wang and Li, 2016). This aberrant expression was further correlated with the depth of tumor infiltration and TNM stage in both univariate and multivariate analyses of a cohort of 183 gastric cancer patients, which indicates the potential of ANXA3 as an independent prognosticator for the survival of gastric cancer patients (Wang and Li, 2016). Likewise, markedly elevated ANXA3 expression was detected in bladder cancer by multiplexed liquid chromatography multiple-reaction-monitoring mass spectrometry assay (LC-MRM-MS). Investigators from this study further suggested that ANXA3 might serve as a reliable non-invasive diagnostic biomarker for bladder cancer (Tsai et al., 2018). Similarly, ANXA3 was overexpressed in colorectal cancer (CRC) tissues compared to adjacent normal tissues, as shown from immunohistochemistry and western blot results (Yang Q. et al., 2018). Moreover, a Max Vision immunohistochemistrybased retrospective analysis of a cohort of 309 breast cancer patients demonstrated that ANXA3 expression in triple negative breast cancer (TNBC) patients was significantly higher than other breast cancer subtypes (Zhou et al., 2017b). Given that overexpression of ANXA3 has a vital impact on tumor progression, we could expect that downregulation of ANXA3 can also exert certain regulatory effects on tumorigenesis. Interestingly, ANXA3 expression level was diminished in prostate tumor tissues and was correlated with increasing pathological stages and Gleason scores (Köllermann et al., 2008). Immunohistochemistry and tissue microarray data further confirmed that ANXA3 could be used as an independent prognostic factor to predict the survival of prostate cancer patients and to support population risk stratification (Köllermann et al., 2008). Downregulation of ANXA3 was also reported in papillary thyroid cancer (PTC), and PTC patients with decreased ANXA3 expression exhibited substantially elevated lymph node metastasis scores and tumor growth (Jung et al., 2010).
Collectively, aberrant expression of ANXA3 plays a crucial role in malignant tumor development. It stimulates tumor cell proliferation, facilitates invasion, migration and metastasis, induces angiogenesis, desensitizes patient response to antitumor treatments and predisposes the emergence of certain inherited familial tumors (Gerke and Moss, 2002;Moss and Morgan, 2004;Mussunoor and Murray, 2008;Tong et al., 2018;Sarquis et al., 2020). Therefore, it is important to shedding Frontiers in Molecular Biosciences | www.frontiersin.org July 2021 | Volume 8 | Article 716415 light on the functions of ANXA3 in tumor biology in order to improve the early detection of preneoplastic tumors, to overcome anticancer therapy resistance and to develop novel, targeted approaches to treat solid tumors. This review focusses on the roles of ANXA3 in cancer.
Sustaining Proliferative Signaling
Sustaining proliferative signaling is a common strategy used by cancer cells to facilitate their progression and aggressiveness. Recent years, the pro-proliferative role of ANXA3 has been substantiated to be heavily implicated in various types of malignancies, and a diversity of evidence has been provided about the underlying mechanisms mediating this process ( Figure 2). In hepatocellular carcinoma, ANXA3 has been shown to activate the Notch and MAPK/ERK/JNK signaling pathway, resulting in enhanced cell proliferation and promotion of stem-cell like characteristics (Pan et al., 2015a;Tong et al., 2015). Support on this finding was delivered by in vitro data of colorectal cancer, in which the phosphorylation of ERK and JNK was found to be significantly reduced once a depletion of ANXA3 was established using small interfering RNA (siRNA) (Xu et al., 2019). This finding was further corroborated by another study on chemoresistant non-small cell lung cancer (NSCLC) cells, which showed that high level of ANXA3 secreted by cancer associated fibroblasts (CAFs) in the tumor microenvironment activated the JNK/survivin signaling, thereby helping cancer cells escaping the cisplatin-induced apoptosis (Wang et al., 2019). Another putative mechanism contributing to the pro-proliferative effect of ANXA3 was provided by the work of Wan et al., in which PI3K/Akt signaling pathway was found to be substantially inhibited in pancreatic cancer patients with overexpression of miR-382, a miRNA that suppresses the expression of the ANXA3 gene (Wan et al., 2020). Investigators in this study further observed a decline of clone formation ability and proliferative behavior in pancreatic cancer cells overexpressing miR-382. Interestingly, a research on breast cancer has described an opposite correlation; the depletion of ANXA3 using short hairpin RNA plasmids has been shown to promote cell proliferation in both cell-line models and mouse xenograft models (Du et al., 2018). However, a series of studies on breast cancer challenged this finding. Collectively, these studies demonstrated that ANXA3 is highly expressed in luminal A, B and triple negative breast cancer subtypes, and that ANXA3 inhibition could significantly impair tumor growth in vivo, concomitant with a lower proliferation index and a higher apoptosis rate and G0/1 cell count in vitro (Zhou et al., 2017a;Li et al., 2018;Zhou et al., 2018). Aside from promoting the pro-proliferative pathways, aberrant expression of ANXA3 has also been shown to downregulate multiple pro-apoptotic proteins and cyclindependent kinases (CDKs), which facilitates the evasion of apoptosis and cell cycle arrest. For example, in vitro investigations on breast cancer cell lines HCC-1954 and MDA-MB-231 demonstrated that ANXA3 silencing using siRNA significantly reduced the expression of CDK4 and enhanced the expression of E2F1 and p27 Kip1 (Kim et al., 2018). Furthermore, high levels of ANXA3 in hepatocellular carcinoma has been demonstrated to attenuate the PKCδ/p38 associated apoptosis (Tong et al., 2018). Of note, p38 not only plays a role in the regulation of apoptosis, but also acts a key regulator in autophagy, which is another defining feature of tumor cells that supplies metabolic fuel sources for unlimited proliferation (Webber, 2010;Webber and Tooze, 2010). Unsurprisingly, investigators from this study further detected a substantially increased level of autophagic marker LC3B in both HCC cells in vitro and mouse xenografts in vivo, thereby confirming the positive correlation between ANXA3 expression and autophagic activity (Tong et al., 2018). The anti-apoptotic ability of ANXA3 was further substantiated by the work of Wang et al., wherein pro-apoptotic proteins caspase 3 and caspase 8 was significantly downregulated in NSCLC cells overexpressing ANXA3 (Wang et al., 2019). Conversely, miRNAinduced silencing of ANXA3 markedly upregulated caspase 3 expression as well as the expression of pro-apoptotic protein Bax, while suppressing the expression anti-apoptotic protein Bcl-2 (Liu et al., 2021). Consistent data were also published by Xu et al., who observed an ANXA3 knockdown-induced upregulation of c-caspase 3 and c-PARP in colorectal cancer cells (Xu et al., 2019).
Promoting Invasion and Metastasis
Invasion and metastasis are the major cause of poor clinical outcomes of malignant diseases. In vitro investigations have revealed that ANXA3 overexpression significantly stimulated the invasion and migration of breast cancer cells (Ibrahim et al., 2012;Guo et al., 2017;Kim et al., 2018). Clinically, ANXA3 overexpression has been demonstrated to be correlated with the occurrence of lymph node metastasis and the clinicopathological stages of breast cancer (Zhou et al., 2017b) and lung adenocarcinoma (Liu et al., 2009). The correlation was further reversely validated by the work of Zhou et al., in which knockdown of ANXA3 by shRNA impaired the invasion and (2) inhibiting PKCδ/p38 pathway leading to decreased apoptosis and increased autophagy; (3) activating JNK/survivin and Raf/ERK/c-myc pathways leading to increased proliferation; (4) activating PI3K/Akt/mTOR pathway leading to increased EMT and decreased apoptosis; and (5) activating HIF-1α/Notch pathway leading to increased proliferation and decreased apoptosis. p38, p38 mitogen-activated protein kinases; PARP, Poly (ADP-ribose) polymerase; bax, Bcl-2-associated X protein; bcl-2, B-cell lymphoma 2; IκBα, nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha; NF-κB, nuclear factor kappa-light-chainenhancer of activated B cells; HIF-1α, Hypoxia-inducible factor 1-alpha; EMT, Epithelial-Mesenchymal Transition; TNF, Tumor necrosis factor.
Frontiers in Molecular Biosciences | www.frontiersin.org July 2021 | Volume 8 | Article 716415 migration abilities of luminal A and triple negative breast cancer cells (Zhou et al., 2017a). Similar approaches employing miRNA and siRNA to probe the effect of ANXA3 depletion have also been applied in in vitro cell-line models of colorectal carcinoma (Yang L. et al., 2018;Xu et al., 2019) and ovarian carcinoma (Jiang et al., 2019), which presented consistent results. Intriguingly, an inverse relationship between ANXA3 expression and the execution of invasion-metastasis cascade in malignant tumors has also been documented. In surgical tumor specimens from 25 thyroid papillary cancer patients receiving thyroidectomy, downregulation of ANXA3 was detected in tumor tissues compared to the adjacent non-tumor tissues (Jung et al., 2010). Further immunohistochemistry results showed that reduced ANXA3 staining was correlated with thyroid papillary tumors with higher lymph node metastasis scores and larger sizes (Jung et al., 2010). Nonetheless, no data are currently available about what factors govern the downregulated expression of ANXA3 and its potentially anti-oncogenic activities in thyroid neoplasms.
A number of attempts has been made to decipher the underlying molecular changes mediating the ANXA3-induced invasion and migration. In ANXA3 overexpressing gastric cancer cells and patient-derived tumor specimens, an enhanced degree of epithelial-mesenchymal transition (EMT) was observed, which was evidenced by western blot results indicating increased expressions of mesenchymal markers vimentin and ß-catenin, a decreased expression of epithelial marker E-cadherin, and increased expressions of EMT-related transcription factors fibronectin, Slug and Snail (Wang and Li, 2016). Contrariwise, silencing of ANXA3 inhibited the expression of N-cadherin and vimentin in pancreatic cancer, while elevating the expression of E-cadherin (Wan et al., 2020). Additionally, ANXA3-knockdown pancreatic cancer cells exhibited decreased expressions of VEGF-C and VEGF-D (Wan et al., 2020), both of which are proteins previously shown to be positively associated with the number of lymph node metastases (Schulz et al., 2011). Similar findings were reported by research involving triple negative breast cancer cell lines MDA-MB-231 and MDA-MB-486, which found mesenchymal-epithelial transition (MET) in cancer cells receiving ANXA3-targetting shRNA plasmids, evidenced by decreased mesenchymal markers (vimentin and N-cadherin) and increased epithelial markers (E-cadherin and c-cadherin) (Du et al., 2018). This reversed transition pattern could be considered as an indicator of diminished invasion and migration abilities. Interestingly, this study further indicated that IκBα knockdown could partially reverse the ANXA knockdown-induced MET state concomitant with an increased phospho-NF-κB p65 expression. These findings suggest that NF-κB signaling might be crucially involved in the ANXA3medicated tumor cell invasion and migration (Du et al., 2018).
Inducing Angiogenesis
Angiogenesis is the essential mechanism that allows continuous nutrients supply and assists tumor cells to combat hypoxia in the tumor microenvironment. Over the last decade, a series of studies has identified novel clues shedding light on the functional link between ANXA3 and tumor angiogenesis. At histological level, less blood vessels were observed in H&E-stained tissue slides of triple negative breast cancer xenografts from mice pretreated with ANXA3-silencing shRNA plasmids . At molecular level, ANXA3 silencing in pancreatic cancer cells resulted in a decrement of the expression of vascular endothelial growth factor receptor 3 (VEGFR3), which is a protein suggested to play a key role in lymphatic vascularization in pancreatic cancer (Wan et al., 2020). In addition, investigators from this study observed an ANXA3-induced upregulation of PI3K/Akt signaling pathway, which is a canonical pathway also capable of promoting neovascularization apart from its pro-proliferative effects (Sun et al., 2016;Wan et al., 2020). The hypoxia-inducible factor-1α (HIF-1α)/VEGF pathway is another well-established response of cancer cell to initiate angiogenesis and thereby survive hypoxia Zhang et al., 2018;Zhang P.-C. et al., 2020). Employing liver cancer stem-like cells, Pan et al. revealed a positive correlation between the expressions of ANXA3 and HIF1α, which further confirmed the pro-angiogenic role of ANXA3 (Pan et al., 2015a). Similar findings were also presented by a research on bone-cancer induced pain (BCP), showing that downregulation of ANXA3 using shRNA substantially inhibited the expression of HIF1α and VEGF in the ipsilateral spinal cord and microglial cells of mice undertaken 21 days of bone cancer induction (Zhang Z. et al., 2020). Besides cancer-related investigations, fundamental studies using human umbilical vein endothelial cell (HUVEC) model provided extra confirming data (Park et al., 2005;Mohr et al., 2017). Employing the hypoxia responsive element (HRE)-luciferase reporter assay, investigators demonstrated that ANXA3 upregulated the HIF1α transactivation activity and VEGF production, give rise to enhanced tube formation and migration of HUVECs (Park et al., 2005). These findings suggest that ANXA3 might act as a key driver in angiogenic processes.
Paradoxically, a negative relationship between the expressions of ANXA3 and HIF1α has been reported in renal cell carcinoma (RCC) (Bianchi et al., 2010). Whereas RCC cultures not expressing HIF1α (HIF1α-negative) exhibited a similar level of ANXA3 expression as the matched noncancerous cortex, a markedly decreased ANXA3 expression was detected in HIF1α-positive RCC cultures compared to the matched counterparts (Bianchi et al., 2010). Moreover, the abundance of 36-kDa ANXA3 was significantly reduced in HIF1α-positive RCC cultures, while the 33-kDa ANXA3 showed a pronounced increment (Bianchi et al., 2010). Most importantly, the total expression of ANXA3 protein was significantly lower in RCC cultures in vitro as well as in surgical excised RCC tissues compared to their paired counterparts (Bianchi et al., 2010). It was speculated that the downregulation of the total ANXA3 protein in RCC was associated with the decrement of its 36-kDa isoform, as the N-terminus of ANXA3 capable of promoting self-expression is present in the 36-kDa isoform but absent in the 33-kDa isoform (Hofmann et al., 2000;Gerke et al., 2005;Bianchi et al., 2010). However, the mechanism underlying the diminution of 36-kDa ANXA3 in RCC remains unclear, and the differences between the two isoforms regarding their expressions and functions in other cancer models are still far from fully understood.
Genetic Predisposition
Genetic predisposition usually serves as a prerequisite in cancer initiation and harbors pronounced clinical relevance especially in aggressive malignancies where an urgent need exists for novel risk screening methods with good predictive performance and clinical utility. Cancer-associated mutations can not only drive accelerated cancer progression but can also exhibit inherited patterns, contributing to familial hereditary tumors (AlHarthi et al., 2020). So far, very limited evidence has been published reporting the association between genetic alterations of the ANXA3 gene with cancer susceptibility. In a case-control study of 29 TNBC patients searching for risk-associated SNPs through microarray-based whole genome SNP sequencing, the non-synonymous SNP exm4087722 was detected in the ANXA3 gene (NM_005139) with a minor allele frequency (MAF) smaller than 0.05 (Aravind Kumar et al., 2018). Whole genome sequencing was also applied in another research on 3 Brazilian families with hereditary papillary thyroid cancer (PTC), which found point mutation p.D283N on the ANXA3 gene (MIM No.106490) (Sarquis et al., 2020). Yet, it is important to note that above observations are both based on small sample sizes and that more conclusive data from larger cohorts are still awaited about the role of genetic alterations of the ANXA3 gene in cancer susceptibility.
To generate a more comprehensive picture of the potentially oncogenic alterations on the ANXA3 gene, a computational analysis was performed using cBioPortal (http://www. cbioportal.org) querying 10,953 patients/10967 samples from 32 TCGA PanCancer Atlas studies (June 2021). Overall, the gene exhibited a low alteration frequency, with a detectable alteration in 121 (1.1%) of the total queried patients. The top three malignant diseases ( Figure 3A) with the highest ANXA3 gene alterations are cervical adenocarcinoma (4.35% alterations including 2.17% mutation and 2.17% deep deletion), endometrial carcinoma (2.73% alterations including 2.56% mutation and 0. 17% structural variant), and melanoma (2.48% alterations including 2.25% mutation and 0.23% amplification). Furthermore, the largest number of mutations on the ANXA3 gene (4 registered cases) occurred at amino acid 288 on its fourth annexin repeat domain, which could be either a nonsense mutation or a missense mutation replacing the arginine into glutamine (R288*/Q) ( Figure 3B). Cancer types in which these mutations were detected are glioblastoma multiforme, uterine endometrioid carcinoma, and breast invasive ductal carcinoma. However, none of the mutations was further annotated with documented clinical implications from the literature, indicating that our understanding of their oncogenic effects is currently still highly rudimentary.
THE ROLE OF ANXA3 IN ANTICANCER DRUG RESISTANCE
Chemotherapy is the standard-of-care treatment for a majority types of tumors, yet its efficacy is frequently plagued by the evolution of fatal drug resistance or temporary, fluctuating, or partial drug responses in patients. Based on the previously described tumorigenic characteristics of ANXA3, it is wellreasoned to hypothesize that ANXA3 also play a role in attenuating the vulnerability of cancer cells against chemotherapeutic drugs. As a result, in the past decade, investigators in this field have generated a substantial body of evidence delineating the involvement of ANXA3 in the development of anticancer drug resistance ( Table 2 and Figure 4). Platinum resistance. ANXA3 has been previously described as a potential marker for platinum sensitivity in patients with hepatocellular carcinoma (Pan et al., 2015b), ovarian carcinoma (Yan et al., 2010;Yin et al., 2012), non-small cell lung carcinoma (Wang et al., 2019;Jin et al., 2020) and colorectal carcinoma (Xu et al., 2019). In an attempt to ascertain the mechanism of platinum resistance in patients with hepatocellular carcinoma, researchers observed an enhanced degree of resistance to cisplatin-induced cell death in ANXA3-overexpressing tumor cells in vitro and in engrafted mice in vivo (Pan et al., 2015b). Support on this finding was provided by Yan et al., in which a significantly lower intracellular accumulation and DNA binding of cisplatin and carboplatin was detected in ANXA3 overexpressing ovarian cancer cells, accompanied by a decrement of the intracellular p53 level (Yan et al., 2010). Likewise, a substantially increased ANXA3 expression has been found in the serum of platinum-resistant ovarian cancer patients compared to the platinum sensitive group, which underscored the value of ANXA3 as a noninvasive marker for drug response prediction (Yin et al., 2012). Mechanistically, high level of ANXA3 released by CAFs has been confirmed to activate the JNK/survivin pathway in both cell-line models (A549, H661 and SK-MES-1) and mouse xenograft models of NSCLC, resulting in a markedly attenuated IC50 of cisplatin in vitro and an augmented xenograft tumor growth under cisplatin exposure in vivo (Wang et al., 2019). Conversely, abrogation of oxaliplatin resistance was achieved with ANXA3-silencing siRNAs in A549 NSCLC cells (Jin et al., 2020). Similarly, in vitro observations in oxaliplatin-resistant HCT116 and SW480 colorectal cancer cells further evidenced that ANXA3knockdown has significantly impeded the phosphorylation of ERK and JNK, which led to decreased cell viability and BrdU incorporation as well as increased apoptosis and impaired migration and invasion abilities (Xu et al., 2019).
TKIs resistance. In a microarray analysis testing the sensitivity of 45 different cancer cell lines to anticancer tyrosine kinase inhibitors (TKIs) targeting the ERBB/RAS pathway, ANXA3 was indicated to be associated with resistance against multiple TKIs including gefitinib, sorafenib, sunitinib, and lapatinib (Pénzváltó et al., 2013). Among these TKIs, only sorafenib resistance possesses a comprehensive in vitro and clinical evaluation, which is provided by the work of Tong et al. In this study, ANXA3 overexpression was detected in sorafenib-resistant HepG2 and Huh7 hepatocellular carcinoma cells as well as in patient-derived xenografts; conversely, inhibition of ANXA3 achieved with shANXA3 re-sensitized the response to sorafenib in vitro and limited tumor growth in vivo (Tong et al., 2018). The mechanistic rationale underlying this phenomenon was the inhibition of PKCδ/p38-mediated apoptosis and activation of p38-mediated autophagy, both of which are driven by a substantially elevated ANXA3 expression (Tong et al., 2018). Clinically, HCC patients with high levels of ANXA3 exhibited inferior outcomes under sorafenib treatment (Tong et al., 2018). Interestingly, investigators from this study further demonstrated that the combination of sorafenib with anti-ANXA3 monoclonal antibody effectively overcomes sorafenib resistance in both ex vivo organotypic cultures and patient-derived mouse xenografts (Tong et al., 2018), suggesting that targeting ANXA3 might be an actionable therapeutic approach for sorafenib-refractory HCC patients. Fluoropyrimidine resistance. The involvement of ANXA3 in 5-fluorouracil (5-FU) resistance has been previously reported in hepatocellular carcinoma, in which enforced ANXA3 expression established by ANXA3-overexpressing lentiviruses significantly enhanced the IC50 of 5-FU in both cell-line models and mouse xenografts models (Pan et al., 2015b). Furthermore, a bioinformatics analysis on the genomic data of 119 fluoropyrimidine-treated gastric cancer patients reported that SNP rs2867461 in the ANXA3 gene exhibited a significant correlation with the sensitivity to fluoropyrimidine treatment and might therefore serve as a potential genetic biomarker in predicting the therapeutic response (Takahashi et al., 2015).
Frontiers in Molecular Biosciences | www.frontiersin.org July 2021 | Volume 8 | Article 716415 wild type PC3 cell line (Thoenes et al., 2010). Moreover, western blot analysis validated the CPA-induced upregulation of ANXA3 in the isolated tumor tissues from mice that were engrafted with PC3-wt cells and subsequently exposed to a 70-day of metronomic CPA treatment (Thoenes et al., 2010). This finding is further consistent with the data from another microarray-based study on CPA-resistant PC3 tumors (Kubisch et al., 2013). Doxorubicin and docetaxel. The knockdown of ANXA3 has been demonstrated to promote the uptake of doxorubicin in human MDA-MB-231 breast cancer cells and 4T1 mouse mammary cancer cells (Du et al., 2018). MTT assay data from this study further indicated a substantially increased sensitivity of ANXA3-knockdown breast cancer cells to doxorubicin and docetaxel (Du et al., 2018). The reduction of the NF-κB signaling activity, which was shown to be a result of ANXA3knockdown, was thought to be a putative mechanism underlying this improved drug response (Du et al., 2018). Despite these findings, current data is extremely limited regarding the role of ANXA3 in drug resistance against antineoplastic antibiotics and taxanes, which indicates the pressing need for further explorations in this field.
THE DIAGNOSTIC AND PROGNOSTIC VALUE OF ANXA3 IN CANCER TREATMENT AND MANAGEMENT
ANXA3 is one of the candidate biomarkers that has been extensively studied in different cancer models over the last decade. Collectively, studies have demonstrated that ANXA3 possesses potential diagnostic and prognostic value in the clinical treatment and management of an array of malignancies, including breast cancer (Pendharkar et al., 2016;Zhou et al., 2017b;Kim et al., 2018;Zhou et al., 2018), prostate cancer (Wozny et al., 2007;Köllermann et al., 2008;Schlomm et al., 2010;Guo et al., 2020), colorectal cancer (Marshall et al., 2010;Yip et al., 2010;Chang et al., 2014;Yang Q. et al., 2018) and some relatively rarer cancers such as bladder cancer (Hofmann et al., 2000), pancreatic cancer (Baine et al., 2011b) and testis cancer (Hofmann et al., 2000). In this section, we will summarize data from various studies which evaluated the clinical value of ANXA3 or provided novel clues on its diagnostic and prognostic potential.
Breast cancer. Currently, stratifying patients into their correct subtypes and discriminating between early and late stages of these subtypes remain as a prime challenge in breast cancer (Pendharkar et al., 2016). This is of considerable importance for an optimal treatment design and patient outcome, especially in those patients with aggressive phenotypes such as triplenegative and basal-like breast cancer. Moreover, multiple shortcomings of the current mammography screening method have been reported, with the main concern remains on its suboptimal sensitivity in young women with dense breasts (Arif et al., 2015). Interestingly, the expression level of ANXA3 has been reported to be different per breast cancer subtype (Pendharkar et al., 2016;Zhou et al., 2017b;Kim et al., 2018;Zhou et al., 2018). Specifically, ANXA3 expression in basal subtype of breast cancer (MDA-MB-231, HCC-70, HCC-1954) was found to be significantly higher than other subtypes (Kim et al., 2018). Other studies further confirmed that a significantly higher expression of ANXA3 was detect in TNBC cells in comparison to luminal A and B subtypes (Zhou et al., 2017b;Zhou et al., 2018). Furthermore, employing 2D-DIGE and iTRAQ approaches, Pendharkar et al. demonstrated that ANXA3 upregulation is a marker for differentiating invasive ductal carcinoma with luminal B HER2 positive (LB) and HER2 enriched (HE) subtypes as well as their early and late stages (Pendharkar et al., 2016). These findings collectively confirmed the potential of ANXA3 in improving the accuracy of subtypes differentiation and risk stratification of breast cancer.
Prostate cancer. The low explanatory power of existing clinical and histological parameters has also been reported in prostate cancer, where prognosis prediction relies solely on clinical stages, Gleason score and serum PSA and no molecular marker has been successfully translated into the routine clinical applications to date (Köllermann et al., 2008;Schlomm et al., 2010). The inferior performance of serum PSA in screening small neoplasms of initial stages (Köllermann et al., 2008) and the patient inconvenience from periodic biopsy sampling for Gleason score and pathological stage evaluation further emphasize the pronounced clinical value of novel molecular biomarkers in prostate cancer. ANXA3 has been demonstrated to be downregulated in prostate cancer tissues (Wozny et al., 2007;Köllermann et al., 2008;Peraldo-Neia et al., 2011) but upregulated in the chemo-resistant PC3-D3 and PC3-D4 cell lines (Thoenes et al., 2010;Kubisch et al., 2013). Immunohistochemistry-based studies have reported a substantially less abundant ANXA3 staining in prostate cancer tissue when comparing with the surrounding epithelium and prostatic intraepithelial neoplasia (Wozny et al., 2007;Köllermann et al., 2008). In addition, the proportion of the ANXA3-negative tissue has been demonstrated to be correlated with advanced pathological stage and Gleason score as well as a reduced PSA-free survival (Köllermann et al., 2008). Using Kaplan-Meier analysis and multivariate cox regression, the study further indicated a significant association between ANXA3 staining abundance and biochemical relapse of prostate cancer (Köllermann et al., 2008). Support on this finding was provided by the work of Schlomm et al., in which ANXA3 was identified as an independent prognostic marker for the postoperative PSA recurrence (Schlomm et al., 2010). Of note, Schlomm et al. further demonstrated that incorporation of ANXA3 into the current risk stratification nomogram predicting the biochemical relapse after radical prostatectomy provided an enhancement of its predictive performance (AUC from 0.71 to 0.73) (Schlomm et al., 2010). Aside from postoperative recurrence risk, the value of ANXA3 in discriminating prostate cancer from benign tumors has also been elucidated by the work Guo et al., in which a 14-gene panel involves ANXA3 was constructed and evaluated on pre-biopsy urine and tissue specimens, displaying desirable predictive performance (AUC 0.854) . Further evaluation of this gene predictor set on prospective and retrospective cohorts confirmed its ability in distinguishing low-risk patients from high-risk patients . Unlike a wide spectrum of cancers overexpressing ANXA3, it is notable that the protein is downregulated in prostate cancer tissues. The putative mechanism underlying this downregulated expression pattern has been proposed in the context of autoimmune reactions in prostate cancer (Köllermann et al., 2008). Autoimmune antibodies against ANXA3 have been previously detected in the serum of prostate cancer patients (Köllermann et al., 2008), which was thought to be a result of the release of ANXA3 by prostatic epithelium cells in the form of prostasomes (Wozny et al., 2007;Köllermann et al., 2008). These granules, which are secreted by both normal and malignant prostate epithelium, have been broadly considered to harbor a strong immunogenicity (Feist et al., 2000;Minelli et al., 2005;Larsson et al., 2006). Studies have shown that, compared to patients with benign prostate hyperplasia and other noncancerous prostate disorders, prostate cancer patients carried significantly higher levels of antiprostasome antibodies in their blood (Feist et al., 2000;Minelli et al., 2005;Larsson et al., 2006). This finding has linked anti-prostasome antibody titer to malignant transformation. It was therefore postulated that some prostate tumors could induce the formation of autoimmune antibodies against ANXA3 to neutralize its tumor-suppressive effects (Köllermann et al., 2008). This speculation also explained why decreased ANXA3 expression is associated with unfavorable clinicopathological features in prostate cancer patients (Köllermann et al., 2008). Nevertheless, the functional rationale of the antitumor activity of ANXA3 in prostate cancer, which is inconsistent to its tumorigenic roles in other malignancies, still necessitates additional experimental analyses.
Colorectal cancer. Colorectal cancer (CRC) is another wellknown malignancy that can benefit most from early detection via population screening, since almost every colorectal tumor arises from a benign adenomatous polyp and is easily surgically resectable once detected (Marshall et al., 2010;Yang Q. et al., 2018). However, the unpleasant and inconvenient nature of current CRC screening procedures has led to a low compliance in screening participation (Kronborg, 2004;Marshall et al., 2010), which results in a higher incidence of advanced tumors and metastases. In an attempt searching for novel tumor-specific immunogens in CRC, mass spectrometry analyses have reproducibly detected the presence of ANXA3 in the tumor protein extracts blotted with patient's own sera (Yang Q. et al., 2018). Further verifications using immunohistochemistry and western blot assays confirmed the increased ANXA3 expression in tumor tissues compared to nontumor tissues (Yang Q. et al., 2018). These findings collectively indicated the potential of ANXA3 as a novel serum antibody screening marker in CRC. Additionally, a series of microarraybased studies have designed novel gene predictor sets that can potentially contribute to improving the performance of the existing screening methods and overcoming the current low screening compliance (Marshall et al., 2010;Yip et al., 2010;Chang et al., 2014). A blood-based 7-gene panel (ANXA3, CLEC4D, LMNB1, PRRG4, TNFAIP6, VNN1, and IL2RB) developed by Marshall et al. exhibited good predictive performance in discriminating CRC from controls (AUC 0.80) and in stratifying patients' current relative risk (Marshall et al., 2010). This gene predictor set was further validated in a Malaysian cohort, showing consistent results (Yip et al., 2010). In a study published in 2014, Chang et al. evaluated 17 previously described CRC-associated genes (including those from Marshall et al.) and further constructed a novel bloodbased 7-gene model comprised of CpEB4, EIF2S3, MGC20553, MS4A1, ANXA3, TNFAIp6 and IL2RB, which showed a more superior performance in logistic regression analyses (AUC 0.99) (Chang et al., 2014).
Other cancers. With the hypothesis that peripheral blood mononuclear cells (PBMCs) act as the first line of defense against early emerged neoplastic cells and thereby capable of more accurately representing the tumor biology in initial stages, Baine et al. investigated the differentially expressed genes in PBMCs of patients with pancreatic cancer (PC) and found that ANXA3 was significantly upregulated in PC patients compared to healthy controls (Baine et al., 2011b). The study further established a blood-based 7-gene predictor panel (ANXA3, SSBP2, Ube2b-rs1, CA5B, F5, TBC1D8, ARG1, and ADAMTS20) with a sensitivity of 83% and specificity of 75% in discriminating PC patients (Baine et al., 2011b). In a LC−MRM/MS analysis of urine samples from 30 bladder cancer patients and 89 noncancer patients, ANXA3 was most frequently detected in bladder cancer samples (Tsai et al., 2018). Validation using western blot analysis further confirmed the elevated level of ANXA3 in both urine and tumor specimens from bladder cancer patients (Tsai et al., 2018). In another bioinformatics study based on previously published microarray data, a significantly augmented expression of the ANXA3 gene was detected in testicular carcinoma in situ (CIS) samples and further validated using RT-qPCR, suggesting the potential of ANXA3 serving as a novel clinicopathological marker for testicular CIS (Almstrup et al., 2007).
CONCLUSION
Cancer is a major public health issue and a highly lethal disease worldwide. The high mortality rate of many cancers can be attributed to the evasion of detection of benign or preneoplastic tumors, the occurrence of intrinsic or acquired therapy resistance, and the suboptimal predictive power of the routinely used screening techniques and clinicopathological parameters. Emerging evidence has suggested that ANXA3 might be a promising biomarker candidate and an attractive therapeutic target that harbors the potential to address these obstacles. According to the gathered data in this review, differential expression of ANXA3 has been demonstrated in a wide array of cancers, which is capable of sustaining cell proliferation signaling, promoting invasion and metastasis, inducing angiogenesis and resistance to various chemotherapeutic agents. Genetic alterations in the ANXA3 gene have also been reported to be associated with an enhanced tumor susceptibility. As a candidate molecular marker, ANXA3 exhibited pronounced clinical value in risk stratification, early detection, patient differentiation and active surveillance. Nonetheless, some limitations to the current findings and conclusions still need to be noted. Despite the substantial body of evidence on its tumorigenic role, the precise upstream and downstream signaling transduction of ANXA3 remains incompletely elucidated, and there is a lack of mechanistic studies on the role of ANXA3 in highly prevalent Frontiers in Molecular Biosciences | www.frontiersin.org July 2021 | Volume 8 | Article 716415 cancers. Besides, the underlying mechanisms of ANXA3-induced chemoresistance are not fully understood, and investigations so far have not covered all chemotherapeutic drugs, nor other anticancer therapies, such as radiotherapy. Furthermore, most of the studies evaluating the diagnostic and prognostic value of ANXA3 suffer from small sample sizes, which indicates that more conclusive data from larger patient cohorts are still urgently awaited. Therefore, the eventual clinical implementation of ANXA3 as a therapeutic target or as a diagnostic or prognostic biomarker still requires further investigations to 1) elucidate the complete picture of the upstream and downstream signaling pathways of ANXA3; 2) obtain deeper insights into the mechanisms underlying ANXA3-induced chemoresistance as well as the role of ANXA3 in other types of anticancer therapy resistance such as radioresistance; and 3) evaluate the clinical significance of ANXA3 as a diagnostic or prognostic biomarker on larger patient cohorts.
AUTHOR CONTRIBUTIONS
LY and PL wrote initial article, XY and KL provided helpful comments on this article and SQ conceived the idea and supervised the project. All the authors have read and approved the final version of this article. LY and PL contributed equally to this work as co-first authors.
|
2021-07-20T13:25:46.062Z
|
2021-07-20T00:00:00.000
|
{
"year": 2021,
"sha1": "b1db61d10960f228375972d6d8e343375bc30202",
"oa_license": "CCBY",
"oa_url": "https://www.frontiersin.org/articles/10.3389/fmolb.2021.716415/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "b1db61d10960f228375972d6d8e343375bc30202",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
231965735
|
pes2o/s2orc
|
v3-fos-license
|
Simulation of Rayleigh Bubble Growth near a No-Slip Rigid Wall
In order to study the role of growing cavitation bubbles in the context of ultrasonic cleaning, we perform two-dimensional, axisymmetric Navier-Stokes simulation for compressible, multicomponent flow and examine the so-called Rayleigh growth of an air bubble (with initial radius 33 µm and pressure 10 MPa) near a rigid wall. The simulation suggests that strong shear stress, which is important in physical cleaning such as particle removal, appears as a result of the bubble-growth-induced shock passage. The parametric study with varying a standoff distance of the bubble to the wall shows that the wall shear stress linearly decreases against the standoff distance.
Introduction
Ultrasonic cleaning is an essential process in semiconductor industry. Earlier studies [1,2] suggested that wall shear stress induced by dynamics of acoustic cavitation bubbles is a main factor to produce cleaning effects of underwater ultrasound. When it comes to studying acoustic cavitation bubbles near rigid boundaries, a main focus has been put on their collapse dynamics that can lead to strong wall shear stress [3,4]. Even though cavitation is expected to be nucleated heterogeneously from micron-sized or even smaller gas bubble nuclei pre-existing in the liquid bulk or at solid boundaries, wall shear stress generation from cavitation bubble nucleation (often called cavitation inception) has not been studied in the context of ultrasonic cleaning, to the author's knowledge. In this study, we aim to quantify wall shear stress as a result of the initial growth of a nucleated cavitation bubble. To do so, we perform two-dimensional, axisymmetric Navier-Stokes simulation for compressible, multicomponent flow, which allows for reproducing the bubble-growth-induced shock wave propagation and wall shear stress generation.
Physical and Numerical Modeling
The computational setup is depicted in Fig. 1. Since the scale separation between the size of cavitation bubble nuclei and the wavelength of ultrasound is very large, it is computationally expensive to fully resolve such bubble-ultrasound interaction. Here, for simplicity, we model the initial growth of a nucleated cavitation bubble as the so-called Rayleigh bubble growth [5]. The bubble growth is driven by pressure imbalance between the bubble and its ambient liquid. Rayleigh advocated the bubble collapse theory in the case of that when homogeneous fluid is at rest and spherical portion of the fluid is suddenly annihilated, the pressure at any point of the mass instantly alters, the pressure at an infinite distance being supposed to remain constant. We adapted and examined his proposed assumption to bubble growth rather than bubble collapse [6]. The bubble is assumed to consist of air that obeys the perfect gas law; the liquid is set initially at the standard temperature and pressure (T =20℃, ∞ =101.3 kPa). The initial pressure of the nucleated bubble is set at 0 =10 MPa and its density and temperature are determined according to the adiabatic relation with specific ratio =1.4. The initial radius of the nucleated bubble (at time =0) is set, as an example, at 0 =33 μm, which is expected to show large-amplitude bubble oscillation for the case of (10 kHz) ultrasound, and used to make length scales dimensionless. The standoff distance from the bubble center to the no-slip rigid wall (aligned with the axis at =0) is treated as a parameter and set at Vol. 314, Revised: 2020-08-12 doi: 10.4028/www.scientific.net/SSP.314.192 Accepted: 2020-08-12 © 2021 Trans Tech Publications Ltd, Switzerland Online: 2021-02-09 Figure 1: Problem setup of the bubble growth near a rigid wall, together with the computational domain and boundary conditions. The problem is axisymmetric with respect to axis. The bubble growth is induced by initial pressure imbalance 0 > ∞ where ∞ stands for one atmosphere in the liquid.
To numerically replicate the bubble-growth-induced shock propagation and shear stress generation, we solve two-dimensional, axisymmetric Navier-Stokes equations for compressible flow consisting of gas and liquid components (with no phase change). To be simple, we exclude terms related with heat conduction and surface tension from the equations, for the heat transfer is gradual in comparison to acoustic phenomena and Laplace pressure is much smaller than the pressure imbalance we consider in this study.
The numerical method we use is based on the solution-adaptive, shock-interface capturing method [7]. For spatial discretization, a third-order finite-volume weighted essentially nonoscillatory (WENO) scheme with the Harten Laxvan Leer (HLL) approximate Riemann solver is adopted only in the vicinity of discontinuities including stocks and interfaces [8]. Otherwise, the fourth-order central differencing scheme is adopted. Time integration is handled by a third-order, total variation diminishing (TVD) Runge-Kutta scheme. The calculation domain is set to be so large that the bubble is not affected by acoustic waves numerically reflected from its boundaries. No-slip, axisymmetric [9], and non-reflecting [10] boundary conditions are applied at the edges of the computational domain as explained in Fig. 1. We apply grid stretching with which the simulation is performed with cheaper cost but the flow near the bubble is resolved with finest grid spacing of Δ 0 ⁄ = Δ 0 ⁄ = 3.5 × 10 −3 .
Results and Discussion
As a representative example, we present the evolution of the velocity and pressure fields for the case of standoff distance 0 ⁄ = 1.25 in Fig. 2, showing the bubble-growth-induced shock propagation/reflection and subsequent liquid flow. Spatial evolution of the wall shear stress at representative times is presented in Fig. 3(a). We can see the acoustic event and its induced fluid flow: a spherical shock (as a result of the pressure imbalance between the bubble and its ambient) is initially emitted and subsequently reflected at the wall and the bubble interface. The fluid motion is induced after the shock passage and larger velocity appears in the vicinity of the growing bubble. The maximum wall shear stress (12.3 kPa) appears at =20 ns and may be strong enough, for example,
194
Ultra Clean Processing of Semiconductor Surfaces XV to remove spherical polystyrene particles of diameter =10 nm from a quartz surface [11,12]. In Fig. 3(b), spatial evolutions of the wall shear stress and pressure when the maximum wall shear stress is achieved ( =20 ns) are shown, suggesting that the strong wall shear stress results from the shock passage with its induced flow along the wall surface. We repeated the simulation, but with varying the standoff distance of the bubble to the wall ( 0 ⁄ ). In Fig. 4, we examine the effect of the standoff distance on the maximum wall shear stress and its generation location. It follows that the maximum wall shear stress decreases as −6.49 / 0 and its generation position gets larger as 0.702 / 0 , in the range of 0 ⁄ of our concern. We may say that the growth of bubbles nucleated closer to cleaning targets will contribute to better cleaning performance.
Conclusion
In short, the two-dimensional, axisymmetric compressible Navier-Stokes simulation shows strong wall shear stress generation as a result of the Rayleigh bubble growth, suggesting that cavitation bubble nucleation can play an important role in acoustic-cavitation-based cleaning. The parametric study on the standoff distance of nucleated cavitation bubbles to the no-slip wall suggests that bubbles nucleated closer to cleaning targets will play a more important role as cleaning agents.
|
2021-02-20T14:07:16.262Z
|
2021-02-09T00:00:00.000
|
{
"year": 2021,
"sha1": "050cab071bdde1f5765653e4fc442e4fe66ad5c6",
"oa_license": "CCBY",
"oa_url": "https://www.scientific.net/SSP.314.192.pdf",
"oa_status": "HYBRID",
"pdf_src": "ScientificNet",
"pdf_hash": "050cab071bdde1f5765653e4fc442e4fe66ad5c6",
"s2fieldsofstudy": [
"Engineering"
],
"extfieldsofstudy": []
}
|
251022272
|
pes2o/s2orc
|
v3-fos-license
|
The effects of the COVID-19 pandemic on the well-being of children with autism spectrum disorder: Parents’ perspectives
The COVID-19-related lockdown interrupted children’s learning progress and discontinued social learning and regular activities that children with autism spectrum disorder (ASD) rely on socially and physically. Negative consequences for children with ASD were reported far and wide. To investigate this problem in Kazakhstan, we conducted a mixed-methods study that drew on data from an online survey with 97 parents and semi-structured interviews with 14 parents. While parent-report quantitative results suggest that children were likely to experience negative impacts of the pandemic due to disrupted educational and therapeutic services, qualitative findings confirm that they have experienced an elevated mental health and behavioral challenges during the lockdown. Remote educational and therapeutic services were not helpful as families coped with pandemic-caused problems on their own. We highlight that continued support and care during and after a crisis is vital not only for children with ASD but also for the families under-resourced mentally and socially.
INTRODUCTION
For nearly 2 years now, the COVID-19 pandemic has affected every aspect of life and continues to redefine the ways people socialize far and wide. Even before the current pandemic hit, children with special educational needs and disorders (SENDs) were among the underserved population with fewer opportunities to thrive in day-to-day life. Due to the pandemic-created social restrictions, children with SENDs and their families may experience unstable mental health and abrupt changes in mood and behaviors, as found widely in the United Kingdom (1,2). In particular, research worldwide shows that persons with autism were severely affected by the current pandemic [Colizzi et al. (3) in Italy; Srinivasan et al. (4) in the United States; Kawaoka et al. (5) in Japan]. For example, a significant amount of studies have reported on the negative repercussions such as interrupted daily routines and increased psychiatric problems (e.g., irritability and sleep disorders) caused by the COVID-19 pandemic on the well-being of individuals with autism spectrum disorder (ASD) and their families [see Amaral and de Vries (6), Levante et al. (7), Vasa et al. (8)].
Children with ASD were particularly vulnerable during the pandemic because of their social, communicative, and executive functioning differences and co-existing conditions, such as physical and intellectual impairments (3,(9)(10)(11). For those with ASD, a surge in severe mental health conditions can be triggered by ineffective time management, unstructured activities, and disrupted daily routines in the home environment (7,12). For instance, after two months of the COVID-19induced lockdown, about 6 in 10 children with ASD reportedly worsened psychiatric conditions in the United States (8). These findings reveal that children's behavioral problems may appear in more severe and intense ways during the global lockdown, and show significant negative consequences for parents as well. For instance, Morris et al. (12) explained this through the emotion contagion mechanism, according to which parents' negative emotions, in particular, translate to their children during stressful situations like a pandemic. It is unknown what kind of influence COVID-19 and the lockdowns had and continue to have on children with ASD to address the educational and therapeutic challenges that may emerge in a post-pandemic future. As a result, research studies attempting to understand the impact of COVID-19 on the social wellbeing of the children, whether positive or negative, are called for (6,12). Indeed, the effects of the pandemic may vary from country to country due to different autism contexts and the spread of COVID-19. Therefore, we first present how autism is understood and treated in Kazakhstan and later provide a COVID-19 timeline.
The context of autism in Kazakhstan
Kazakhstan, a post-Soviet country, is undergoing a massive shift in creating an inclusive society, once unbeknownst to the majority as a concept. In Kazakhstan, ASD is poorly understood as a developmental impairment (13). The legal framework does not include the term "ASD" with respect to disorders in official documents, partly because they group children with special needs or mental health impairments under the same group of disorders (14). This generalization of developmental and mental impairments may delay diagnosis and consequently treatment. Children with physical and mental barriers are being treated with a deficit mindset. A recent Kazakhstani study shows that individuals with autism continue to face stigma and discrimination from other people (15).
According to official statistics (available since 2010) in Kazakhstan, one or two in 100,000 children were diagnosed with ASD from 2010 to 2012 (16). However, there are contrasting data on the number of children diagnosed with ASD and receiving the treatment. There is no reliable and publicly available data on how many children and adults live with this diagnosis in the country. Most of the information we report in this section comes from media sources. According to the latest data from the Ministry of Health, the number of children with autism has increased more than 20 times in the last 13 years. 1 They also reported that the screening tests such as ADOS -2 ADI-R had been introduced only recently. Children with ASD receive treatment through a special support scheme Psychological-Medical-Pedagogical Commissions (PMPCs), which are predominantly under-resourced, and practitioners are often oblivious of the latest developments in autism diagnosis and treatment (17). Furthermore, despite some improvements to integrate students with special needs in the traditional classrooms, most students still go to the so-called correctional or remediation classes or study in separate "inclusive" classes within mainstream schools (18). Pedagogical institutions have introduced new modules on inclusive education in the curriculum of pre-service teachers. According to the Ministry of Education, the number of teaching assistants and staff with special qualifications is increasing. 2 Inconsistent autism diagnosis, the lack of legal framework, and support in terms of social care and education provision prevent people on the autism spectrum from living in a more inclusive and equitable society. Of course, the challenges Kazakhstan faces in implementing inclusive education are also found in other contexts (19).
Some promising reforms are to be implemented. In particular, we have noticed several grassroots-level campaigns led by parents of children with ASD. We believe that they are accomplishing truly significant achievements bringing together other parents and thus expanding their network and visibility. For example, a mom of an autistic child started the National Association "Autism Kazakhstan" 3 and advocates for the rights of the children living with developmental disorders, including autism. Another great project is "Autism Pobedim" 4 which organizes educational, therapeutic, and recreational activities for children with ASD. In addition, there are regional autism centers such as "Asyl Miras" funded by Bulat Utemuratov Fund. 5 Here, children with ASD until the age of 15 can receive free educational and therapeutic support from professionals. Despite these significant improvements, there remains much work to be done. More research is needed to support the activities of these organizations and further identify effective mechanisms, best practices, and sustainable solutions to scale up.
A roadmap for the provision of care for children with autism was introduced for multiple purposes, including to collect data on the child population diagnosed with autism and other mental disorders, to increase the provision of specialists, build a national Autism Center, and analyze and improve state laws and documents related to the status of children living with neurodevelopmental disorders (20). However, the current status is unknown. These activities are highly needed to recognize the needs of families raising a child with ASD. As such, these new steps reinforce the importance of continuing autism research to provide better outcomes that guide traditional and alternative ways of support for the healthcare system and clinical professionals to meet effective decisions.
The COVID-19 pandemic in Kazakhstan
The World Health Organization (WHO) declared COVID-19 a pandemic on 11 March 2020; by that time over a hundred thousand cases were registered worldwide (21). The pandemic hit Kazakhstan at about the same time world countries announced the state of emergency (see Figure 1). A national "lockdown" started on 19 March 2020, in two major cities, with a later announcement for lockdown in other regions. All educational institutions, including special-care centers were required to stop all services on the same day the lockdown began. School-aged children had early spring break 2 weeks before the abrupt change to learning remotely in April 2020. Similar to other countries, the transition to remote education was ad hoc and often unsuccessful due to extensive gaps in digital access and skills in the country.
In May 2020, children with special needs were allowed to visit daycare centers and treatment facilities upon parents' consent to follow the restrictions set by the facilities. In fact, Kazakhstan experienced the first wave between July and August 2020, when other parts of the world as many countries in Europe started to gradually reopen their economies after the first shocking COVID-19 outbreak in Spring 2020. Although the country eased quarantine measures from Fall 2020, there have been significant fluctuations in daily cases, while quarantine measures have been still in place for most public activities. In Fall 2020, most schools and universities did not reopen, leaving online learning the only choice for many. While the mass vaccination started early in February 2021, the number of vaccinated people remained low until summer 2021. Vaccination has been accelerated due to mandatory workplace requirements and the implementation of a tracing app (Ashyq) to enter public places. All schools returned to traditional face-toface learning in September 2021, yet some schools were closed when the school-wide COVID-19 situation was out of control.
According to Ministry of Health data, a total confirmed cases of COVID-19 stand at about one million people. Of them, a total of 7,291 children were registered with a confirmed diagnosis of COVID-19 in 2020. Since the beginning of 2021, about 45,000 cases of coronavirus infection among children have been registered in Kazakhstan, a sevenfold increase compared to the whole period of 2020.
The current study
In the past decade, Kazakhstan has increased its awareness of autism and initiated special support mechanisms to support children with ASD and their families. However, the 2 years of the pandemic thwarted that progress and disproportionately affected countries and their social care systems. More active commitment to health care and education is needed to learn though this experience and prevent the same challenges from happening in future. Considering all of the above, we seek to explore how the COVID-19 pandemic influenced the well-being of children with ASD, as reported by their parents. Specifically, we seek to understand how parents of children with autism perceive the impact of COVID-19 on child behaviors, treatment, and daily life in general. The study reveals some profound insights into the importance of continued care for children on the autism spectrum and the need for support mechanisms to serve their needs in times of crisis and beyond. This study contributes to the growing body of autism research conducted worldwide during the pandemic to enable system-wide and evidence-informed decisions and support mechanisms for both children and their parents during emergencies.
Materials and methods
This mixed-methods study was conducted using an online survey and qualitative interviews. Ethics approval for this study was obtained by the Nazarbayev University Institutional Research Ethics Committee. Informed consent forms were collected before taking the survey and participating in interviews. All participants were provided specific information about the research purpose, procedures, benefits and risks, privacy considerations, potential contribution, and contact details of the principal investigator.
Survey
The survey was developed on Qualtrics. It was available from August to September 2021. The survey and interview questions were piloted with one parent via Zoom and adapted based upon feedback. To ensure privacy and confidentiality, settings did not collect personal identifiers such as names, Frontiers in Psychiatry 03 frontiersin.org Timeline of COVID-19 restrictions in Kazakhstan from 2020 to 2021.
contacts details, and IP addresses. That is, the survey link was anonymized. There were no forced-choice questions, and instead, participants were allowed to leave the survey at any stage. There were 26 questions in total in the survey (see Supplementary Material) which included data demographics about the child and parent and the effect of the COVID-19 pandemic on child wellbeing. Here, the human/subjective wellbeing is defined as a multidimensional construct combining psychological, physical, and social dimensions (22,23) in which self-report measures and ratings are indispensable (24). Our understanding draws on subjective parent-report data that indicate how child conditions such as mental health and social behaviors are affected by a novel pandemic. We self-developed survey questions to assess "well-being" through 5-point Likert scales. It included agreement scales, perceived mood rate, satisfaction, therapy quality, importance of educational, and therapeutic services. The surveys were bilingual so that parents could choose the preferred language, Kazakh or Russian.
Interviews
The interviews were conducted concurrently with the survey. Participants who wished to participate in the followup interview provided their email or telephone number at the end of the survey for us to contact them. The parents could choose the preferred language, Kazakh or Russian. Interviews consisted of 16 open-ended questions (see Supplementary Material). They lasted between 17 to 60 min, with an average of 30 min. Prior to the interviews, all parents filled out the consent forms. All parents requested to be interviewed via WhatsApp audio call rather than Zoom, mostly due to technical constraints. The researcher also asked for verbal assent before recording the conversation. Names, locations, and other identifiers were masked by applying pseudonyms. We used pseudonyms when presenting the interview results, and any resemblance to actual names is a mere coincidence. The interviews were conducted either in the Kazakh or the Russian language.
Participants
Participants were recruited using convenience and snowball sampling. To recruit participants, we first contacted parents with whom we worked earlier on studies of robot-assisted therapy for children with ASD (25,26). As we needed to recruit more participants for the survey, we reached out to potential gatekeepers, such as the parental community, autism centers, and non-profit organizations, via email and Facebook. We contacted them directly using the details on their official sites. In particular, the local network of autism centers helped us send out the anonymous survey link to parents' groups and then recruitment flyers (see Supplementary Material). A total of 109 parents of children with ASD in Kazakhstan participated in the survey. Considering that 12 participants did not respond to any question, a total of 97 respondents were included in the final analysis. The mean age of the parents is 34.7 (SD = 6.8). Of 97 participants, 14 parents took part in the semi-structured interviews (see Table 1).
Quantitative analysis
Quantitative data obtained from the online surveys were analyzed with SPSS. A series of Kolmogorov-Smirnov tests were conducted to check if the data were normally distributed. We ran Mann-Whitney U and Kruskal-Wallis tests for nonnormally distributed data. Predictor variables included: urbanrural, parents (education background and employment status), family (language background and number of children), and children (gender, current age group, age diagnosed with ASD, schooling status, and access to therapy before and after pandemic).
Qualitative analysis
Qualitative data collected from interviews were analyzed using thematic analysis (27,28). We used inductive and deductive coding strategies (29,30). Themes and codes were checked for accurate interpretation of results by consulting interview excerpts. The interview questions can be found in Supplementary Material. The interview findings were transcribed and then coded by hand using thematic analysis. One researcher analyzed over 500 min of audio recordings. The emerging themes are related to the impact of quarantine measures on children with ASD and their families, as shown in Table 2.
Results
In this section, we provide first the results from an online survey and then move on to discuss interview findings.
Perceived impact of the COVID-19 pandemic
Parents reported an agreement score of 3.65 (SD = 1.21) (out of 5 for "Strongly Agree") for the question of how severely the COVID-19 affected their children with ASD. To explore what socio-demographic characteristics of the surveyed families differentiated their responses, we conducted a series of Mann-Whitney U and Kruskal-Wallis tests. The majority of independent variables did not reveal any statistically significant results. However, we found a statistically significant difference in the perceived impact of the pandemic between respondents with children of varying schooling status: χ 2 (2) = 8.199, p = 0.017. Parents of children without formal schooling reported a significantly stronger impact of the pandemic (4.00 ± 1.09) compared to those parents of children that attended mainstream schools (3.00 ± 1.15), as presented in Figure 2.
Child and parent mood during the pandemic
The mean of parents' mood was 2.67 (SD = 0.851) out of 5, with 0 being very negative and 5 being very positive. Interestingly, there was a significant difference in parents' mood between the reported number of children: U = 721.000, p = 0.037. Parents with two or more children had significantly higher levels of positive mood (2.75 ± 0.847) than those with only one child (2.31 ± 0.793). Parents reported their child's mood to be 2.81 (SD = 0.753) out of 5 with 0 being very negative and 5 being very positive. It was significantly predicted by parents' age groups [χ 2 (2) = 6.631, p = 0.036] and children's schooling status [χ 2 (2) = 7.957, p = 0.019]. Children of young parents aged less than 30 years old had higher mood ratings (3.13 ± 0.619) than children of older (3 ± 0.784), and middle-aged (2.62 ± 0.747) parents. Another interesting finding is that children attending mainstream schools were reported to have higher levels of positive mood rate (3.31 ± 0.480) than their peers in special centers (2.68 ± 0.764) and those without any formal education (2.75 ± 0.775). Pairwise comparisons showed significant differences between children at special centers and mainstream schools, as shown in Figure 3.
Frequency and satisfaction with autism spectrum disorder therapy before and during the pandemic We found no significant differences in the frequency of ASD therapy before and during the pandemic. The small sample size might explain this. However, there is a marginal difference between receiving the treatment in both periods. The frequency of everyday treatment dropped after the pandemic from 17 to 7 as conducted at the treatment center. There has been a double increase in the category "several times, " from 12 before March to 24 after March 2020 (see Figure 4). The number of treatment sessions slightly increased in other categories, which shows a greater need for therapeutic services during the pandemic. However, the paired samples t-test showed that the parents' satisfaction with ASD treatment decreased significantly during the pandemic (3.12, SD = 1.142) compared to the pre-pandemic time (3.42, SD = 1.136): t(58) = 2,125, p < 0.050.
Quality of online autism spectrum disorder therapy during the pandemic Overall, the satisfaction rate with the quality of ASD therapy was low and stood at 4.87 (SD = 2.876) out of 10. The results show a highly significant difference in the reported quality of online ASD therapy depending on children's age groups: χ 2 (2) = 7.038, p = 0.034. Parents of children aged five or younger rated online services more favorably than those having children in older age groups (see Figure 5). Perceived impact of the pandemic grouped by schooling status. *Means statistically significant pair-wise comparisons. Child and parent mood during the pandemic. Asterisk indicates a 0.05 level of significance.
Support mechanisms during the pandemic
We also tested how parents rated the importance of support mechanisms in terms of access to therapeutic and educational services (see Figure 6). We found clear trends in the ratings of the importance of educational services [χ 2 (2) = 7.311, p = 0.026] and regular therapy FIGURE 4 Therapy frequency in pre-pandemic and pandemic periods.
[χ 2 (2) = 6.626, p = 0.036] between children's age groups. Parents of children aged nine and older rated that those services were significantly less important than parents of younger age groups. Compared to older age groups, parents of children aged 6-8 rated regular therapy and educational services significantly more important.
Interview findings Theme 1: Coping with quarantine measures and restrictions
Pandemic-caused restrictions in daily life Self-isolation was exercised widely, requiring people to stay at homes to contain the spread of COVID-19. It has been discouraging for many people, but our findings show that the stay-at-home measure heavily influenced children with ASD. They are particularly vulnerable to the quarantine measures that restrict access to educational, recreational, and therapeutic services because their well-being depends on regular social and health support. A recurrent concern for all families was restrictions applied to free movements when the national lockdown was announced. Parents reported that their children were frustrated while staying indoors. These restrictions were lifted after some time in quarantine, and children with special needs like autism could go outside, as illustrated in this quote: When the pandemic began, the only difficulty was that we could not go out after 6 p.m. Once when we were walking in the playground, a police officer, who was doing a quarantine compliance check, told us to go back home and not walk around. We were following all the quarantine measures and wearing a mask though. It was only then [after lockdown] that children with special needs were allowed to go outside (Alena).
Some emergency services were limited as cities had to set up special quarantine checkpoints, which prevented any transport from crossing designated places. For instance, Aidana shared her experience of calling an ambulance from the city, which was returned to the city: "Once my son hurt his leg when he rode the scooter inaccurately and fell down. We called an ambulance. At the checkpoint, they [police officers] did not allow the ambulance to pass because it came from the city." (Aidana)
Parental concerns over children's health
A common view among the parents was that they endured higher fear levels concerning children's health during the quarantine. Such fear likely stems from the medical observation, according to which children with ASD are commonly diagnosed with co-occurring health problems (e.g., ADHD). Such stress and fear are illustrated in the comment below: "My child has primary immunodeficiency Quality of online ASD therapy during the pandemic. The importance of support mechanisms. *Statistically significant pair-wise comparisons.
disease (PIDDs) and other diseases like VRI intolerance.
When quarantine announcements began to appear, a week before the quarantine, we already canceled lessons [with the specialist]. I feared for her health" (Galiya). Besides health issues, parents had to restrict outdoor activities for other reasons. As one parent stated, her family moved into a new district during the quarantine: "Last year [during the quarantine], we movedy to a new district. It was a new place. We didn't know our neighbors. Thus I didn't let my child out. We didn't go outside during the quarantine." (Aliya)
Home environment
We found differences in quarantine experiences of families that reside either in private homes or apartment buildings. The impact of variations in home environments during the quarantine was tangible. Some parents lived in single-room apartments that further affected their life in quarantine and their children's well-being. The following quotes describe these experiences vividly of their children with ASD not having sufficient space for physical and emotional needs: "As we lived in a single-room apartment, we were restless and couldn't find peace." (Zhansaya); "During the quarantine, we were stuck in a single-room apartment with an autistic child and a toddler. It was an unpleasant experience; even a typically developing child got bored." (Mira)
Theme 2: Effects for children and their families
Negative effects of the quarantine on children with autism spectrum disorder Hyperactivity was a regular condition as children showed atypical behaviors that surprised and concerned their families: "she [tried] climbing ceilings" (Dariya), "he ran, jumped, turned around, did not comply" (Zhansaya), "he ruined everything he touched" (Aisulu While for most children the reason for their negative behavior was the lack of time outdoors, a few children had to spend time with people who were unaware of approaches to helping children with ASD: He spent summer in the village last year [during the quarantine]. He forgot the skills he learned. In September, when he returned, he had aggression and tantrums because he didn't want to study in the [autism] center. When he was in the village, no one demanded anything from him because his grandmother didn't know how to deal with him. (Mira) A substantial number of parents reported increased screen time on gadgets and TV. This subsequently resulted in the lack of attention to the surrounding environment. Most children seemed to get stuck in such events leading to such extreme instances of not eating: Since we didn't go outside last year, we gave her the phone we took from her earlier [in the year]. Then she was stuck on the phone and didn't look around. She didn't care whether we came by or not. She was enough for herself. She didn't even eat food. (Aliya) Although the majority of mothers were housewives with a spouse/family to financially support the household, some were the primary breadwinner, working and supporting their children on their own. Those parents were desperate to rely on gadgets to keep their children engaged while working from home during the quarantine. This is clearly discussed in the following comment: I worked online, and the kids were on their own. To ensure that they didn't scream and fight, I gave the older one [with ASD] the phone, and the [TD] younger one took care of himself. The child [with ASD] got addicted to the gadget because of it. Certainly, it shouldn't be that way. But when you have kids to feed, you have to work (Mira) Positive effects of the quarantine on children with autism spectrum disorder and their families Despite the adverse consequences of the pandemic, a few children improved their behaviors and learned new skills while at home. However, such progress should not be only associated with the pandemic but also with their development. The following quotes refer to the improvements, such as fewer extreme behaviors, more compliance, and new learning: "I mirrored his actions when he screamed and stomped his feet at home. I asked older children to do the same. He somehow started to understand that those actions were not good. Now he does them less often." (Aisara); "He started to comply with my instructions a bit. He began to use his forefinger. He couldn't show it before. He also started to pronounce some words like 'Dad, ' 'Mom' and 'Give."' (Zhansaya); "Primarily, we practiced fine and gross motor skills. He can hold the pen; until then, he held it like a shovel. Now he holds it with three fingers as it should be." (Aliya) In our study, almost all families had two or more children. The participating parents reported that life in quarantine was helpful to reconnect with family members and understand each other, contributing to their children's well-being. In normal times, fathers or siblings seem not to be as close in physical proximity and emotionally as mothers do with their children with ASD. However, during the pandemic, with parents working from home, some parents commented on a positive change in family values: "Most of all, I liked that my family reunited as if we understood each other better. Family values became tangible. . . We were constantly engaged. For instance, we performed physical exercises and played games with the child [with ASD] just to keep him interested. My husband and everyone [siblings] contributed, not just me alone." (Aidana); "The positive thing is that her dad got to spend more time with her because he was in quarantine, too" (Galiya). Dilyara reported the importance of parental involvement in delivering remote learning: [During remote learning], we became our child's teacher and taught him, for example, all kinds of football activities. We video-recorded him at the same time. Two parents must be involved. One of them explains everything to the child, where to score the ball, how to do the exercises, while the other has to record the process. (Dilyara) Almost all parents explained efforts to develop their children's social and life skills regularly. They reported that their children became more independent and skillful to fulfill their own self-care needs. For instance, with fewer cars and people on the road, Dariya could teach his son street safety. "Last year during the quarantine, I taught him how to behave in the streets, like ''Stop, ' 'Look out, ' 'Wait, there's a car.' I taught a lot during the quarantine because there were a few people outside." (Dariya) In other cases, children mastered self-care skills themselves, as reported by Alena: Positive changes related to self-care and hygiene like washing hands. We attended remedial kindergarten for two years, but it may happen that he hadn't been taught. He probably was still a little child back then. But during this time [quarantine], when we stayed at home, my child learned how to take care of himself and be independent, so he could take care of himself and dress himself. (Alena) Theme 3: Experiences with therapy and learning during the pandemic
Remote therapy challenges
Almost all parents received asynchronous support from teachers and specialists to whom they sent out the tasks via social networks like WhatsApp. Technical constraints made it difficult for teachers to observe children's behaviors online. Therefore, this format is not sustainable for children with ASD, says one parent.
We had a WhatsApp video call for 30-40 min or so. The only thing was that I did everything without a therapist alongside. The teacher did not fully observe the child because I put the phone on from a different angle. The observation gets worse because of the video call (Galiya). Some parents thought online therapy was better than nothing, supporting their child on their own relying on therapeutic support online. But the therapy time decreased a bit, likely due to restrictions set on platforms like Zoom, limiting free accounts to 45 min.
I was happy to have online therapy because we didn't waste time. Even though my child didn't participate directly, I learned and taught my child myself. But offline [treatment] is better. Due to the pandemic, therapy time decreased to 45 min; earlier, it was one hour. (Aliya) For some children with ASD, their sensing abilities are a significant source of their learning processes, which could disadvantage them during online therapy: "I don't think it's [online therapy] functional because autistic children are susceptible. They need visual and tactile contact. My child likes touching everything." (Zhazira)
Remote learning challenges
It would not be an overstatement to say that online learning was of great tension for children with communication and attention barriers distinct to autism. All parents unanimously were against adopting online learning as an alternative to traditional learning. One parent described her child's experience in an online dance club as a failing experience: It's us -a dad and a mom -who do tasks while online. It can't be a lesson. He didn't understand anything and had no interest at all. Therefore he quit the dance club. In-home and online dancing was challenging. He could perform physical activities but not dance movements. (Dariya) With remote learning, parents had to fill in for the newfound roles of teachers for their children. What works for some children may not work for those children with special needs, and many parents were in a trying situation during the pandemic. Two parents confessed that they wanted to remain mothers, not teachers. For instance, a child showed negative attitudes toward his mother who had to act as a teacher: I want to remain a mom to my child. . . During the quarantine, I demanded him practice motor skills. I made him write, draw, and imitate, which he needed to do based on his age range. I tried to do those activities in the Centre by myself. But it was not good for us. It backfired because he started to develop negative feelings towards me. For such children, it is better to do activities with a stranger, not their mom. (Lara) Child inattention seemed to be a major reason for the unsuccessful transition to online learning. For children with ASD, it is hard to keep undivided attention to activities. In most cases, children tend to perceive online learning unseriously because they are usually exposed to technologies such as a mobile phone as a source of entertainment to watch videos or play games: "Online learning is not applicable because the child sees the phone, and he loses engagement. He wants to watch YouTube. Otherwise, he screams hysterically" (Mira). One parent even perceived it causing an opposite effect: We certainly had a few online classes. But it did not work. He was freaking out that he was not allowed to watch what he wanted and instead was forced to watch the classes online. It was stressful for him and for us. We gave up on online classes because they had no effect. It was even the opposite. It was detrimental. (Lara)
Key findings
• Pandemic-caused restrictions disrupted access to essential and non-essential services, particularly the access to therapeutic and medical facilities that are vital for children with ASD. In particular, the overall well-being of children with no schooling experience were likely to worsen compared to those in mainstream schools during extended lockdowns.
• Both parents and children had reportedly lower mood ratings during the pandemic. Parental concerns over a child's health and safety were elevated because some children have cooccurring conditions such as immunodeficiency and anxiety, while the imposed restrictions in daily life triggered the atypical behaviors among the children.
• Although we collected data on the employment status of participants, it did not reveal any significant differences in the perceptions of employed and unemployed parents. Yet, the qualitative findings show that families from lower socioeconomic backgrounds had fewer opportunities to support their children while at home. In the study, most families raising a child with ASD have two or more children and usually reported to struggle to have ends meet.
• The time allocated for most free treatment services was shortened. Online therapies were also short partly due to time restrictions in platforms like Zoom. The number of daily treatment sessions marginally dropped threefold when the pandemic hit.
• Parents reported distance learning experiences discouraging and not helpful both for their children and themselves. Although parents attempted to replace teachers and therapists at home, they had no training to serve their children's needs in the best possible ways. Parents' satisfaction with ASD treatment decreased significantly when we compared how parents rated their satisfaction in the pre-pandemic and the pandemic period.
• Both educational and therapeutic services were found to be unanimously important during the pandemic. This finding clearly explains how salient is the access to key support mechanisms for children with ASD; as Baweja et al. (9) noted, "it takes a village" to raise a child with ASD. The quality of online ASD therapy was rated notably low. It might be affected by how unprecedented the shift from traditional to online treatment was.
Discussion
The study explored how children with ASD experienced the COVID-19 pandemic from their parents' perspectives. We particularly investigated whether children with ASD are vulnerable to pandemic-imposed social restrictions that were found to bring overwhelming challenges to regular treatment and learning across countries. Most parents reported that the restrictions related to therapeutic and educational services were particularly prominent. We found quantitative evidence as represented in significant differences in the parent-report severe effect of the pandemic on children who did not have formal schooling. Interview findings confirm that the lockdown measures have resulted in disrupted ASD treatment, which is critical to achieving positive benefits in the long term (31)(32)(33).
The degree to which children with ASD in Kazakhstan experienced adverse effects of the pandemic is not uniform. We could not identify if there were differences in the perceived effects of the pandemic between employed and unemployed parents, particularly due to small sample size. But qualitative findings show varying home environments may affect considerably. For instance, parents living in small apartments reported fewer opportunities to support their children in regular practices of new skills, while those in private houses had resources and physical spaces to arrange recreational activities outside. This may account for a lower socio-economic situation in the families. For instance, lowincome families tend to report greater struggles when caring for their autistic child during the quarantine. This confirms prior research emphasizing intensified inequities in coping with stress and behavioral changes between economically weaker and better-off families raising a child with ASD (8,(34)(35).
The results show that technology-based autism treatment modalities cannot replace in-person services. That being said, our data further shows that only certain groups may have relatively positive online experiences, for instance, young children under 5 years old. This finding is entirely unexpected, however, the learning content for young children with ASD might be more entertaining and easy to follow in daily life compared to older children who need more advanced and complex learning guidance. For instance, parents of older children rated the importance of educational and therapeutic services relatively higher than that of younger children. Some inherent limitations with telehealth-based intervention mainly include child inattention during sessions and technical issues like video setup. This problem might not be unique to these children as their neurotypical peers may lose interest easily. Internationally, when children with ASD could not attend schools, many parents reported having online lessons and activities via digital platforms such as Zoom and Skype (12). In our study context, parents could only access asynchronous learning and therapy services through WhatsApp. In this case, therapists or tutors send out activities that children need to perform (e.g., physical activities and daily skills) while their parents/siblings either assisted or recorded children's activities. The quantitative results also show that the parent-report satisfaction rate with the therapy decreased during the pandemic.
Distance learning environments are less likely to meet the needs of children with ASD if they are not given adequate guidance and support in managing the new environment they have never experienced before. Since rehabilitation centers and schools were partially or fully closed, parents of children with ASD had to replace teachers to continue education from home. Our study suggests that parents of children with ASD should be supported more during crises. Some parents reported being exposed to elevated stress due to the quarantine and its discouraging impacts. This experience was overwhelming for most parents -who are largely -not prepared to delegate demanding parental and educational responsibilities, to name a few. Parents were conditioned to care for their children on their own while juggling multiple responsibilities for which they had not been trained. The unique struggles of parents of children with ASD are echoed in similar studies that report family tensions (36), lacking a resourceful state (37), and increased stressors during the pandemic (38). Parents often need extra support from other people, particularly professionals who can aid and counsel when raising a child with a developmental impairment. It is true that without the challenges of COVID-19, parents usually face many obstacles in obtaining continuous care.
We conclude that children with ASD and their parents endured mental and behavioral problems that seem to be predicted by disrupted therapeutic and educational opportunities. This is supported by the quantitative results reporting higher agreement with the perceived adverse effects of the pandemic on both children and parents. Almost all parents reported that remote educational and therapeutic opportunities did not satisfy their children's needs due to the challenges associated with unique autistic traits, technical problems and parental stress. System-wide and evidence-informed support mechanisms for both children and their parents need to be implemented to support telehealth therapy modalities.
The main limitation of this work is that we could not deliver standardized tests evaluating medical conditions such as anxiety level and other behavioral changes during the pandemic. Yet, to our knowledge, this is the first study that explored a relatively large sample of parents of children with ASD in Kazakhstan to date. This study could be extended by conducting a followup study on post-COVID treatment options and what has been learned after the series of lockdown measures.
The results from the current study can help identify therapeutic needs of the children in future interventions embracing remote technologies and in-home treatment options for parents and other caregivers. More research and development are called for to advance the current limitations of therapeutic and educational opportunities in Kazakhstan and elsewhere.
Data availability statement
The original contributions presented in this study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.
Ethics statement
The studies involving human participants were reviewed and approved by the Nazarbayev University Institutional Research Ethics Committee. The patients/participants provided their written informed consent to participate in this study.
Author contributions
AA: data collection, data visualization, and writing the first draft. AC: conceptualization, and writing-review and editing. AS: conceptualization, supervision, funding acquisition, data curation, and writing-review and editing. All authors contributed to the article and approved the submitted version.
Funding
This work was supported by the Nazarbayev University Collaborative Research Program grant (award number 091019CRP2107).
|
2022-07-25T13:41:58.224Z
|
2022-07-25T00:00:00.000
|
{
"year": 2022,
"sha1": "a8ed51623aae1ff1c3b4e5b68b9c60ab0918bd5e",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Frontier",
"pdf_hash": "a8ed51623aae1ff1c3b4e5b68b9c60ab0918bd5e",
"s2fieldsofstudy": [
"Psychology"
],
"extfieldsofstudy": []
}
|
261380272
|
pes2o/s2orc
|
v3-fos-license
|
Phytochemical Profile and Composition of Chickpea (Cicer arietinum L.): Varietal Differences and Effect of Germination under Elicited Conditions
Germination is a simple process that improves the nutritional and medicinal values of seeds such as chickpeas. However, the detailed analysis of the phytochemical profile after chemical elicitation during chickpea germination is indispensable when making inferences about its biological properties. Therefore, an evaluation was made of the effect of the chemical inducers salicylic acid (SA, 1 and 2 mM), chitosan (CH, 3.3 and 7 μM), and hydrogen peroxide (H2O2, 20 and 30 mM) during germination at 25 °C with 70% RH for 4 days on the content of antinutritional and bioactive compounds, including phenolics, sterols, and saponins, in three Mexican chickpea varieties (Blanoro, Patron, and San Antonio) using UPLC-ELSD-ESI-QqQ-MS/MS, UPLC-DAD-ESI-QqQ-MS/MS, and HPLC-DAD-sQ-MS. The highest increase in phenolics and saponins was found in the Blanoro sprouts induced with SA 2 mM, whereas the highest phytosterol content was detected in San Antonio sprouts induced with CH 7 μM. In addition, significant increases in mono-, di-, and oligosaccharides and decreases in antinutritional contents were achieved after germination with most of the elicitation conditions. More importantly, we identified new compounds in chickpea sprouts, such as the lignans matairesinol and secoisolariciresinol, the phenolic compounds epicatechin gallate and methyl gallate, some phytosterols, and the saponin phaseoside 1, which further increased after chemical elicitation.
Introduction
Chickpea (Cicer arietinum L.) is one of the most ancient and consumed legumes around the world; it can be broadly divided into light-yellow-coated (kabuli) and dark-browncoated (desi) types and grows mainly in areas with temperate and semiarid climates [1,2].In Mexico, the northwest area is the major production region of white chickpea for exportation, followed by the west-central region [3].In this regard, the development of light-yellow chickpea grains with superior characteristics has generated new varieties, such as Blanoro, with a grain color and size that is suitable for the international market [4,5].Furthermore, Blanoro was one of the most promising kabuli types with the highest values of phenolic acids among the selected chickpea genotypes from Mexico [6].San Antonio is another variety with a pigmented, coated (desi type) seed that is resistant to root rots and has a higher yield than other varieties [7].In addition, the introduced variety, Patron, is a pigmented seed that originated under genetic improvement in Mexico.
Chickpeas are considered a good source of carbohydrates and proteins of a higher quality than those of other grain legumes, and they are a source of important vitamins and Plants 2023, 12, 3093 2 of 26 minerals [2,8]; nevertheless, chickpeas contain antinutritional compounds that can impair the utilization of nutrients.These include the protein antinutritional factors, such as trypsin, chymotrypsin, the amylase inhibitors, and lectins, and the non-protein antinutritional factors, such as tannins and phytic acid.However, these compounds can be reduced or eliminated by using processing methods like soaking, cooking or toasting, fermentation, and germination [2,8,9].In this regard, there is an increased interest in the production of germinated edible seeds for human consumption.
Germination is a simple process that improves the nutritional and medicinal values of seeds and provides edible sprouts that can be consumed as functional foods [10].This is of major relevance since the determination of the nutritional composition, as well as the bioactive components, of functional foods has a key role in defining daily nutrient intakes at the population level and their association with health effects [11].In this regard, the phenolic acid profile and many other bioactive compounds that might be considered beneficial as antioxidants are dramatically augmented during germination; in addition, an increase in the isoflavonoid diversity and overall antioxidant capacity of chickpea sprouts (CS) has been reported [10,[12][13][14][15][16].Furthermore, a few other compounds have been identified in chickpea raw seeds, such as lignans, phytosterols, and saponins [2,8,[17][18][19], which might also exert health benefits [20,21].Nevertheless, the effect of germination on the profile of the phytosterols and saponins of germinated chickpeas, as well as after chemical elicitation, has not been reported.
In addition, the synthesis of secondary metabolites can be augmented through the exogenous application of elicitors.In this regard, salicylic acid (SA), chitosan (CH), and hydrogen peroxide (H 2 O 2 ) have been proven to enhance seedling growth and to increase the content of total polyphenols and flavonoids in common beans, soybeans, and chickpea seeds during germination by activating the enzymes involved in phenolic compound biosynthesis [17,18,22,23].However, a well-characterized phytochemical profile, including that of phytosterols and saponins, after chemical elicitation during chickpea germination has not been evaluated.
The evaluation of the influence of agricultural practices, intra-species biodiversity, and environmental factors, among others, might impact the relationships between the food quality and the health benefits [11].In this regard, the effect of agricultural practices, such as germination, and environmental factors, such as chemical elicitation, on the enhancement of the nutritional and bioactive component composition might also depend on the chickpea variety.Therefore, the aim of this study was to evaluate the effect of chemical elicitors (SA, CH and H 2 O 2 ) on three Mexican varieties of chickpea, Blanoro, San Antonio, and Patron, to improve the nutrimental quality and phytochemical profile of the resulting sprouts.
Increased Percentage of Germination and Radicle Size of Chickpea Seeds after Chemical Elicitation
The percentage of germination and the radicle size of chickpea seeds treated with three different elicitors were evaluated in order to determine the improvement of the seedling sprouts.The three Mexican chickpea cultivars presented germination percentages higher than 90% after 4 days of germination with or without elicitation; the Blanoro cultivar had the highest increase in germination percentage, particularly the seeds treated with 2 mM SA and 7 µM CH (p < 0.05, Figure 1).Similarly, the Patron and San Antonio cultivars elicited with 2 mM SA and 7 µM CH, respectively, showed higher germination percentages compared with their control germination sets, although significant differences were only observed in Patron CS (p < 0.1, Figure 1).
According to Figure 1, the 1 mM SA-elicited Blanoro cultivar showed the largest radicle size increase (16.2 vs. 14.0 cm), whereas Patron and San Antonio CS elicited with 7 µM CH had the largest radicle size compared to their control CS.Similarly, with regard to germination percentage, elicitation with H 2 O 2 showed a negative effect on Patron CS.According to Figure 1, the 1 mM SA-elicited Blanoro cultivar showed the largest radicle size increase (16.2 vs. 14.0 cm), whereas Patron and San Antonio CS elicited with 7 μM CH had the largest radicle size compared to their control CS.Similarly, with regard to germination percentage, elicitation with H2O2 showed a negative effect on Patron CS.
Decreased Content of Antinutritional Compounds in Chickpea Sprouts after Chemical Elicitation
Table 1 shows the levels of trypsin inhibitors (TIA), lectins (HU), and phytic acid (PA) in the three raw Mexican chickpea cultivars, and it is clearly observed that the germination process by itself significantly decreases the content of the antinutritional compounds in all CS with respect to their contents in the raw seeds (Table 1).TIA significantly decreased up to 26.2%; the Patron CS had the lowest activity (14.67 ± 0.69 TIU/mg) and the San Antonio CS had the highest decrease (26%) in TIA (from 26.40 ± 1.67 to 19.50 ± 0.12 TIU/mg).Interestingly, all three of the chemical elicitors had an impact on TIA (up to 50.8% inhibition) at the highest applied doses of each elicitor (2 mM SA, 7 µ M CH, and 30 mM H2O2) during germination, as compared to the raw seeds and despite the different initial TIA levels.
Decreased Content of Antinutritional Compounds in Chickpea Sprouts after Chemical Elicitation
Table 1 shows the levels of trypsin inhibitors (TIA), lectins (HU), and phytic acid (PA) in the three raw Mexican chickpea cultivars, and it is clearly observed that the germination process by itself significantly decreases the content of the antinutritional compounds in all CS with respect to their contents in the raw seeds (Table 1).TIA significantly decreased up to 26.2%; the Patron CS had the lowest activity (14.67 ± 0.69 TIU/mg) and the San Antonio CS had the highest decrease (26%) in TIA (from 26.40 ± 1.67 to 19.50 ± 0.12 TIU/mg).Interestingly, all three of the chemical elicitors had an impact on TIA (up to 50.8% inhibition) at the highest applied doses of each elicitor (2 mM SA, 7 µM CH, and 30 mM H 2 O 2 ) during germination, as compared to the raw seeds and despite the different initial TIA levels.
Similarly, the content of lectins significantly decreased during germination (up to 51.3%); Blanoro CS had the lowest lectin content.Furthermore, all the chemical elicitors had a significant impact on the lectin content of the CS at the highest concentrations applied (2 mM SA, 7 µM CH, and 30 mM H 2 O 2 ), as compared to those of the raw seeds (Table 1).In addition, all three Mexican CS showed the highest decrease in lectin content at 2 mM SA; in particular, the Blanoro sprouts had the highest decrease of 81.2%, as compared to the Patron (73.1%) and San Antonio (61%) CS.
The PA contents in the Blanoro and Patron raw and control CS were similar but were lower for the San Antonio cultivar (Table 1).Despite the differences in the initial content, the PA levels in the CS reached a decrease value of about 40% after 4 days of germination; furthermore, a significant decrease of 69% was observed in Blanoro CS elicited with 1 mM AS and 7 µM CH.In addition, the 30 mM H 2 O 2 -stressed San Antonio sprouts had the largest decrease of 73.1% in PA content (from 4.80 ± 0.46 to 1.29 ± 0.21 mg/g dry flour).Overall, we observed a significant reduction in PA contents after chemical elicitation. Trypsin inhibitory units (TIU)/mg of dried flour, 2 Haemagglutinating units (HU)/g of dried flour, 3 mg of phytic acid (PA)/100 g of dried flour. 4Control: germinated seeds with distilled water.SA: salicylic acid, CH: chitosan, H 2 O 2 : hydrogen peroxide.Samples were analyzed on fourth day of germination.
Increased Mono-, Di-, and Oligosaccharide Contents after Germination and Chemical Elicitation
In chickpeas, seed germination exhibits a tight temporal regulation on the release and utilization of sugars (reducing sugars, sucrose, and RFOs).As expected, sucrose occupied first place in terms of abundance in the raw seeds, followed by oligosaccharides and monosaccharides, particularly stachyose and fructose, respectively (Table 2).Interestingly, germination under a control condition (distilled water) significantly increased all the carbohydrate contents in the three Mexican chickpea cultivars; they were further increased or maintained after chemical elicitation.In this regard, the sucrose content was higher in the Blanoro raw seeds; however, the chemically stressed San Antonio and Patron CS reached similar levels of sucrose.Furthermore, San Antonio CS had the highest content of stachyose after chemical elicitation.On the other hand, mannose was only detected in the pigmented San Antonio and Patron raw seeds; however, Blanoro CS had similar mannose contents to that of the pigmented sprouts.It is important to mention that in the present study we identified and quantified mannose, a monosaccharide not previously reported or detected in chickpea seeds and sprouts.
Chemical Elicitation and Chickpea Varietal Effects on Phytochemical Profile of Chickpea Sprouts
Chickpea seeds are a rich source of several polyphenolic compounds.In this regard, the isoflavones biochanin A and formononetin were the major compounds present in the three Mexican CS (Table 3), followed by genistein and daidzein in lower concentrations.The Blanoro CS showed the highest contents of formononetin and biochanin A, followed by the San Antonio and Patron CS, respectively.Regarding the isoflavone genistein, the highest concentration was detected in the control CS of San Antonio, followed by the Blanoro and Patron CS; a similar pattern was observed for the isoflavone daidzein.As expected, the application of chemical elicitors during germination increased the contents of the three major isoflavones (genistein, biochanin A, and formononetin) (Table 3).The greatest increases were detected in the 2 mM SA-stressed Blanoro sprouts (14, 43, and 80%, respectively), the Patron sprouts with 30 mM H 2 O 2 and 3.3 µM CH treatments (up to 17, 78, and 114%, respectively), and the 7 µM CH-stressed San Antonio sprouts (35, 22, and 45%, respectively).
Similarly, the germinated San Antonio seeds treated with 7 µM CH showed increases of 21 to 129% in the contents of chlorogenic acid, p-hydroxybenzoic acid, matairesinol, ethyl gallate, kaempferol, epicatechin gallate, epigallocatechin gallate, quercetin, and ellagic acid; the latter compounds had the highest increases, followed by ethyl gallate and kaempferol.Interestingly, the flavonoid quercetin was the phenolic compound that commonly showed the highest concentration increases in all three chemically elicited Mexican chickpea cultivars.
As shown in Table 4, β-sitosterol was the major saponin component in the control Mexican CS.Likewise, the compounds campestenyl glucopyranoside and sitosteryl glucopyranoside were identified and quantified in the control Blanoro CS; the compounds campestenyl glucopyranoside, sitosteryl glucopyranoside, and brassicasterol were detected in the control Patron CS, whereas the compounds campesteryl glucopyranoside, fucosterol, and sitosteryl glucopyranoside were also identified in the control San Antonio CS.Although the contents of the phytosterols increased or decreased independently of the elicitor and concentration applied (Table 4), the highest increase in the content of the major component β-sitosterol was detected in all three Mexican CS chemically stressed with 7 µM CH (from 20.5 to 71%), whereas the other phytosterols did not show a clear pattern of induction.Interestingly, we also observed varietal differences in the level of phytosterol induction; the chemically stressed Patron CS had the highest increases (33-3300%) in most of the detected phytosterol compounds, and the San Antonio CS had the lowest increases (8.6-519%) among the three Mexican CS.
Saponins were also detected and quantified in the three Mexican CS, and these included soyasaponins Bb (I), βg (VI), Ba (V), and αg; soyasaponins VI and Af were the major saponins present in the three control Mexican CS (Table 5), followed by soyasaponins Ba (V) and αg; the control Blanoro CS had the highest contents.As expected, the application of chemical elicitors during chickpea germination increased the contents of the major saponins (soyasaponins Af and VI) in all three Mexican CS; the greatest increases were detected with the 2 mM SA treatment for the Blanoro (38 and 25%, respectively); 3.3 µM CH for the Patron (53 and 25%, respectively); and 7 µM CH for the San Antonio cultivars (59 and 35%, respectively).Interestingly, the phaseoside I compound had significantly higher increases in the Blanoro CS (62%), Patron CS (655%), and San Antonio CS (160%) under the same chemical elicitation conditions, respectively.For the rest of the saponins identified in the present study, no significant increases were found in the chemically stressed CS of the three Mexican cultivars.
Increased Percentage of Germination and Radicle Size of Chickpea Seeds after Chemical Elicitation
There is an increased interest in the production of germinated edible seeds for human consumption since germination is a simple process that improves the nutritional value of seeds, obtaining edible sprouts that can be consumed as functional foods [10].Therefore, it is important to identify the elicitors that stimulate the growth and yield of chickpea sprouts.According to Soltero-Díaz et al. [7], the San Antonio cultivar is a variety with a higher yield than other varieties.In this work, the Patron cultivar showed the highest germination percentages among the three Mexican chickpea cultivars.Interestingly, we observed higher germination percentages in the control seeds of the three Mexican cultivars than those previously reported for other chickpea landraces or varieties [18,22,24].Although the germination percentage of the three Mexican chickpea seeds in our study was higher (90.0-96.3%)than that previously observed when 5 to 30 mM H 2 O 2 concentrations were applied (80.0-91.1%) between 2 and 4 d of germination time [23], elicitation with H 2 O 2 showed the lowest improvement of the seedling sprouts in the three chickpea cultivars, compared to the control seed germination (90.7-95.7%).Similarly, Amjad et al. [17] evaluated the effects of the seed priming technique with H 2 O 2 on the seed germination and vegetative growth of chickpea seeds and found non-significant improvement compared with the control condition (distilled water).The germination percentage range of the three Mexican chickpea cultivars elicited with SA (91.3-97.7%)was similar to that of chickpeas from Iran germinated with 1.5-3.0mM SA (90-97%) [22].
On the other hand, the radicle length in the control CS (11.6-14.0cm) was like those previously reported [24].In agreement with the results in this work, SA (0.1 and 2 mM) induced a larger radicle size in common bean sprouts; however, no significant difference was observed in sprouted common bean seeds with CH (3.3 and 7 µM) treatment when compared to the control condition (distilled water) [23].
According to the literature, germination growth and radicle size are influenced by environment, including the exposure to chemical substances [17,23,24].Chemical compounds such as SA, CH, and H 2 O 2 participate in the regulation of physiological processes such as cell growth, respiration, seed germination, seedling development, and the cell wall formation of radicles [25][26][27].The evidence points to cell redox status as being the main mechanism associated with plant responses to many abiotic and biotic stresses [26,28] because reactive oxygen species (ROS), particularly H 2 O 2 , participate as signaling molecules involved in the controlling of many different physiological processes, both biotic and abiotic stress responses, throughout the activation of mitogen-activated protein kinase (MAPK) cascades, Ca2+ release, and nitric oxide (NO) synthesis, which are involved in the initiating of environmental stress responses during plant growth and development [26].In this regard, SA signaling is also connected to cell redox status, and the phenylalanine ammonia lyase (PAL; EC 4.3.1.5)is the key enzyme involved in SA biosynthesis [28].Similarly, CH induces the accumulation of ROS, such as H 2 O 2 in the cell wall, upon the wounding of cell tissues.This leads to the induction of plant defense enzymes, including PAL, which is also a main enzyme involved in phenolic compound biosynthesis [27].In addition, treatment with H 2 O 2 causes the accumulation of SA [25], thus activating SA signaling within the cell.
Decreased Content of Antinutritional Compounds in Chickpea Sprouts after Chemical Elicitation
The majority of legume plants, including chickpeas, have the ability to synthesize certain biologically active substances which are considered to be antinutritional compounds, causing deleterious consequences to the human digestive system, among other health issues [8,9].In this regard, the most widely recognized antinutritional compounds from chickpeas are the protease and amylase inhibitors, phytolectins, phytic acid, oligosaccharides, and few other compounds in traces [8,9].As observed, pigmented or desi type chickpea seeds (Patron and San Antonio) had lower levels of trypsin inhibitors (TIU) than the Blanoro raw (kabuli type) seeds, whereas the mean values for TIU in the seed samples of the desi type were higher compared with the kabuli cultivars grown in India [29].Even though all three cultivars had higher TIU values than those of the cultivar Blanco Sinaloa 92, also grown in Mexico [18], we can categorize the three cultivars into the low group on the basis of TIU values [29].Although we did not observe a clear tendency in the hemagglutinin activity (HU) and PA content that depended on the desi or kabuli type, the PA content in all three raw Mexican chickpea cultivars was in the range previously reported [8,[29][30][31].Even though these three cultivars also had a higher PA content than that of the cultivar Blanco Sinaloa 92 [18], we can categorize the three cultivars into the low group on the basis of PA content [29], thus confirming the influence of agricultural practices, wild species, intra-species biodiversity, and environmental factors in the overall composition of chickpeas [11].
According to the literature, sprouting is highly effective in reducing antinutritional compounds from chickpeas and is comparable to the other processing methods [9].As mentioned before, all three of the chemical elicitors applied had a major inhibition effect on TIU (up to 50.8%, p < 0.05) at the highest applied doses of each elicitor (2 mM SA, 7 µM CH, and 30 mM H 2 O 2 ) during germination.These results are in accordance with those previously reported by Mendoza-Sánchez et al. [23] in chemically elicited bean seeds.Conversely, optimal germination conditions (30 mM H 2 O 2 and 72 h germination time) did not significantly reduce the TIU value in the CS of Blanco Sinaloa 92 [18].
Although lectins are present at low levels in chickpeas [32], previous works have observed a decrease in CS from 77% up to the complete elimination of hemagglutinin activity (HU) after 3 and 8 days of germination, respectively [31,32].However, there are few studies that have investigated the effectiveness of chemical elicitation during germination to reduce lectins in pulses, particularly in CS.According to Mendoza-Sánchez et al. [23], chemical elicitation with SA, CH, and H 2 O 2 treatments further enhanced lower lectin contents in Dalia common beans, with the highest reduction of 48% in 0.1 mM SA-stressed sprouts as compared to control sprouts (6%) after 3 days of germination at 25 • C in darkness.
Reductions in the PA contents of cereals and legume seeds with sprouting have been frequently reported [1,17,23,27]; this has been attributed to an increase in phytase activities.In this regard, the PA contents in the three Mexican CS were in the range previously reported [26].In addition, it was found that this antinutrient is more prone to hydrolysis in the case of the kabuli (light-yellow-coated) type than in the desi (dark-coated) type of chickpea [1].In this work, we found a higher reduction in PA content in the pigmented San Antonio (desi type) than in the non-pigmented Blanoro (kabuli type) CS.
Overall, sprouting reduces the amounts of several antinutrients in the seeds; however, the reduction in PA has been found to be more profound than that of the other compounds [31].Similarly, a reduction in PA content of up to 40% was higher than those of the TIU values (up to 26%) and HU values (up to 33%) in all three Mexican CS, with the exception of the lectin content in the Blanoro CS (>50%).Furthermore, we observed decreases up to 50.8% and 73% for the TIU values and PA content in the chemically stressed sprouts, respectively, which further supports the fact that chemical elicitation during germination enhances the decrease in antinutritional compounds in legumes [18,19].In this regard, the results indicated that the 2 mM SA treatment was most effective in reducing the content of lectins and PA, as well as the TIU in the three Mexican CS.These results are in accordance with those previously reported for common beans [18], showing a decrease in the TIA value and PA content by 41.3 and 35.9%, respectively, after germination, which was further enhanced in the 2 mM SA-stressed sprouts by 54.3 and 56.5%, respectively, as compared to the raw bean seeds.
Increased Mono-, Di-, and Oligosaccharide Contents after Germination and Chemical Elicitation
Chickpeas are considered a better source of soluble carbohydrates than other grain legumes, particularly oligosaccharides and sucrose [2,8].The raffinose family oligosaccharides (RFOs) belong to low-molecular-weight, non-reducing saccharides that in seeds perform very important physiological functions in plant acclimation during seed development, maturation, and germination, among others [33,34].In this sense, the mono-, di-, and oligosaccharide contents in the three raw Mexican chickpea seeds were in the ranges previously reported for several genotypes and cultivars worldwide [2,31,32,35], but they were higher than the contents of stachyose and raffinose previously reported for fresh Blanoro raw seeds [5].
The evidence on the increases in sucrose after the sprouting of chickpea seeds was reported previously [31,33,35]; the increases are the result of the degradation of RFO catalyzed by the alpha(α)-galactosidase enzyme (EC 3.2.1.22,α-GAL), which provides carbon and energy to the growing seedling [33,36].In this regard, Arunraj et al. [35] measured α-GAL, RFO, sucrose, and reducing sugars (monosaccharides) during various stages of seed germination in chickpeas and identified the fact that growing tissues immediately start accumulating both sucrose and raffinose after imbibition (24, 48 and 72 h).Similarly, we also observed an increased in the sucrose, RFO, and monosaccharide contents of the three Mexican CS after 4 days of germination, as compared to those values for the raw seed.Similarly, Kalaivani et al. [36] noticed that the accumulation of sucrose and the breakdown of RFOs was regulated to maintain a constant supply of reducing sugars as an energy source during germination.This is of major interest as recent studies have reported that RFO can serve as a prebiotic and can improve the intestine microbial composition in healthy adults; other biological activities, such as anti-allergic, anti-obesity, and anti-diabetic activities and the prevention of non-alcoholic fatty liver disease through the inhibition of lipid accumulation, have also been reported [34,35].
Chemical Elicitation and Chickpea Varietal Effects on Phytochemical Profile of Chickpea Sprouts
According to the literature, sprouting is a highly effective process for increasing bioactive compounds in chickpeas [12][13][14][15][37][38][39].Therefore, in the present study the phytochemical profile of the germinated seeds was analyzed through an HPLC-MS system, confirming that the isoflavones biochanin A and formononetin [12,16,18,[37][38][39] were the major compounds present in the three Mexican CS (Table 3).Although the isoflavone contents of biochanin A, formononetin, and daidzein in the three control Mexican CS were in the range previously reported for chickpeas germinated under similar conditions (≈4 days of germination), it is worth mentioning that the control Blanoro CS showed the highest values compared to those reported in the literature [12,13,38,39].On the other hand, the genistein content in the three control Mexican CS were higher than those previously reported [38,39], and the San Antonio CS had the highest content.
The health benefits associated with chickpea consumption are attributed to the two main isoflavones, biochanin A and formononetin, which exert hypolipidemic, anticancer, anti-inflammatory, and antioxidant activities [2,8,13,14,37].Additionally, the phytoestrogens daidzein and genistein can be extensively metabolized in humans through the activity of the intestinal microbiota to produce metabolites such as equol and enterolactone [2,40].Similarly, the enterolactone metabolite is formed from the precursors secoisolariciresinol and matairesinol, and these microbial metabolites have increased antioxidant, anti-inflammatory, antineoplastic, and/or apoptotic activities compared to their precursors [40].As mentioned above, a high increase in the content of the flavonoid quercetin was consistently detected in the chemically stressed Mexican CS (Table 3).This phenolic compound possesses antioxidant and anticancer activities through the induction of the antioxidant enzymes and apoptosis, respectively [41,42].Likewise, important increases were obtained for the flavonoids kaempferol, epicatechin gallate, and epigallocatechin gallate, which exert anti-viral, anti-bacterial, anti-fungal, anti-inflammatory, antioxidant, anticancer, and cardioprotective properties [41][42][43][44]; epigallocatechin gallate is one of the most widely studied phytochemicals.More recently, the therapeutical anti-obesity and anti-diabetic potential of ethyl gallate was discovered [45], as was the anti-inflammatory and protective effect on intestinal mucosal integrity [44].
Phytosterols previously reported in chickpea oil and seeds were also detected and quantified in the three Mexican CS, and these included β-sitosterol, β-campesterol, avenasterol, stigmasterol, and avenasterol [2,16].It is important to mention that in the present study we also identified and quantified phytosterol compounds not previously reported or detected in chickpea sprouts, and these included campestenyl glucopyranoside, sitosteryl glucopyranoside, brassicasterol, and fucosterol (>20 µg/g), as well as ergosterol, stigmastanol, and stigmasteryl glucopyranoside (<20 µg/g).
Plant components such as phytosterols, particularly β-sitosterol, exert an anti-tumor effect on multiple malignant tumors, such as those of breast, gastric, lung, kidney, pancreatic, prostate, and other cancers, through molecular mechanisms, such as those which are pro-apoptotic, anti-proliferative, anti-metastatic, and anti-invasive on tumor cells [8,46].Likewise, the anti-diabetic, antioxidant, anti-inflammatory, and anti-lipidemic activities of phytosterols, such as β-sitosterol, stigmasterol, and their combinations, have also been demonstrated [20,47].
Regarding the content of saponins, the compounds previously reported in the literature in chickpea seeds were also detected and quantified in the CS, and these included soyasaponins Bb (I), βg (VI), Ba (V), and αg [48,49].Similarly, the germination increased the saponin content in several legumes such as Dalia beans after 3 days of germination [23] and chickpeas after 6 days in the dark [48].The observed increase in saponin level is most likely due to the activation and synthesis of the various enzymes that enhance the production of secondary metabolites, including the saponins, in response to abiotic stress and are associated with the chickpea cultivar [21].These results confirm that the three Mexican cultivars evaluated in our study differ in their response to chemical induction in terms of the elicitor and its concentration.In addition, it is important to mention that in the present study we identified and quantified compounds not previously reported or detected in chickpea sprouts, and these included phaseoside 1 and the soyasaponins Bb (I), Af, Bd, and αg.
The health benefits associated with dietary saponins include a wide range of activities, such as anticancer, antimutagenic, hypoglycemic, hypocholesterolemic, hypolipidemic and appetite suppressant, hepatoprotective (against fatty liver formation), immunomodulatory, neuroprotective, anticoagulant, anti-inflammatory, and antioxidant activities, in experimental studies on animals and in in vitro models [21,37,50].
It is important to highlight that the main contribution of our study was the identification of the effect of the germination process; the chemical elicitation in the enhancement of the germination yield, oligosaccharide content, and the concentration and diversity of some phenolic compounds, while diminishing antinutritional compounds, is highly dependent on the chickpea variety.In addition, one of the main findings was the increases detected in the concentration and the diversity of some saponins and phytosterols.Undoubtedly, our study contributes to the field of yield improvement and the possible functional properties of this legume due to the effect of chemical elicitation during germination.Nevertheless, we also consider the fact that our study has several limitations that are related to the sample size analyzed due to limited resources; therefore, it was only possible to focus on specific types of seed or sprouts, with the limitation of having to evaluate the phytochemical profile and some antinutritional factors under one controlled condition of temperature-RH at 4 days of germination, instead of over time; the analytical techniques used, such as UPLC and mass spectrometry, were not readily accessible in all the research laboratories, thus limiting the scope of the investigation; additionally, there was the limitation of scientific evidence on the relationship between the chemical composition of sprouts and their biological activity in in vivo or epidemiological studies through bio-directed phytochemical studies.Further studies are needed to overcome these limitations and to provide more comprehensive insights into the phytochemical composition of sprouts and their health benefits.
Materials
The chickpea seed (Cicer arietinum L.) varieties Blanoro, Patron, and San Antonio were developed and donated by the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP), Celaya, Guanajuato, Mexico.The chickpea seeds utilized in the trials were obtained during the fall-winter season of 2019-2020 at the Bajío Experimental Station located at 20 • 34 45.69 N. 100 • 49 11.49W and 1767 masl.The soil at the site is a Pelic Vertisol with a 1.79% O.M. content.The crop received only the pre-sowing irrigation and a fertilizer rate of 30-30-00 units of N-P-K at sowing.Weeds and insects (leaf miners) were taken care of using the conventional methods.According to the literature, the Blanoro variety is a larger non-pigmented seed which is similar to the kabuli type [5], whereas the Patron and San Antonio cultivars are smaller pigmented seeds like those of the desi type (Figure 2A-C).
tional properties of this legume due to the effect of chemical elicitation during germination.Nevertheless, we also consider the fact that our study has several limitations that are related to the sample size analyzed due to limited resources; therefore, it was only possible to focus on specific types of seed or sprouts, with the limitation of having to evaluate the phytochemical profile and some antinutritional factors under one controlled condition of temperature-RH at 4 days of germination, instead of over time; the analytical techniques used, such as UPLC and mass spectrometry, were not readily accessible in all the research laboratories, thus limiting the scope of the investigation; additionally, there was the limitation of scientific evidence on the relationship between the chemical composition of sprouts and their biological activity in in vivo or epidemiological studies through biodirected phytochemical studies.Further studies are needed to overcome these limitations and to provide more comprehensive insights into the phytochemical composition of sprouts and their health benefits.
Materials
The chickpea seed (Cicer arietinum L.) varieties Blanoro, Patron, and San Antonio were developed and donated by the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP), Celaya, Guanajuato, Mexico.The chickpea seeds utilized in the trials were obtained during the fall-winter season of 2019-2020 at the Bajío Experimental Station located at 20°34′45.69′′N. 100°49′ 11.49′′ W and 1767 masl.The soil at the site is a Pelic Vertisol with a 1.79% O.M. content.The crop received only the pre-sowing irrigation and a fertilizer rate of 30-30-00 units of N-P-K at sowing.Weeds and insects (leaf miners) were taken care of using the conventional methods.According to the literature, the Blanoro variety is a larger non-pigmented seed which is similar to the kabuli type [5], whereas the Patron and San Antonio cultivars are smaller pigmented seeds like those of the desi type (Figure 2A-C).
Germination Process and Chemical Elicitation Treatments
The seeds (50 g) were disinfected with 1.5% NaClO (1:6 w/v) for 30 min at room temperature, drained and washed until neutral pH, and then soaked in H 2 O for 24 h.The hydrated seeds were placed in trays between two layers of filter paper moistened with H 2 O or chemical elicitor solutions and were incubated at 25 • C with 70% RH in darkness in a temperature-controlled cabinet for 4 days.Germination was considered when the radicle emerged (Figure 2a-c).
Chemical elicitors were dissolved in distilled water at the following concentrations: salicylic acid (SA, 1 and 2 mM); chitosan (CH, 3.3 and 7 µM); and H 2 O 2 (20 and 30 mM) [23].Elicitors were freshly applied every day.Distilled water was used to germinate the seeds of the control group.Triplicate observations were carried out in independent experiments.
The germination percentage was determined based on total number of seedlings that fully emerged.The lengths of the radicle and plumule (cm) of the germinated seeds were measured and recorded daily.At the end of the experiment, the germinated seeds were immediately plunged into liquid nitrogen, ground in a mill, and passed through a mesh with a particle size of 1 mm; then, the flours were stored at −70 • C to ensure phytochemical stability until further analyses.
Quantitation of Antinutritional Compounds
The determination of trypsin inhibitor activity (TIA) was carried out as reported by Kakade et al. [51].The TIA was expressed as trypsin inhibitory units (TIU) per mg of dry sample, and one unit of trypsin was defined as the increase of 0.01 absorbance units at 410 nm of reaction mixture.The determination of lectins was carried out using the hemagglutinating activity technique according to the method previously reported by Grant et al. [52].One unit of hemagglutinating activity (HU) was defined as that contained in the amount of sample in the last dilution which caused 50% agglutination of the blood cells, and the results are expressed as HU/g of dried flour.The phytic acid (PA) content was determined according to the method of Frühbeck et al. [53], and the results are expressed as mg of PA/g of dry sample.
Sample Preparation and Carbohydrate Profile and Mono-, Di-, and Oligosaccharide Contents
The samples (25 mg of dried flours) were extracted with 0.5 mL of 80:20 methanol: water, and sonicated three times for 30 s, with resting of 15 s between cycles.The mixtures were centrifuged at 25,000× g for 5 min at 4 • C. The supernatants were recovered, and the extraction procedure was repeated with the residue; then, both supernatants were mixed.After the extraction, the samples were concentrated by using a rotary vacuum evaporator to evaporate the ethanol.Prior to injection, the samples were filtered through a 0.45 µm membrane (Millipore, Bedford, MA, USA).
Monosaccharides (fructose, glucose, and mannose), sucrose, and oligosaccharides (raffinose and stachyose) were identified and quantified by using an ultra-performance liquid chromatograph (UPLC) coupled to an evaporative light-scattering detector (ELSD; Xevo TQ-S, Waters Co., Milford, MA, USA).The samples (2 µL) were injected into a chromatographic column (Acquity BEH Amide, 100 × 2.1 mm, 1.7 µm, Waters Co) using (A) 80:20 (v/v) acetonitrile:water with 0.1% of ammonium hydroxide and (B) 70:30 (v/v) acetonitrile:water with 0.1% of ammonium hydroxide as a mobile phase under the following gradient conditions: 5% B at 0 min, 60% B at 5 min, and 60% B at 6 min, followed by a reset and equilibration step for 5.5 min at 250 µL/min.Standard solutions were freshly prepared and injected into the chromatographic system; the resulting peak areas were plotted against concentration for the calibration curve using the external standard method.The results are expressed as µg/g of dry flour [54].
Sample Preparation and Analysis of Polyphenol Profile
The sample preparation and the procedure extraction were performed according to the method reported by a previous study [23].The polyphenol profile was achieved in an HPLC system with a diode array detector (DAD) coupled to an ESI-sQ MS (Agilent 1200 and 1100 SL).
The samples (10 µL) were analyzed in a Zorbax ODS-C18 (150 × 4.6 mm, 5 µm) column at 30 • C. The mobile phase consisted of (A) 99:1 (v/v) water:formic acid and acetonitrile (B) at 1 mL/min under gradient conditions.Standard solutions were injected into the chromatographic system; the resulting peak areas were plotted against concentration for the calibration curve.The identification was carried out by comparing their spectroscopic and chromatographic characteristics with those of the standards.Those compounds without commercially available standards were tentatively identified using retention time arrangement, peak spectra, mass-to-charge ratio, MS fragmentation, and online metabolite databases.The results are expressed as µg/g of dry flour [55].
Sample Preparation and Analysis of Phytosterol Profile
samples (50 mg of dried flours) were extracted with 1 mL of 20:80 dichloromethane/ methanol and sonicated three times for 15 s, with resting of 30 s between cycles.Then, the mixtures were centrifuged at 14,000× g for 5 min at 4 • C, and the supernatants were evaporated to a concentrate.The samples were dissolved in 1 mL acetonitrile:acetic acid (99.9:0.1) and immediately injected (adapted from Tan et al. [56]).
For the phytosterol separation, the samples were analyzed in a high-performance liquid chromatograph (HPLC) coupled with a DAD and single-quadrupole mass spectrometer (sQ-MS, Agilent 1200).The samples (20 µL) were injected into a Zorbax ODS-C18 (150 × 4.6 mm, 5 µm) column at 35 • C. The mobile phase consisted of methanol (A) and water: acetonitrile 99:1 (B) at 0.8 mL/min under gradient conditions, as previously reported.The measurement of absorbances was performed at 205 nm.The identification was carried out by comparing their spectroscopic and chromatographic characteristics with those of the standards.Those compounds without commercially available standards were tentatively identified using retention time arrangement, peak spectra, mass-to-charge ratio, MS fragmentation, and online metabolite databases [55].For the quantification of phytosterols, β-sitosterol was used as a standard for the construction of the calibration curve using the external standard method.The results are expressed as µg/g of dry flour.
Sample Preparation and Analysis of Saponin Profile
For saponin extraction, the samples (100 mg of dried flours) were extracted with 1 mL of 80% methanol and stirred for 30 min.Then, the mixtures were centrifuged at 15,000× g for 5 min at 4 • C, and the solvent was evaporated.The samples were washed with 1 mL of 70:30 acetone: water, and the solvent was evaporated [23].
The samples (20 µL) were injected into a Zorbax ODS-C18 (150 × 4.6 mm, 5 µm) column at 35 • C. The mobile phase consisted of water containing 8 mM of ammonium acetate (A) and acetonitrile containing 0.1% formic acid (B) at 0.4 mL/min under gradient conditions.The measurement of absorbances was carried out at 310 nm.The identification was carried out by comparing their spectroscopic and chromatographic characteristics with those of the standards.Those compounds without commercially available standards were tentatively identified using retention time arrangement, peak spectra, mass-to-charge ratio, MS fragmentation, and online metabolite databases.For the quantification of the saponins, soyasaponin I was used as standard for the construction of the calibration curve using the external standard method.The results are expressed as µg/g of dry flour [23].
Statistical Analyses
All the results are expressed as mean values ± standard deviation (SD) of triplicate observations.All the variables were parametric; therefore, the data were analyzed by oneway analysis of variance (ANOVA), and differences among the treatments were determined by comparison of the means using the Tukey test and the Dunnet's test.The statistical significance level was α = 0.05.All the statistical analyses were carried out with JMP 5.0.1 software.
Conclusions
In conclusion, among the elicitors applied to chickpea seeds, SA induced the highest increase in the percentage of germination, radicle size, and oligosaccharide contents and decreased antinutritional compounds such as lectins, trypsin inhibitors, and phytic acid in all three Mexican chickpea varieties.In addition, SA induced the highest increase in the contents of phenolic compounds and saponins in the non-pigmented Blanoro cultivar.CH elicitation, on the other hand, exerted the greatest effect on the content of phenolic compounds, phytosterols, and saponins of the pigmented Patron and San Antonio varieties, whereas H 2 O 2 exerted the lowest effect on most of the contents.The diverse effects exerted by the elicitors SA, CH, and H 2 O 2 on the agronomical parameters, antinutritional compounds, and phytochemical profile suggest that these Mexican chickpea varieties
Figure 1 .
Figure 1.Percentage of germination and radicle size of chemically stressed chickpea sprouts (CS) of three Mexican cultivars.(A) Blanoro germination percentage; (B) Patron germination percentage; (C) San Antonio germination percentage; (a) Blanoro CS radicle size; (b) Patron CS radicle size; (c) San Antonio CS radicle size.Chickpea sprouts (CS) during 4 days of germination at 25 °C with 70% RH in darkness.Data are expressed as mean ± standard deviation of three experimental replicates.* Indicates significant difference (p < 0.05) using the Dunnet's test against control germination.Control: germinated seeds with distilled water.SA: salicylic acid; CH: chitosan; H2O2: hydrogen peroxide.
Figure 1 .
Figure 1.Percentage of germination and radicle size of chemically stressed chickpea sprouts (CS) of three Mexican cultivars.(A) Blanoro germination percentage; (B) Patron germination percentage; (C) San Antonio germination percentage; (a) Blanoro CS radicle size; (b) Patron CS radicle size; (c) San Antonio CS radicle size.Chickpea sprouts (CS) during 4 days of germination at 25 • C with 70% RH in darkness.Data are expressed as mean ± standard deviation of three experimental replicates.* Indicates significant difference (p < 0.05) using the Dunnet's test against control germination.Control: germinated seeds with distilled water.SA: salicylic acid; CH: chitosan; H 2 O 2 : hydrogen peroxide.
Table 1 .
Content of antinutritional compounds in chemically stressed chickpea sprouts (CS) of three Mexican cultivars.
Data are presented as mean ± standard deviation of three experimental replicates.Different letters in the same column indicate significant (p < 0.05) difference by Tukey test.
Table 2 .
Carbohydrate profile of chemically stressed chickpea sprouts (CS) of three Mexican cultivars.
Table 3 .
Polyphenol profile of chemically stressed chickpea sprouts (CS) of three Mexican cultivars.
Data are presented as mean ± standard deviation of three experimental replicates.Results are expressed as mg/g dry flour.Different letters in the same row indicate significant difference (p < 0.05) by Tukey test.Control: germinated seeds with distilled water.SA: salicylic acid, CH: chitosan, H 2 O 2 : hydrogen peroxide.Samples were analyzed on fourth day of germination.LDL: lower than detection limit.
Table 4 .
Phytosterol profile of chemically stressed chickpea sprouts (CS) of three Mexican cultivars.
Data are presented as mean ± standard deviation of three experimental replicates.Results are expressed as µg/g dry flour.Different letters in the same row indicate significant difference by (p < 0.05) Tukey test.Control: germinated seeds with distilled water.SA: salicylic acid, CH: chitosan, H 2 O 2 : hydrogen peroxide.Samples were analyzed on fourth day of germination.LDL: lower than detection limit.
Table 5 .
Saponin profile of chemically stressed chickpea sprouts (CS) of three Mexican cultivars.
Table 5 .
Cont.Data are presented as mean ± standard deviation of three experimental replicates.Results are expressed as µg/g dry flour.Different letters in the same row indicate significant difference by (p < 0.05) Tukey test.Control: germinated seeds with distilled water.SA: salicylic acid, CH: chitosan, H 2 O 2 : hydrogen peroxide.Samples were analyzed on fourth day of germination.LDL: lower than detection limit.
|
2023-08-31T15:23:39.511Z
|
2023-08-29T00:00:00.000
|
{
"year": 2023,
"sha1": "8a8b82aba74cbd1f9a5fa640c2afbb02c2a202d4",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2223-7747/12/17/3093/pdf?version=1693276310",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "14d0cc94512ebcc3e65e50def7d2b4d89ed27b9b",
"s2fieldsofstudy": [
"Agricultural And Food Sciences"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
16144708
|
pes2o/s2orc
|
v3-fos-license
|
Innovative psycho-educational program to prevent common postpartum mental disorders in primiparous women: a before and after controlled study
Background Universal interventions to prevent postnatal mental disorders in women have had limited success, perhaps because they were insufficiently theorised, not gender-informed and overlooked relevant risk factors. This study aimed to determine whether an innovative brief psycho-educational program for mothers, fathers and first newborns, which addressed salient learning needs about infant behaviour management and adjustment tasks in the intimate partner relationship, prevented postpartum mental health problems in primiparous women. Methods A before and after controlled study was conducted in primary care in seven local government areas in Victoria, Australia. English-speaking couples with one-week old infants were invited consecutively to participate by the maternal and child health nurse at the universal first home visit. Two groups were recruited and followed sequentially: both completed telephone interviews at four weeks and six months postpartum and received standard health care. Intervention group participants were also invited to attend a half-day program with up to five couples and one month old infants, facilitated by trained, supervised nurses. The main outcome was any Composite International Diagnostic Interview (CIDI) diagnosis of Depression or Anxiety or Adjustment Disorder with Depressed Mood, Anxiety, or Mixed Anxiety and Depressed Mood in the first six months postpartum. Factors associated with the outcome were established by logistic regression controlling for potential confounders and analysis was by intention to treat. Results In total 399/646 (62%) women were recruited; 210 received only standard care and 189 were also offered the intervention; 364 (91%) were retained at follow up six months postpartum. In women without a psychiatric history (232/364; 64%), 36/125 (29%) were diagnosed with Depression or Anxiety or Adjustment Disorder with Depressed Mood, Anxiety, or Mixed Anxiety and Depressed Mood in the control group, compared with 16/107 (15%) in the intervention group. In those without a psychiatric history, the adjusted odds ratio for diagnosis of a common postpartum mental disorder was 0.43 (95% CI 0.21, 0.89) in the intervention group compared to the control group. Conclusions A universal, brief psycho-educational group program for English-speaking first time parents and babies in primary care reduces de novo postpartum mental disorders in women. A universal approach supplemented by an additional program may improve effectiveness for women with a psychiatric history. Trial registration ACTRN 12605000567628.
Background
Postnatal mental health problems in women are associated with disability, reduced social participation and diminished caretaking capacity [1] and constitute a significant public health problem [2], which has proved difficult to prevent [3].
Nature and prevalence of postpartum mental health problems
The predominant focus of research, policy initiatives, clinical practice recommendations and health education has been postnatal depression, but there is increasing evidence that postnatal anxiety disorders are at least as common, but less well recognised than depression [4,5]. Brockington [1] in a review of postnatal psychiatric disorders concludes that women identified through screening as depressed actually have heterogeneous conditions including posttraumatic stress disorder, panic, phobic, obsessional and generalised anxiety disorders, adjustment disorders and depression. These are situationfocused, disabling and often reflect adversity [6]. Even among those who meet diagnostic criteria for major depression, severity ranges from mild to severe and most depression after childbirth is minor and not major [7]. Brockington [1] argues that 'postnatal depression' therefore has value as a lay term, but is imprecise as a clinical or a research construct.
This lack of clarity is reflected in widely divergent estimates of prevalence for probable depression as assessed by the Edinburgh Postnatal Depression Scale [8] in women in high income countries from 3.7% to 36% [9] and for diagnoses of postnatal generalised anxiety disorder from 4.4 to 8.5% [10]. There is less evidence about the prevalence of panic disorders in women after childbirth, but a much higher de novo onset of panic attacks in the first twelve weeks postpartum (10.9%) than expected by chance (0.92%) is reported [10]. In women with a history of panic attacks, symptoms tend to increase after childbirth, but not during pregnancy [10]. The prevalence of adjustment disorders arising in response to the birth of a baby has not been established. However, there is recognition that a proportion of mothers-of-infants who seek help for early parenting difficulties do not meet diagnostic criteria for depression or an anxiety disorder, but do have higher than population average scores on the Edinburgh Postnatal Depression Scale (EPDS) [8]. An expanded conceptualisation, including adjustment disorders, is required to recognise their needs [11].
Universal interventions to prevent postnatal depression
In addition to treatment for people with current mental health problems, a comprehensive approach to mental health service delivery must include mental health promotion and the prevention of mental disorders [12]. A range of interventions to prevent postnatal mental health problems, principally 'postnatal depression', have been tested in randomised controlled trials. These have included secondary prevention via indicated interventions for women with current clinically significant depressive symptoms, selective interventions for women identified by screening as at risk of developing depression and universal interventions offered to all women to reduce population prevalence [13][14][15][16][17]. Universal strategies are preferred; because even small reductions in population prevalence have a greater public health benefit than treating individuals who are already symptomatic [18]. They are also less stigmatizing and more likely to be used [19]. Systematic reviews have concluded that screening measures administered in pregnancy have low positive predictive values, probably because events after childbirth are more salient determinants of postnatal depression [15].
There have been seven trials of universal postnatal interventions offered to unselected populations of women who have recently given birth. Five tested strategies for use with individual women: debriefing with a psychologist about childbirth experiences [20] or a midwife listening visit during the postnatal hospital stay [21]; earlier-than-usual postnatal visit to a primary care physician [22]; ten three-hour home visits involving practical assistance and emotional care from a specifically trained support worker in the first postpartum month [23]; and an information pack containing specific, salient written information about maternal health, sleep and support needs and management of infant crying, with or without an invitation to attend a facilitated new mothers' group [24]. The other two trials assessed comprehensive community-based interventions which involved increasing the skills of primary health care nurses to identify women's physical and mental health conditions and initiate referral to appropriate health services [25,26]. Lumley et al [26] also provided specific training for general practitioners and community development aimed at enhancing local facilities and services for parents of newborns.
Of these varied strategies, only Lavender et al's midwife-listening visit [21] and MacArthur et al's primary health care intervention [25] were associated with reduced mental health problems in the treated compared to the control group. In Lavender et al's study, there was apparent selection bias in that 60% of participants were single women and the very high rates of selfreported depression and anxiety in the control arm (50% classified as having clinically significant symptoms) led reviewers to conclude that it is a 'true outlier' [27].
McArthur et al's intervention [25] was embedded in the UK National Health Scheme in which women are more likely to have an established relationship with a primary health care practice than in other health systems. All studies were adequately powered, analyzed by intention to treat, had properly concealed random allocation to trial arms and blinded assessment of outcomes. There were some methodological limitations: attrition greater than 20% at final assessment [22][23][24][25] and in one trial there was poor compliance with the intervention [24]. Dennis [3] concludes that these studies are generally of good methodological quality. The findings raise questions about why most of the interventions were not effective. A number of explanations emerge.
First, most conceptualised postnatal mental health problems as depression and/or general mental health morbidity and role functioning assessed by the SF-36 [28]. Anxiety was an outcome in only two trials. Lavender [21] reported reduced anxiety symptoms, but Priest [20] found no differences between groups in acute stress disorders. It is possible therefore that the interventions might have had undetected benefits for mental health problems other than depression or acute stress disorders.
Second, none of the trials analysed outcomes by prior psychiatric history, so it is also possible that the interventions had differential effects in women with and without a history of mental health problems, which were not detected [27,29].
Third, the mechanisms by which the interventions were proposed to reduce mental health morbidity did not target modifiable risk factors directly. Of the four well-established risk factors for depression after childbirth: personal psychiatric history, coincidental adverse life events, quality of relationship with the intimate partner and insufficient social support [30], the latter two are the most readily modified in the postpartum period. Most of the universal interventions addressed low social support, but through the provision of enhanced professional care outside the domestic sphere particularly in primary health care consultations, and not by seeking to improve the quality of a woman's intimate relationships [20,21,[23][24][25][26]29].
New avenues for prevention Intimate partner relationship
There is consistent evidence that the quality of relationship with the intimate partner is associated with postnatal mental health in women. It has been found to act both protectively and to increase risk. Women, who experience their partners as welcoming the pregnancy, and providing empathic support and encouragement, have better mood [31][32][33]. In contrast, women, who feel unable to confide in their partners or are experiencing conflict, poor communication or dissatisfaction in the relationship have worse mood [31,[34][35][36][37][38][39][40][41][42]. Although the evidence is consistent, few investigators have operationalised how difficulties in the intimate partner relationship are enacted in day-to-day behaviours. Only two randomised controlled trials of universal interventions for the prevention of postnatal mental health problems, both offered during pregnancy in North America, included partners. Fifty years ago Gordon et al [43] included men in two additional childbirth education classes, not only modelling that the work of parenting is a shared obligation, but also providing guidance to assist men to be sensitised to the demands of this life change for women. There were significantly fewer 'emotional upsets' in women in the intervention than the standard care group six months postpartum. Midmer [44] tested the effects of two additional 3-hour classes which focused on increasing: couples' appreciation of potential feelings of isolation, ambivalence, conflict, resentment and guilt in new mothers; and skills for managing relationships with extended family, a fretful baby, and the redistribution of household chores, using role-play and practice in problem solving and communication techniques in a standard childbirth education program for women and men. There was lower anxiety in women and men in the intervention than in the standard care group six weeks and six months postpartum. Gordon et al used nonstandardised outcome assessments and neither study controlled for cluster effects, but Gordon et al's study is cited as evidence of the importance of including women's partners in strategies to promote postpartum mental health [27].
Unsettled infant behaviour
To date, most investigations in this field have presumed that infants' behaviour reflects parenting factors [45], in particular that prolonged infant crying is a consequence of maternal depression [46]. Few have acknowledged that the relationship might be reciprocal and that infant behaviour might exert an adverse effect on a mother's confidence and affect. Infant behaviour, especially prolonged inconsolable crying, frequent night-time waking, short daytime sleeps and feeding difficulties are very common reasons for mothers of infants to seek help [47,48]. Mothers of infants who cry excessively report significantly more parenting stress and less sense of competence and efficacy than other mothers, and do not experience their infants as a source of positive reinforcement [49].
Occupational fatigue
Profound fatigue is widespread among mothers of newborns but is often normalised or trivialised, despite the adverse impact it exerts on normal daily functioning [50]. It has been regarded as symptomatic of depression, but an alternative view is that it arises because the unpaid workload of mothering a newborn is severely underestimated [51]. Poor infant sleep and maternal fatigue have been shown in a prospective investigation to precede the onset of depressive symptoms in women [52] Social theory of depression Brown and Harris's [53] social theory proposes that depression in women arises from experiences of entrapment and humiliation, which we argue are salient to the circumstance of mothering a newborn. The work of infant care is intrinsically confining. If the baby is responsive and rewards the mother by quieting to her soothing, smiling, interacting, suckling easily, and developing along at least an average trajectory, the baby provides gratification. In contrast, an infant who resists soothing, cries inconsolably, or is difficult to breastfeed can be experienced as critical and unappreciative. The work of mothering an infant and managing a household in which an infant lives is repetitive, isolated, never complete, and can be ungratifying. A mother of a newborn depends on her partner for recognition and affirmation of her endeavours and is especially vulnerable to his criticism, which can be humiliating. At this life phase women have increased dependence on intimate relationships, and reduced interactions with workplaces and the broader community.
None of the universal postnatal interventions included partners or attempted to modify day-to-day interactions in this relationship, included infants and addressed infant behaviour, or attempted to prevent occupational fatigue. We postulate that depression and anxiety in mothers of newborns can be conceptualised as reflecting poorly functioning intimate relationships, which are potentially modifiable mediated by fatigue.
The aim of this study was to assess whether What Were We Thinking! (WWWT) a brief, novel, highly structured, universal psycho-educational intervention for mothers, fathers and a first newborn, which addresses the intimate partner relationship, infant behaviour management, and thereby the mediating effects of occupational fatigue (see [29] for a detailed description) is effective in reducing the common maternal mental health problems of Depression or Anxiety or Adjustment Disorder with Depressed Mood, Anxiety, or Mixed Anxiety and Depressed Mood.
Study design
The What Were We Thinking! intervention was designed to be highly diffusible amongst families and social networks and includes attractive take home materials for ongoing reference. To prevent contamination of the standard care group with the intervention, we used a before and after controlled study design [54]. We first recruited and followed a control group who received standard primary postnatal care. Immediately after this, a second group was recruited and followed in the same way, but in addition to standard postnatal care, this group was invited to attend the intervention. Outcomes in the two groups were compared, controlling for baseline differences.
Setting
The study was conducted in seven local government areas (LGAs) in the Australian state of Victoria (population 5.2 million [55]). Diverse LGAs were selected by Socio-Economic Indices for Areas (SEIFA) to represent a range of areas across the spectrum of socioeconomic advantage and disadvantage [56]; three were from rural Victoria and four were in metropolitan Melbourne.
Recruitment to the control group took place from February to December 2006, and to the intervention group from February to December 2007. The intervention was conducted when babies were approximately four weeks old in easily accessed Maternal and Child Health Centres in the participating local government areas. Six-month follow-up of control group participants was completed in June 2007 and of the intervention group in June 2008.
Participants
All couples with healthy firstborn infants, sufficient English language proficiency to complete telephone interviews, both partners of which were willing to participate in the study and aged over 18 years, were eligible. Maternal and child health nurses provided verbal and written information about the study at the universal home visit and in the postnatal ward of a private maternity hospital. Interested couples provided contact details and were telephoned by research staff within one week. Women and men who agreed to participate returned individual signed consent forms by mail.
The intervention program
What Were We Thinking! (WWWT) is a highly structured, interactive, gender-informed, couples-based, psycho-educational program for parents and a first newborn to promote confident parental caretaking, optimise functioning in the intimate partner relationship, improve infant manageability and reduce common postnatal mental disorders in women [29].
Theoretical principles of the intervention
The theoretical principles of the WWWT program are, first, that postpartum anxiety is as important as depression and requires explicit attention; however, as depression and anxiety are not easily distinguished, they are addressed most effectively together. Second, that partner and infant behaviours can be modified to decrease those that contribute to maternal depression and anxiety and increase those that promote maternal confidence and sense of competence. Third, that women desire care from and gratification within these relationships and not increased care from health professionals. Fourth, that improvements in on-going day-to-day interactions are of fundamental importance to mental health promotion. Fifth, that this knowledge needs to be made available at a critical developmental stage and in a readily comprehensible form. Finally, that the language of the intervention is crucial and needs to challenge gender stereotypes and honour the work of mothering newborns.
Psycho-educational approach
WWWT is psycho-educational in addressing theoretically plausible psychological mechanisms using an educational approach to meet parents' learning needs. The program aims to: minimise experiences of humiliation through increasing fathers' understanding and empathy; counter experiences of entrapment by promoting infant care as a shared endeavour in which parents with comparable competence can permit each other independent or shared leisure [53]; and promote cognitive-rather than emotion-focused responses to infant crying by building skills to respond in non-avoidant ways. Together these strategies are expected to promote gratifying and rewarding intimate interactions rather than frustrating and diminishing ones, minimise maternal fatigue and thereby lead to increased parental confidence; more settled infant behaviour; and reduced depression, anxiety and adjustment disorders [29].
The educational approach addresses the provision of salient knowledge and opportunities to learn new skills. These are difficult to acquire through self-learning at this life stage because of fatigue, and because it is difficult for most people to distinguish between resources that are evidence-based, and those that constitute personal experience or opinion. Principles of adult learning are used and include group discussion, focused tasks to be undertaken individually and then discussed as a couple; practice in problem solving and negotiation, handson supported practice in infant settling, short talks and practical demonstrations. Anxiety is contained by a supportive, non-judgemental and knowledgeable facilitator. Very careful language-use is prescribed so that gender stereotypes are challenged, fathering and mothering are positioned as different but of equal importance and emotions are named and normalised without the use of psychiatric labelling.
Specific content
WWWT has 13 sections, grouped into two components: "About Babies" and "About Mothers and Fathers". About Babies includes sections about infant temperament, crying and fussing, recognition of tired cues, sleep needs, establishing feed-play-sleep routines of daily care and sustainable settling strategies. Opportunities to practise wrapping their babies and settling them to sleep are provided in the session. About Mothers and Fathers includes sections about differences between how parenthood had been imagined and reality; recalling the difficult and pleasing aspects of the baby's birth; recognising, naming and renegotiating the unpaid workload; acknowledging the losses and gains of parenthood; thinking about experiences within their families of origin that they wish to duplicate or to leave behind; and identifying gaps in support. This component provides language and strategies to assist couples to understand and respond effectively to changed needs for support, re-negotiate the paid and unpaid household workload fairly and manage the losses and gains associated with becoming parents. A folder containing a short book covering program content in accessible plain language and illustrations, and worksheets for each section is used during the program and taken home by couples for later reference.
Program delivery
Antenatal education for women and their partners is well established and there are high participation rates in Australia, but it does not continue postpartum when parents have high learning needs. Programs were held on Saturday mornings to maximise fathers' participation with groups of up to five couples and their babies. Materials were sent by mail to those who did not attend the face-to-face intervention.
Program fidelity
Program facilitators were three maternal and child health nurses, experienced in leading groups, who had attended a half day training session conducted by JF (clinical psychologist) and HR (adult educator). Training included didactic sessions to address relevant theory, role play to promote use of appropriate language to describe household work and challenge gendered stereotypes, and practice in supporting parents in infant settling techniques. The program was pilot tested with eleven couples and their infants and feedback was incorporated prior to implementation.
Fidelity to the program was upheld by adherence to the Facilitators' Handbook containing program theory, learning outcomes, group strategies and interactive worksheets for each section. Telephone and email communication with lead investigators was available to facilitators for an immediate response to unanticipated problems. The research coordinator provided informal weekly supervision and support, and formal supervision took place in face-to-face settings with JF and HR at bimonthly intervals throughout the implementation phase.
Standard care
Participants in the control group received usual primary health care.
Data sources
Data were collected by computer-assisted-telephoneinterviews (CATIs) conducted at approximately two weeks (baseline interview) and six months (follow up interview) postpartum.
The primary outcome was any diagnosis of a disorder meeting DSM IV criteria [57] for a Specific or Social Phobia, Panic with or without Agoraphobia, Generalized Anxiety Disorder, Dysthymia, Major or Minor Depressive Episode in the first six months postpartum as assessed by the relevant module of the Composite International Diagnostic Interview (CIDI) [58]. Adjustment disorders were diagnosed as the DSM IV criteria of feeling low and sad most of the day, nearly every day for at least two weeks since the birth of the baby and that it had not followed bereavement (Adjustment Disorder with Depressed Mood); a period of at least a month since the baby's birth of feeling worried, tense or anxious about everyday problems such as work, family or life with the baby (Adjustment Disorder with Anxiety), or both (Adjustment Disorder with Mixed Anxiety and Depressed Mood). These diagnoses were not made if the participant's other symptoms met criteria for Major or Minor Depressive Episode or Generalised Anxiety Disorder.
Psychiatric history was assessed by both study-specific questions and the CIDI as self reported lifetime history of treatment for alcohol or drug dependence, depression, eating disorder, or symptoms meeting criteria for panic attack in non life-threatening situations.
Potential confounders were assessed using study-specific questions at the baseline interview. Maternal factors included: age, country of birth, language spoken at home, marital, educational and occupational status, selfrated health, gravidity, appraisal of partner support and self-rated confidence on discharge from maternity hospital. Infant factors included: multiple birth, sex, birthweight, gestation at birth, age, health status and method of feeding. Standardised psychometric instruments were used to assess personality; depressive symptoms, infant behaviour and quality of relationship with the intimate partner (see Table 1).
Fidelity of intervention delivery was assessed by standard program evaluation checklists, completed by facilitators after each implementation. Facilitators rated how well the objectives of each of the 13 individual components of the program had been achieved, using a scale of 1 to 5 (1 = not at all; 5 = completely), and recorded details of unforeseen events that influenced delivery of the intervention.
Potential effect modifiers occurring between interviews were assessed at the follow up interview including: selfreported adverse life events, which were rated according to number and severity of events [59], and use of mental health or early parenting services. Reliably completed by parents, high correlation with tape recordings: for frequency (r = 0.85, p = 0.002) and duration (r = 0.90, p = 0.001) of episodes [74]. Procedure Interviews were conducted uniformly by trained telephone interviewers who had no other involvement in the study. Repeat contact attempts were made for up to one month at the preferred time that had been nominated at recruitment to participants who were unavailable. Attrition was minimised by the use of participant-provided additional contact telephone numbers. An AUD 25 shopping voucher to compensate for inconvenience was posted to participants who completed all interviews. CATIs were identical for intervention and control participants with the exception of specific questions in the follow up interview about the intervention program for participants in the intervention group.
Sample size
The intervention program was designed for groups of five couples. The sample size calculation included a correction to adjust for any correlation between responses within the same group. Assuming an intraclass correlation of 0.1 and an average cluster size of 5, the inflation factor for the intervention group was 40%. Thus the required ratio of control group participants to intervention group participants was 1:1.4. For a change of 10% in the prevalence of common mental disorders in women in the first six months postpartum, a two group continuity corrected chi-sq test with a 0.05 two-sided significance level will have 80% power to detect the difference between a control group proportion of 0.2 (20%) and a WWWT program group proportion of 0.1 (10%) (OR 0.444), with groups of 193 and 246, respectively, a total sample size of 439.
Ethics
Approval to conduct the study was provided by the Department of Human Services Victoria Human Research Ethics Committee and the University of Melbourne's Human Research Ethics Committee.
Data management and analysis
Maternal age and psychometric measures of maternal mood, personality and infant crying and fussing were continuous variables. Binary variables were constructed for: occupation, which was coded by the Australian Standard Classification of Occupations (ASCO 1 to 4 and ASCO 5 -8) [60]; whether English or another language was spoken at home; any or no psychiatric history; primi-or multigravid; unexpected or intended pregnancy; self rating as confident or anxious on discharge from maternity hospital; any breastfeeding or formula feeding and sufficient or insufficient help and support from partner at baseline. Study group was intervention or control and primary outcome was presence or absence of depression, or anxiety or adjustment disorders in the first six months postpartum. Baseline characteristics of those retained in the study were compared to those lost to follow-up. Univariate analyses were used to establish all significant differences in baseline variables between control and intervention groups, and to test associations between baseline variables and the primary outcome. Chi-square tests were used for binary variables; Mann-Whitney tests for ranked and ordered categorical variables, and t-tests for continuous variables. Statistical significance was set as p < 0.05. Between group differences are presented as means (95% confidence intervals) for continuous variables, and proportions (95% confidence intervals) for categorical variables The relationship between participation in the intervention and the outcome was tested by logistic regression in STATA [61], adjusting for potential confounders selected a priori from the univariate analyses. Variance Components Estimation was specified in the model and robust standard errors were used to adjust for clustering of individuals attending the same intervention program. Analysis was by intention to treat. Univariate analysis confirmed that the relationship between study group and the primary outcome was mediated by psychiatric history. An interaction term for psychiatric history and study group was therefore included in the model.
The model was adjusted for variables that differed between groups at baseline and for those associated independently with the primary outcome. All baseline data from a small set of cases were lost owing to server failure. Under an assumption of Missing Completely at Random, we used complete case analysis, thus excluding these cases and an additional ten cases with a small amount of missing data. Results of the model are presented as adjusted odds ratios and 95% confidence intervals and according to the TREND guidelines [62]. As a sensitivity analysis, we re-ran the model on the same number of cases, but excluding from the analysis the variable with the most missing cases. The odds ratios in this analysis were compared with the results of the original analysis.
Participants
Of the 646 eligible couples invited to participate, 399 women completed the first interview giving an overall recruitment fraction of 61.8%. Of these, 364 (91.2%) women completed the follow up interview. Women who could not be contacted by telephone for the follow up interview (n = 35) had significantly lower educational attainment, higher self-rated confidence on discharge from maternity hospital and reported fewer breastfeeding problems at the baseline interview than those who had completed both interviews (n = 364). Server failure led to loss of baseline data from eight participants and data for at least one baseline variable in the final model was missing for eighteen participants (5%). These were excluded, leaving 346 cases with complete data in the final model (Figure 1).
The baseline interview was conducted at mean (SD) 4.1 (2.3) weeks and the follow up interview at 27.6 (5.5) weeks postpartum. There were no significant betweengroup differences in infant age at either interview. Women in the intervention group were significantly older, more likely to speak English at home, to have completed post-secondary education, to be in professional or managerial employment and to be multigravid, and less likely to report that the index pregnancy had been unintended than women in the control group. Babies in the intervention group were more likely to be breastfed and cried and fussed for longer periods than those in the control group (see Table 2).
Intervention program fidelity, participation and satisfaction
A total of 37 intervention programs were implemented with women, their partners and babies, with a mean (SD) group size of 2.7 (1.6) families, at mean (SD) 6.6 (2.5) weeks postpartum. Facilitator evaluation checklists, which were available for 36/37 (97%) programs, showed that in each program all the individual sections had been delivered, and that the objectives of the individual sections were achieved (rated at least 4/5) in almost all of these (491/504; 97%).
A total of 120/189 (63.5%) of women in the intervention group attended the program and received a folder of written materials for take-home reference. Most of those who booked and confirmed, but did not attend a program, did not provide a reason, but those who did, cited unexpected illness or partner work or sporting commitments. The folder of written materials was posted to all those who did not attend in person. Most of the women (54/61; 89%) who were sent materials by mail reported at the follow up interview that they had received them.
Anonymous participant evaluation questionnaires completed by 98/120 (82%) women at the end of the intervention program revealed that 92 (94%) reported increased understanding of infant sleep needs, 81 (83%) increased understanding of infant temperament, 91 (93%) increased understanding of infant sleep and settling strategies, 71 (72%) could now talk more effectively about parenting with their partners and 64 (66%) already reported increased confidence in infant care.
Mental health outcomes
The primary outcome was a CIDI diagnosis of Depression or Anxiety or Adjustment Disorder with Depressed Mood, Anxiety, or Mixed Anxiety and Depressed Mood in the first six months postpartum (Table 3). Of the 117 women diagnosed as having experienced a disorder, 52 (44.4%) had no psychiatric history and were classified as having a de novo condition. The remainder (65/117; 55.5%) had histories of depression, panic, eating or substance abuse disorders and were classified as having a recurrent mental health problem.
In the group without a psychiatric history, the absolute risk reduction associated with the intervention was 0.14 (14%), and the relative risk reduction was 0.48 (48%). The original effect size on which our power calculation was based was conservative (10% difference); we showed a larger-than-postulated effect size, which reached significance in our smaller-than-necessary sample size.
Factors associated with mental health outcomes at 6 months
Use of mental health and residential early parenting services since the birth were potential effect modifiers as they had been used by significantly more participants in the intervention than in the control group. Univariate analysis revealed that use of both these services was associated with psychiatric history. Specifically, 69% of those who consulted a mental health practitioner (p < 0.001) and 83% (p < 0.01) of those who attended a residential early parenting service had a psychiatric history. There were no significant differences between study groups in the use of these services by women without a psychiatric history and therefore these were not included in the model. There were no significant between-group differences in number or severity of coincidental adverse events experienced between baseline and follow up.
In the final model, adjusting for all other factors, three variables remained significant predictors of the primary outcome: EPDS score at baseline interview, study group, and the interaction term for psychiatric history and study group, indicating that the effect of the intervention varied by whether or not respondents reported a psychiatric history (see Table 4).
For participants with no psychiatric history, being in the intervention group was associated with a significantly reduced odds (OR 0.43; 95% CI 0.21, 0.89; p = 0.022) of a diagnosis of a mental disorder. A linear combination of estimates was used to calculate the odds ratio associated with diagnosis of a mental disorder for participants with a psychiatric history in the intervention group (OR = 1.8; 95% CI 0.92, 3.71; p = 0.082) and demonstrated that being in the intervention group was not associated with significantly increased odds, compared to the control group, and thus the intervention did not cause harm.
Sensitivity analysis
The model was re-run on data from these same 346 respondents, excluding as predictor the number of hours the infant had cried or fussed in the past 24 hours, which had the highest number of missing values and was not a significant predictor. None of the odds ratios changed by more than 5%. The model was therefore re-run on all cases for which all the remaining predictors were available (n = 353) and the results were not different from those already reported.
Discussion
This before and after controlled study provides the first evidence that a brief, structured, universal, salient, gender-informed psycho-educational intervention offered in local settings within the first month postpartum appears to be effective in reducing the onset of the common postpartum mental health problems of Depression or Anxiety or Adjustment Disorder with Depressed Mood, Anxiety, or Mixed Anxiety and Depressed Mood in partnered mothers of a first infant who have no history of psychiatric illness.
There are insufficient comprehensive data to establish current and lifetime prevalence estimates of DSM IV Axis I disorders, including, anxiety and adjustment disorders in women in the first six months after childbirth. However, the lifetime prevalence (36%) and 6-month prevalence (32%) reported in this study are consistent with available evidence. A systematic investigation of 1066 women attending for routine antenatal care in Pisa, Italy reported that lifetime prevalence was 50.4% and that 26.3% met criteria for current disorders [63]. In Australia Hiscock and Wake [64] found in a systematic community-based survey of 738 mothers of sevenmonth-old infants that 15% scored more than 12 and another 18% scored 10 -12 on the EPDS.
It is well established that evaluation of complex health promotion interventions in real world settings is challenging [65]. In accordance with the criteria for evaluation of before and after controlled studies [66], we argue that the findings of this study are important, of notable magnitude and that relevant determinants were not ignored; participants in the first and second group met identical eligibility criteria and were recruited systematically from the general population of primiparous mothers and there was no co-occurring service change that might have contributed to a general trend in improvement of common maternal mental health problems over the time these data were collected. It is reasonable therefore to attribute the outcome to the intervention.
We acknowledge nevertheless that this study has limitations. Although attrition was low and participation fractions were adequate, the strength of this evidence is limited by potential selection bias because couples were not randomised to intervention or control conditions. There were differences in baseline characteristics which might have influenced the outcomes. Some of these might have been protective of mental health: women in the intervention group were older, were more likely to be in higher status occupations and were more likely to speak English at home and therefore to have easier social participation. Fewer had unintended pregnancies and more had established breastfeeding than women in the control group. However, the babies of women in the intervention group cried and fussed for longer in twenty-four hours than those in the control group, which might have increased the risk of depression in these women [51].
Although differences in a comprehensive set of relevant baseline characteristics were controlled for in analyses, the possibility that people with an unknown, but better adaptive capacity and lower likelihood of developing a common postpartum mental disorder, were recruited to the intervention group remains. However, the well-established determinants of postpartum depression: past psychiatric history, quality of intimate relationship (IBM Care and Control scores and ability to confide in and perceived support from the partner), current mood (EPDS scores and self-rated confidence in infant care) and vulnerable personality characteristics (VPSQ Vulnerability score) did not differ between groups at baseline. In addition, all participants received the "benevolent attention" of participation in detailed structured interviews about matters of direct relevance to their current experiences. Overall, we believe that the potential for bias is unlikely to account for the magnitude of the effect that was found. However, the findings should be interpreted with caution.
It is perhaps unsurprising that a brief (half day) intervention was insufficient to reduce the elevated risk of postpartum mental disorders in women with a history of mood, panic or eating disorders. However, there was very high satisfaction with the program in those who attended the face-to-face session: almost all participants found the knowledge and learning opportunities it provided relevant, timely and valuable. It did not cause harm.
The limited effectiveness for prevention of postnatal mental health problems in women with a past psychiatric history, suggests that a stepped approach in which a universal program is one element in a comprehensive mental health care system might be beneficial. The group with additional needs can be readily identified by primary care professionals by simple questioning, and referred for additional assistance, perhaps including specifically tailored supplementary programs. These could include other psychosocial interventions, for example, structured peer support which has been shown by Dennis et al [67] to be effective in preventing postnatal depression in women identified by screening as being at high risk; and individual consultations with maternal and child health nurses about how to manage infant sleep problems, found by Hiscock et al [68] to lead to significantly lower levels of depressive symptoms.
However, this novel intervention has merit. It appears that the approach is sound, and that offering a salient, acceptable, well-theorised, gender-informed, timely, nonstigmatising, psycho-educational program to couples and their first babies promotes optimal interactions with the benefit of reduced postpartum anxiety, depression and adjustment disorders in the majority of women. It suggests too that our hypothesised mechanisms of seeking to optimise interactions between mother, father and newborn so that empathy and affirmation are increased and criticism, irritability and insensitivity decreased; and to encourage both partners to participate more equally in the increased unpaid workload and infant care were effective.
This intervention is innovative in several ways. First, it includes partners and babies, and focuses on the modification of social risk factors, specifically the quality of day-to-day interactions in a woman's intimate relationships with her partner and her first infant. Second, it recognises that postpartum anxiety and adjustment disorders are common, but less well recognised than depression and require direct attention. Third, it is informed by plausible causal mechanisms that have not been delineated in previous trials. These include: that infant crying and resistance to soothing can arouse anxiety, helplessness and a sense of incompetence; that women experience many unrecognised losses in having a baby; and that disabling occupational fatigue is widespread. Together these can be conceptualised as experiences of entrapment, humiliation and grief which increase potential for depression and anxiety [53]. Rather than positioning men and infants as victims of a woman's mental state, it conceptualises intimate relationships as reciprocal and modifiable. Fourth, the intervention is gender-informed in naming infant care and household tasks as work and making it explicit that failure to recognise the unpaid workload or to share it fairly contributes to occupational fatigue and interpersonal conflict. Finally, rather than just offering support, the intervention was psycho-educational in providing salient knowledge, active learning opportunities and skills training at a critical life-stage. It is framed as a health promotion activity of universal relevance in response to heightened learning needs common to all new parents and is intended not to be stigmatising. This intervention sought to address possible limitations in the existing prevention trials and distinguished between de novo and recurrent mental health problems. It addressed diverse mental health problems including anxiety and adjustment disorders and not just depression. It aimed to modify salient aspects of a woman's intimate social environment, rather than aspects of her individual functioning. It is integrated into primary health care in a local setting and capitalises on an optimal shared learning environment [69].
The study was conducted in seven study sites, which were chosen to achieve diversity in socioeconomic status and in rural and metropolitan locations, and involved systematic recruitment of all couples meeting inclusion criteria. Participation in the study and the intervention appears to have been more appealing to people who were better educated and occupied higher socioeconomic positions. Although the program is intended to be universal, it is unlikely that the face-to-face professionally-facilitated model will reach everyone. This suggests that other modalities might be required to make this knowledge and these skills more widely available to people with lower language skills and emotional literacy. The effectiveness of the intervention was tested when implemented by trained, highly skilled practitioners and it is unknown whether this intervention will be effective when integrated into existing standard primary health care. This novel approach now requires testing in a pragmatic cluster randomised controlled trial.
Conclusions
A universal, brief psycho-educational group program for English-speaking, first-time parents and babies in primary care appears to reduce de novo common postpartum mental health problems in women. A universal approach supplemented by an additional program may improve effectiveness for women with a psychiatric history.
|
2017-08-28T12:01:00.743Z
|
2010-07-23T00:00:00.000
|
{
"year": 2010,
"sha1": "96a33a736e902c2781bc30295ee03ad1d32f8d2e",
"oa_license": "CCBY",
"oa_url": "https://bmcpublichealth.biomedcentral.com/track/pdf/10.1186/1471-2458-10-432",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "96a33a736e902c2781bc30295ee03ad1d32f8d2e",
"s2fieldsofstudy": [
"Medicine",
"Psychology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
252596279
|
pes2o/s2orc
|
v3-fos-license
|
On the general position numbers of maximal outerplanar graphs
A subset $R\subseteq V(G)$ of a graph $G$ is a general position set if any triple set $R_0$ of $R$ is non-geodesic in $G$, that is, no vertex of $R_0$ lies on any geodesic between the other two vertices of $R_0$ in $G$. Let $\mathcal{R}$ be the set of general position sets of a graph $G$. The general position number of a graph $G$, denoted by $gp(G)$, is defined as $gp(G)=\max\{|R|:R\in\mathcal{R}\}$. In this paper, we determine the bounds on the gp-numbers for any maximal outerplane graph and characterize the corresponding extremal graphs.
Introduction
All finite connected graphs considered in this paper are simple and undirected. Let G be a graph with vertex set V (G) and edge set E(G). For simplicity, set n = |V (G)|, the order of a graph G. For a vertex v ∈ V (G), let N G (v) and N G [v] denote the open neighborhood and the closed neighborhood of v, respectively; thus N G (v) = {u : uv ∈ E(G)} and N G [v] = {v} ∪ N G (v). We denote by d G (v) = |N G (v)| the degree of v. A vertex with degree d is said to be a d-vertex in G. By ∆(G) we denote the maximal degree of G. The distance d G (x, y), as usual, is the number of edges on a shortest path between x and y in G. A shortest x, y-path is called an x, y-geodesic in G. For any positive integer i ≥ 2, let P i be a path on p 1 , p 2 , . . . , p i with natural adjacencies. As usual, K n denotes the complete graph on n vertices.
A graph is planar if it can be embedded in the plane so that no two edges intersect geometrically except at a vertex to which they are both incident. A plane graph is a planar graph with a fixed embedding in the plane. A plane graph divides the plane into connected regions called faces. The unbounded region is called the outer face and each bounded region is called an inner face. A plane graph G is outerplanar if it has an embedding in the plane such that all vertices belong to the boundary of its outer face (the unbounded face). An outerplanar graph G is maximal if G + uv is not outerplanar for any two non-adjacent vertices u and v of G and such graph is called a maximal outerplanar graph (or just an MOP in [7]). A maximal outerplanar graph embedded in the plane is called a maximal outerplane graph.
We note that any MOP has a unique Hamiltonian cycle [25] and any MOP is a triangulation graph (or a triangulated disc), that is, a plane graph such that all its faces, except the outer face, are bounded by K 3 . Let f be an inner face of a maximal outerplane graph G. Then f is isomorphic to a K 3 . If f is adjacent to the outer face, i.e., there exists at least one common edge between f and outer face, then we say that f is a marginal triangle; otherwise we say that f is an internal triangle. An MOP G without internal triangles is called striped.
Since the definition of outerplanar graph was proposed, it has been intensively studied [2,6,10,11,18,22,30]. Motivated with the application of maximal outerlpanar graph, we consider the general position problem for maximal outerlpanar graph in this paper.
The general position problem in graphs was recently introduced by Manuel and Klavžar [19] and is now well studied in graph theory (see, for example, [12,13,16,17,21,23,24,29] for recent papers on this topic). A subset R ⊆ V (G) of a graph G is a general position set if any triple set R 0 of R is non-geodesic in G, that is, no vertex of R 0 lies on any geodesic between the other two vertices of R 0 in G. Let R be the set of general position sets of a graph G. The general position number of a graph G, denoted by gp(G), is defined as gp(G) = max{|R| : R ∈ R}. The general position number will be denoted by gp-number briefly. The general position set of order gp(G) will be called gp-set of a graph G.
By now some results on the gp-number are obtained for various classes of graph operation. Several general bounds on the gp-number were presented [19]. The gp-numbers were determined in [20] for a large class of subgraphs of the infinite grid graph and the infinite diagonal grid graph. And in [1], a characterization of general position set was given and the gp-numbers of bipartite graph and its complement graph were proved. Recently the gp-numbers in Cartesian products of graphs have been further investigated in [8,14,27,28]. A sharp lower bound on the gp-number for the Cartesian products of graphs was determined and the gp-numbers for joins of graphs, coronas over graphs, and line graphs of complete graphs were also characterized [8]. The gp-number of the Cartesian product of n-dimensional grid graph was proved in [14]. In particular, it was proved in [28] that the upper bound on gp-number for Cartesian products of two graphs is sharp and the equality holds if and only if two graphs are both generalized complete graphs. Moreover, the authors showed that the gp-number is additive on Cartesian products of trees [27]. In addition, the gp-numbers of other product graphs were investigated and connected with strong resolving graphs [15].
In this paper we consider the gp-number of maximal outerplane graphs. Some definitions and basic results are given in Section 2. In subsequent Section 3, we determine the bounds on the gp-numbers for any plane graph and characterize its corresponding extremal graphs. In Section 4, we determine an upper bound on the gp-number for any maximal outerplane graph, and also give some corresponding extremal graphs when the upper bound is achieved. In addition, we also prove that the bounds on the gp-number for a maximal outerplane graph containing internal triangles or not, respectively.
Preliminaries
In this section, we define some concepts and introduce the notations, as well as some results needed later.
For a subset S ⊆ V (G), let G[S] be the induced subgraph of a graph G by S. Let H G (for short H G = H) be a Hamiltonian cycle of a maximal outerplane graph G. Suppose u and v are two vertices on H. We call the path on H that goes in the clockwise direction from u to v the u, v-segment of H, and other path the v, usegment of H, denoted by S uv and S vu , respectively. The interval I G (u, v) between vertices u and v is a vertex subset which consists of all vertices lying on some u, vgeodesic of G, that is, A graph F is said to be a minor of a graph G if F can be obtained from G by a series of vertex deletions, edge deletions, and edge contractions. Given a graph F , a graph G is said to be F -minor-free if no minor of G is isomorphic to F . A graph G is outerplanar if and only if it does not contain K 4 and K 2,3 as minors ( [26]), thus any MOP is K 4 -minor-free and K 2,3 -minor-free.
Next, we will give the definition of fan which is important for characterizing of extremal graphs of our main results.
Definition 1. A connected graph G of order at least 3 is a fan if it has only one (n − 1)-vertex, only two 2-vertices, other vertices (if exist) have degree 3.
For any positive integer n ≥ 2, it is observed that a fan is {v} ⊕ P n−1 (the join of an isolated vertex v and P n−1 ), denoted by F n−1 , and here the vertex v is called central vertex of fan. A maximal fan subgraph of an MOP is a fan which cannot be enlarged by adding a vertex.
In this paper, we always set [n] = {1, 2, . . . , n} for a positive integer n. For notations and terminologies not defined here, see [4]. With the above concepts and notations we can recall the following known results.
Lemma 2. ( [5]) Let G be a maximal outerplane graph of order n ≥ 4. If G has k internal triangles, then G has k + 2 2-vertices. (ii) gp(G) = n − 1 if and only if G ∼ = K k n with 1 ≤ k ≤ n − 3 or G is a non-trivial generalized complete graph.
It is easily seen that gp(G) ≥ 3 for any maximal outerplane graph G. Note that G is 2-connected since a maximal outerplane graph has Hamiltonian cycle. Then we have ∆(G) ≥ 2 for any maximal outerplane graph G. Next, we will give a general lower bound on the gp-number for any maximal outerplane graph G with respect to ∆(G). Proof. Assume, without loss of generality, that d G (u) = ∆(G) with u ∈ V (G). It is easy to see that the result holds if ∆(G) ≤ 4. So we may let ∆(G) ≥ 5 in the following. Let N G (u) = {v 1 , v 2 , . . . , v k−1 } with k − 1 = ∆(G). By Lemma 4, we know that N G [u] forms a maximal fan G u with the central vertex u in G. Assume Then |S 1 | = 2j and |S 2 | = 2j + 1 since k ≥ 6. If k = 3j or k = 3j +1, we have d G (w 1 , w 2 ) ∈ {1, 2} for any two vertices {w 1 , w 2 } ⊆ S 1 . Furthermore, if d G (w 1 , w 2 ) = 1, then w 1 and w 2 are two adjacent vertices on P k−1 .
It is a contradiction. Therefore, S 1 is a general position set of G and gp(G) ≥ 2j = 2k 3 . Analogously, S 2 is also a general position set of G, we have gp(G) ≥ 2j + 1 = ⌊ 2k 3 ⌋, as desired. Lemma 8. Let S be a general position set of maximal outerplane graph G with Proof. Note that d G (x) ≥ 2 for any x ∈ V (G). So we may suppose d G (x) ≥ 3, as otherwise the result holds obviously. Assume further, without loss of generality, that |N G (x) ∩ S| ≥ 3 and N G (x) ∩ S = {v 1 , v 2 , . . . , v k } with k ≥ 3. There must be two vertices, say v i and v j , such that d G (v i , v j ) = 2 with {i, j} ⊆ [k], otherwise the induced subgraph on N G (x) has a K 4 minor. It is impossible.
By Lemma 8, it is also immediate that the following holds.
Corollary 9. Let S be a general position set of any maximal outerplane graph G.
Next, we will characterize the induced subgraphs on the general position set for a maximal outerplane graph G.
Suppose first that d H (x, y) = d H (y, z) = 1. Then d H (x, z) = 2. As {x, y, z} ⊆ V (K 3 ), our assumption implies that d G (y) = 2 and d G (x, z) = 1. Since |S| ≥ 4, there must be another vertex w ∈ S \ {x, y, z}. Applying Lemma 5, we have Now assume, without loss of generality, that d H (x, y) = 1 and 2 ≤ d H (x, z) ≤ ⌊ n 2 ⌋. There also must be another w ∈ S \ {x, y, z} since |S| ≥ 4. It follows from This completes the proof.
Maximal outerplane graphs
In this section, we first give the formula of the gp-number for a fan. Then we will determine the upper bound on the gp-number of a maximal outerplane graph G and characterize some corresponding extremal graphs when the bound is achieved. For convenience, G denotes a maximal outerplane graph in this section unless otherwise specified.
Observation 11. Let H be a Hamiltonian cycle of G and S uv be an u, v-segment By symmetry, based on the above Observation 11, we also have Next, we will prove that there is one common neighbor vertex for two adjacent vertices on H as follows.
The proof is simple if n ≤ 6. So we may assume that n ≥ 7 in what follows.
Without loss of generality, Next, we will give the formula of the gp-number of fan which plays a crucial role in the proof of our main result.
Assume further that R be a gp-set of G. Suppose n ≥ 8, as otherwise the proof is simple. Note that G is a fan and ∆(G) ≥ 7. It follows from Lemma 7 that |R| ≥ 5. Applying Corollary 9, it is obvious that v ∈ R. For any positive integer k ≥ 2, based on the value of n, we consider the following cases.
Suppose first that n = 3k. Since G is a fan and ∆(G) = n − 1, by Lemma 7, it is clear to see that gp(G) ≥ 2k. Let V 3 (i) be the set of consecutive three vertices of Analogously, we can get gp(G) = 2k, n = 3k + 1 2k + 1, n = 3k + 2.
In consequence, we can conclude that gp(G) = ⌊ 2n 3 ⌋, completing this proof. We now construct two classes of connected graphs, which we will use in the main result of proofs. For the sake of clarity, we give the definitions as follows.
These graphs G 1 (j) and G 2 (j) are shown in Figure 1.
Lemma 7 gives a general lower bound on gp-number for any maximal outerplane graph. Next, we will give the proof of our main result. Proof. Let H be the Hamiltonian cycle of G and h 1 ,h 2 ,. . . ,h n be its vertices in a cycle clockwise order. Assume further that R is a gp-set of G and V 3 is the set of consecutive three vertices on H. Recall that gp(G) ≥ 3 for any maximal outerplane graph. The upper bound holds obviously if gp(G) ∈ {3, 4} since n ≥ 6. So we may assume that |R| ≥ 5. In order to obtain the upper bound, the following claim is important for our proof.
. Assume first that n = 3k. Note that k ≥ 2 is a positive integer as n ≥ 6.
For the case n = 3k + 2, we can prove that |R| ≤ 2k + 1 = ⌊ 2n 3 ⌋ as similar to the above. Consequently, we get gp(G) ≤ ⌊ 2n 3 ⌋. Next, we will prove that the necessary and sufficient condition when the equality holds. Let k ≥ 2 be a positive integer. Assume first that n = 3k or n = 3k + 2. By It is easy to prove that S 1 of order 2k is a general position set of G. By the above argument, we can get So we prove the sufficiency only in what follows. Let G be a maximal outerplane graph, which satisfies gp(G) = ⌊ 2n 3 ⌋. Suppose n ≥ 12, as otherwise the proof is simple. Note that |R| ≥ 8. Indeed, |V 3 (i) ∩ R| ≥ 1 for any i ∈ [k]. If there exists one V 3 (i) such that |V 3 (i) ∩ R| = 0, by Claim 1, we can get It contradicts our assumption. Combining this with Claim 1, we have 1 ≤ |V 3 (i) ∩ R| ≤ 2 for any i ∈ [k]. Similar to V 3 (i), V 4 is defined as the set of consecutive four vertices of H. To prove our result, the following claim will be useful to the rest proof.
z} with clockwise order on H. By Claim 1, we may assume, without loss of generality, that Then there must be a vertex u * ∈ N G (u) \ {v} or z * ∈ N G (z) \ {w} such that d G (u * , z) = 1 or d G (u, z * ) = 1. Without loss of generality, assume that there exists a vertex u * ∈ N G (u) \ {v} such that d G (u * , z) = 1. By Observation 11, u * = h ′ . Since G is a maximal outerplane graph and d G (v, z) = 2, we have d G (h ′ , w) = 1. Hence V 4 ⊆ N G (h ′ ). Moreover, the result also holds if d G (u, z * ) = 1 for some z * ∈ N G (z) \ {w}. This completes the claim.
Assume now that h ∈ S \{h 3 }, and without loss of generality let h = h 3i for some i ∈ [k]\{1}. Then it follows that h 2 is a 3-vertex in G since {h 1 , h 2 , h 3 , h 4 } ⊆ N G (h). As h 2 , h 3 , h 4 and h 5 are four consecutive vertices on H, by using Claim 2, we have h 5 ∈ N G (h). Similarly, we can get h 3j−2 , h 3j−1 ∈ N G (h) for any j ∈ [k] and h 3j ∈ N G (h) for any j ∈ [k] \ {i}. Therefore G ∼ = F n−1 .
Suppose on the contrary that |R * | ≥ 3, and without loss of generality let h 1 , h s , h t ∈ R * with 3 ≤ s ≤ t ≤ ⌊ n 2 ⌋. It follows from Claim 1 that gp(G) ≤ 2× (n−7) 3 +3 = 2k − 1 if s = 3 and t = 5, a contradiction. While s = 3 and t ≥ 6, by Claim 1, we It is a contradiction. Using the similar arguments as the above, we can get the same contradiction if 4 ≤ s ≤ t. Thus, we can obtain |R * | ≤ 2.
Based on the order of R * , we divide into following subcases to proof this case.
In this subcase, we may, without loss of generality, let The proof is simple if n ≤ 9. Assume that n ≥ 10 in the following.
Assume that x = 3, that is, {h 1 , h 3 } ⊆ R * . According to Claim 1, we may without loss of generality, let R As h 3 , h 4 , h 5 and h 6 are four consecutive vertices of H, applying Claim 2 and Lemma 10, then we have {h 3 , h 4 , h 5 , h 6 } ⊆ N G (r) for somer ∈ R. Ifr = h 2 , then h 3 is a 2-vertex with two neighbor vertices h 2 and h 4 in G. It also implies that h 5 is a 3-vertex having three neighbors h 4 , 6 and h 2 in G. By Lemmas 5 and 10, we have d G (h 5 , h 8 ) = 2 since {h 5 , h 6 , h 8 } ⊆ R, which means that h 8 ∈ N G (h 2 ). Then we also have h 7 ∈ N G (h 2 ) since G is a maximal outerplane graph. By similar, we can get h 9 ∈ N G (h 2 ). Repeating the similar discussion above, it is observed that h ′ ∈ N G (h 2 ) for any h ′ ∈ V (H) \ {h 2 }. Thus we have G ∼ = F n−1 .
Let ] for somer, r 1 ∈ R. Analogously to the previous discussion, we can get G ∼ = F n−1 ifr =r 1 .
Next assume thatr ∈ R 2 andr 1 ∈ R 1 . Ifr = h n andr 1 = h 2 , it implies that h n−1 and h 3 are two 2-vertices of G. Similar to the above, we can obtain that , which leads to h 1 ∈ I G (h n−1 , h 3 ) contradicting with our assumption. Thus we have d G (h n , h 2 ) = 1, then G ∼ = G 1 (1). For the remaining cases, we can apply the similar method as the above repeatedly and obtain that G ∈ G * . Case 3. n = 3k + 2. In this case, gp(G) = 2k + 1. We can get G ∼ = F n−1 , this proof is similar to that of Case 1, hence omitted here.
Maximal outerplane graphs without internal triangles
In this subsection, we determine the bounds on the gp-numbers for striped maximal outerplane graphs. For that, we prove one lemma that will be useful in this proof.
Having proved Lemma 7, it would be interesting to know when the lower bound is achieved. For the sake of this, we introduce the following definition.
if and only if 0 < |i − j| ≤ 2, then we call G n the straight linear 2-tree.
We observe that the straight linear 2-tree G n is a striped maximal outerplane Proof. By the definition of G n , it is clear to see that ∆(G) = 4 if G ∼ = G n . For the converse, suppose that the maximal degree of G is 4. Let H be a Hamiltonian cycle of G with V (H) = {h 1 , h 2 , . . . , h n } and S uv be an u,v-segment on H for any u, v ∈ V (H). In order to obtain our result, we will prove that two claims in the following.
Claim 1. There are exactly two 2-vertices in G.
Suppose that there exists an internal triangle
T is a marginal triangle contradicts our assumption. Since T is an internal triangle and ∆(G) = 4, As G is a maximal outerplane graph, there must be one vertex u s such that the induce subgraph on {h i , h j , u s } is a K 3 for s ∈ [ℓ]. It follows that max{d G (h i ), d G (h j )} ≥ 5, which is a contradiction. Thus, there exists no internal triangle in G. By Lemma 2, G has exactly two 2-vertices, completing this claim. It means that max{d G (z), d G (w)} ≥ 5, which contradicts ∆(G) = 4. Thus we have d H (z, w) = 1, as desired. Assume that h i , h j and h k are any consecutive three vertices of H in clockwise order. Then N H Note that n ≥ 7. There must be another ver- Otherwise, there must be a cycle C 4 in G, which is impossible. Thus, we can get Thus {h i , h k , h} forms an internal triangle in G, which contradicts Claim 1. Hence, there exactly one 3-vertex in h i and h k . Applying Claim 1, it means that G has exactly two pairs of consecutive vertices u and v such that u is a 2-vertex and v is an 3-vertex.
Without loss of generality let h i be an 3-vertex of G and N H ( Let h i and h j be two 2-vertices of G with i < j. Then we will prove that . v ℓ h i are two segments of G with its clockwise order on H and min{k, ℓ} ≥ 1. Then k + ℓ = n − 2. It also follows that Suppose, without loss of generality, that ℓ > k, i.e., ℓ = k + 2. Since h i and h j are two 2-vertices of G, then we have d G (u 1 , v ℓ ) = 1 and d G (u k , v 1 ) = 1. By the above argument, we know that both h i and h j have a neighbor vertex of degree 3 in G. Without loss of generality, let d G (u 1 ) = 3. Then d G (v ℓ ) = 4. It implies that there must be another vertex h ∈ N G (u ℓ ) except h i , u 1 and v ℓ−1 , by Claim 2, we can obtain d H (u 1 , h) = 1. It leads to h = u 2 , that is, Recall that ℓ > k and G has n − 4 4-vertices. Analogously, d G (u s , v ℓ−s+1 ) = 1 and d G (u s , v ℓ−s+2 ) = 1 for any s ∈ [k] \ {1}. Since d G (u k , v 1 ) = 1 and ℓ − k + 1 ≥ 2, it follows that d G (u k ) ≥ 5, which is impossible. Similarly, we can get the same contradiction if d G (u 1 ) = 4. In consequence, we can conclude that d H (h i , h j ) = ⌊ n 2 ⌋. By the above Claims and arguments, we can obtain that G ∼ = G n if ∆(G) = 4, as desired.
With the help of Lemma 18, we can present an extremal graph for achieving the lower bound in Lemma 7. According to Theorem 16, we determine an upper bound on the gp-number of G without internal triangle and characterize its corresponding extremal graphs in the following.
Theorem 19. Let G be a striped maximal outerplane graph of order n ≥ 5. Then we have with left equality iff G ∼ = G n and right equality iff G ∈ {F n−1 , F (1; n), F (n − 2; n), Proof. Let R be a gp-set of G. By Theorem 16, the upper bound holds clearly. Furthermore, since G has no internal triangles, then we have G ∈ {F n−1 , F (1; n), F (n − 2; n), G 1 (1), G 2 (t)}. We observe that |R| ≥ 3. Next, it needs to prove that the left equality holds. Assume first that G is a maximal outerplane graph satisfying gp(G) = 3. Recall that G is a 2-connected graph. Applying Lemma 7, we can get gp(G) ≥ 4 if ∆(G) ≥ 5. Thus, based on our assumption, we have 2 Assume, without loss of generality, that V (S h 0 h d )∩R = {x, y} and V (S h d h 0 )∩R = {z, w} with clockwise order on H. Since H is a cycle, there must be three vertices in R ∩ V (H) such that they lying on the same geodesic in H. Without loss of generality, let y ∈ I H (x, z). By the structure of G, it is obvious that d G (x, z) = d G (x, y) + d G (y, z). Then we have y ∈ I G (x, z), it contradicts {x, y, z} ⊆ R. The similar contradiction can be obtained if z ∈ I G (y, w) or w ∈ I G (x, z). Thus we have |R| ≤ 3.
Using the similar arguments as above, we can obtain the same conclusion if n is odd. Hence, |R| = 3 if G ∼ = G n . We complete this proof of the theorem.
Maximal outerplane graphs with internal triangles
In this subsection, we concentrate on the bounds on the gp-numbers of G with internal triangles. We will give a necessary condition for attaining the lower bound. And we also characterize some extremal graphs when the upper bound is achieved.
It is easy to see from Lemma 2 that the number of 2-vertices is related to the number of internal triangles in G. Hence, the characterization of the bound on the number of 2-vertices in G is very important in our remaining proof. First we give some useful definitions and lemmas as preparation in what follows. Let m ≥ 3 be a positive integer. The sunflower graph SF 9 can be depicted as in the left graph of Figure 3. Figure 3: Sunflower graph SF 9 and generalized sunflower graphs GSF 8 and GSF 7 .
Based on the definition of sunflower graph proposed by Gallian in [9], we give the following definition of generalized sunflower graph. In this paper, we denote the generalized sunflower graph of order n by GSF n . According to the Definition 21, it is easy to see that n ∈ {2m−1, 2m}. Moreover, the generalized sunflower graph is a maximal outerplane graph with internal triangles if m ≥ 4. As an example, GSF 8 and GSF 7 can be depicted as in the middle and right of Figure 3, respectively.
Next, we prove that the upper bound on the total number of 2-vertices in G with internal triangle, and characterize its corresponding extremal graphs when the upper bound is achieved.
Lemma 22. Let G be a maximal outerplane graph with order n and k ≥ 1 internal triangles. Then k ≤ ⌊ n 2 ⌋ − 2 with equality holding if and only if G ∼ = GSF n . Proof. By Lemma 2, there must be k + 2 2-vertices in G since G has k internal triangles, and thus G has k + 2 marginal triangles. Applying Lemma 3, we have n − 2 − k ≥ k + 2, that is, k ≤ n−4 2 . It is obvious that k ≤ ⌊ n 2 ⌋ − 2 since k is a positive integer.
Let H be a Hamiltonian cycle of G and V (H) = {h 1 , h 2 , . . . , h n } with natural adjacencies. Assume that D 2 is the set of 2-vertices in G and V 2 is the set of two consecutive vertices of H. Note that |V 2 ∩ D 2 | ≤ 1 for any V 2 ⊆ V (H). By the structure of GSF n , it is easy to see that GSF n has ⌊ n 2 ⌋ − 2 internal triangles. Next, we may assume that G has k internal triangles with k = ⌊ n 2 ⌋ − 2. We will consider the following two cases according to the parity of n.
Case 1. n is even. In this case, k = n 2 − 2. Applying Lemma 2, it is observed that |D 2 | = k + 2 = n 2 . According to the property of D 2 , it is easy to see that d H (h, h ′ ) ≥ 2 for any two vertices h, h ′ ∈ D 2 . Then we have |V 2 ∩ D 2 | = 1 since |D 2 | = n 2 . So we may assume, without loss of generality, that Let u, v and w be any three consecutive vertices of H with clockwise order. It is obvious that d G (u, w) = 1 if v ∈ D 2 . Similarly, we can get d H (h 2i−1 , h 2i+1 ) = 1 and d H (h n−1 , h 1 ) = 1 for any i ∈ [k + 1]. It follows that there exits a cycle C := h 1 h 3 . . . h n−1 h 1 in G. Since G is a maximal outerplane graph, then the induced subgraph on D 2 is also a maximal outerplane graph and C is a Hamiltonian cycle of it. Furthermore, the induced subgraph on {h 2i , h 2i−1 , h 2i+1 } is a K 3 for any i ∈ [k + 1], and h n−1 , h n and h 1 three vertices of G also induces a K 3 . Therefore, we have G ∼ = GSF n . Case 2. n is odd. In this case, we can get k = n−5 2 . It follows from Lemma 3 that n − 2 − k = k + 3, which means that G has k + 3 marginal triangles. According to Lemma 2, there must be k + 2 marginal triangles containing 2-vertex in G. Then we will claim that there exactly one V 2 such that V 2 ∩ D 2 = ∅, where V 2 ⊆ V (H).
Theorem 19 gives a necessary and sufficient condition when the bounds on the gp-numbers for a striped maximal outerplane graph are achieved. Then we focus on the bounds on the gp-number for a maximal outerplane graph G with internal triangles. By the structure of GSF 7 (see the right graph of Figure 3), it is easy to verify that gp(GSF 7 ) = 4.
Proof. Let H be a Hamiltonian cycle of G with V (H) = {h 1 , h 2 , . . . , h n } and D 2 be the set of 2-vertices of G. By Lemma 2, we can get |D 2 | = k + 2. It is easy to verify that D 2 is a general position set of G, thus we have gp(G) ≥ k + 2. And in view of Theorem 16, the upper bound holds clearly and its corresponding extremal graphs are characterized. Assume that G ∼ = GSF n satisfies n ≥ 8 in the remaining proof. Then we will prove that gp(G) = k + 2 in what follows. Let R be a gp-set of G. By the above argument, we obtain |R| ≥ k + 2. Next, it suffices to prove that |R| ≤ k + 2. The proof is simple if 8 ≤ n ≤ 15, so we may let n ≥ 16 in the following.
|
2022-09-30T01:15:25.822Z
|
2022-09-29T00:00:00.000
|
{
"year": 2022,
"sha1": "001b026daae6e6c3e435715dfb3abcddba6a4b46",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Arxiv",
"pdf_hash": "001b026daae6e6c3e435715dfb3abcddba6a4b46",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
7134408
|
pes2o/s2orc
|
v3-fos-license
|
Oral Motor Intervention Improved the Oral Feeding in Preterm Infants
Supplemental Digital Content is available in the text
INTRODUCTION
T he early survival rate of preterm infants, in recent 30 years, has been greatly increased as the development of assisted reproductive technologies. 1-3 However, immature oral feeding ability has severely negatively impact on the normal development of preterm infants and even obviously increased the morbidity in this given population. 4 Coordination of sucking-swallowing-breathing (SSB) movements, which usually tend to mature until 32 to 34 gestational weeks, is an essential to develop the delicate oral feeding in infants. 5 The full-term infants can successfully complete the SSB activity, but preterm infants cannot.
Published studies revealed that nonnutritive sucking (NNS), which is closely associated with gestational age (GA), may improve the efficacy of oral feeding in preterm infants. [6][7][8][9] However, the preterm infants characterized by immature cardiorespiratory system, central venous system, and oral musculature will suffer from some threatening clinical outcomes which included bradycardia, apnea, and low oxygen saturation when changed feeding approach from tube to totally oral feeding, [10][11][12] and this condition was defined as oral feeding difficulty which is associated with the longer length of hospital stays (LOS), more medical costs, and serious psychological stress of parents after parturition, as well as even caused long-term oral feeding difficulties both related to bottle and breast feeding. 13,14 Some studies published previously suggested that early oral motor intervention (OMI), which consists of oral stimulation, oral support, and NNS, can better the effects of oral feeding in preterm infants and shorten the LOS. [15][16][17] However, the powers of conclusions were impaired due to some shortages such as small sample size existed in across studies. Although a systematic review was performed by Arvedson et al 18 to determine whether the OMI can improve the oral feeding ability of preterm infants, only studies published between 1960 and 2007 and in English, however, met its criteria. The Cochrane Collaboration recommended that a systematic review should be updated every other year, 19 and it is important that a plenty of randomized controlled trials (RCTs) have been developed since then. Hence, we aimed to further evaluate the potential of OMI for oral feeding in preterm infants by undertook this metaanalysis with trial sequential analysis (TSA).
MATERIALS AND METHODS
We performed our meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement 20 and Cochrane Handbook for Systematic Reviews of Interventions. 21 The prospective protocol of this topic has been registered on PROSPERO database and the register number of CRD42014014356 has been approved (available at: http://www.crd.york.ac.uk/prospero/). Ethical approval and informed consent were not required because all analyses were carried out based on these data extracted from published previously studies and no clinical prejudice was put on patients.
Studies Identification
An appropriate selection criterion is the key factor to guarantee the accurate studies identification. So we established this inclusion criteria according to the PICOS acronym: P (Population): all the patients diagnosed as preterm infants; I (Interventions) and C (comparisons): OMI compared with route interventions only; O (Outcomes): the transition time, LOS, feeding efficiency, intake of milk, and weight gain were assessed in our meta-analysis; S (Study Design): only RCTs with appropriate random sequence generation met the criteria. In addition, for duplicate data reported by the same author or 1 medical center, the article with high quality was included. Meanwhile, we included studies published in English and Chinese language.
Literature Screened and Data Extraction
Search was conducted, and data were extracted by 2 independent investigators (XT and L-JY). Each trial captured in the search stage was evaluated for author, publication year, the number of participants, allocation method, and patients' age, and interventions, period of treatments, eligibility criteria, baseline, and outcome measures of interest. Any divergences concerning the eligibility of a trial occurred in the any phases were resolved through discussion or consulting a third investigator (LZ).
Assessment of Risk of Bias
Two independent investigators (J-GZ and LM) assessed the methodological quality of trials included in our meta-analysis by using the Cochrane Risk of Bias Tool. 21 The procedure was performed based on the following 7 domains: randomization sequence generation, allocation concealment, blinding of participants and study personnel, blinding of outcome assessors, incomplete outcome data, selective reporting, and other biases. Based on the information extracted from each eligible trial, each domain was rated as ''high risk,'' ''unclear risk,'' or ''low risk.'' These domains will be classified as high risk unless appropriate methods were used; in contrast, corresponding domains will be graded as low risk when no obvious mistakes were detected; moreover, associated domains will be rated as unclear risk if lack of sufficient information to make a clear judgment on the risk of bias. Agreement on any domain was identified based on consensus or consulting a third investigator (Y-XO).
Statistical Analysis
All extracted data were entered into RevMan 5.3 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2013) for statistical analysis. Mean difference (MD) or standard mean differences (SMD) with 95% confidence interval (CI) for continuous outcomes were selected to estimate the pooled effect size. Heterogeneity in the included studies was evaluated using the x 2 , corresponding P value and I 2 statistic. If I 2 ! 50%, the eligible studies were considered to be heterogeneity and a random-effects model based on Mantel-Haenszel (MH) or inverse variance (IV) statistical approach was selected. In contrast, the studies were considered to be homogeneous, and a fixed-effects model based on MH or IV statistical approach
TSA
Repeated significance test of sparse and accumulated data has a risk to yield random errors which cause false positive or negative results. [23][24][25] For single primary trial, sequential analysis based on group sequential is similar to interim analysis that may increase the risk of type I errors. So the monitoring boundaries were developed and applied to determine if the trial should be ended early under the condition of a diminutive P value, which indicates statistical significant difference between study groups to minimize the type I error. 26 It is possible that sequential analysis, which can also be titled with TSA, can be adopted to analyze the pooled results of meta-analysis. 23 The quantification of the required information size (RIS) is a major factor to realize the TSA. We calculated the RIS adjusted for diversity because the heterogeneity adjustment with I 2 will underestimate the RIS value. 19 The TSA was performed at the level of an overall 5% risk of a type I error and 20% of the type II error (a statistical test power of 80%). 27 If the Z-curve across the monitoring boundary, then we can draw the conclusion of getting credible conclusion before surpassing the RIS line. If the Z-curve across the futility boundary, then we can come to the conclusion of this intervention have no effect for this outcome even though the RIS was not reached. The reliable conclusion can be drawn if the adjusted monitory boundary was surpassed and/or RIS was reached. Because effect measures selected in this meta-analysis fall into continuous data category, and the outcomes of transition time, LOS, feeding efficiency, intake of milk, and weight gain, we estimated the RIS based on the empirical data autogenerated from software according to the data input. 28 TSA software (version 0.9 beta) was available at http://www.ctu.dk/tsa/.
Assessment of Risk of Bias
A total of 855 participants were included into our metaanalysis. Basic characteristics of all eligible studies are shown in Table 1. Most of them have problems about blinding. Lowlevel literatures were resulted from incomplete outcome data. Although reported the dropouts, these studies have no intentionto-treat (ITT) analyses. The methodological quality assessments of included trials were shown in Figure 2.
Meta-Analysis on Transition Time
Ten of all trials involving 780 participants reported the transition time. Heterogeneity was identified across the included studies (P ¼ 0.01; I 2 ¼ 57%). Hence, a random-effects model, which indicates all the participants from included trials were sampled from the different population, was selected to summarize mean effect size because the resources caused heterogeneity were not detected among included trials. Metaanalysis suggested that OMI can effectively shorten the transition time, with statistical significant difference (MD, À4.03; 95% CI, À5.22 to À2.84) (Figure 3). Sensitivity analysis based on different pooled model was adopted to test the robustness of pooled result. Pooled result from fixed-effects model was the same with that of random-effects model, and well suggested that the summary effect size is robust (MD, À4.03; 95% CI, À5.22 to À2.84). We undertook a TSA at the level of a of 0.05, b of 0.2, and then an RIS of 279 was calculated. This pooled result was considered to have reliability resulted from Z-curve across conventional statistically significance test boundary, TSAadjusted boundary value, and the cumulative number of patients reached RIS of 279 ( Figure 4). So, OMI has the potential for transition time on preterm infants and worth clinical use.
Meta-Analysis on LOS
Six trials, which include 436 participants, were enrolled in our meta-analysis calculating the LOS. We identified homogeneity in the 6 studies assessed (P ¼ 0.27, I 2 ¼ 21%). Therefore, a fixed-effects model, which indicates all the participants from included trials were sampled from the same population, was performed to calculate mean effect size. Meta-analysis revealed that OMI effectively shortened the LOS, with statistical difference (MD, À3.64; 95% CI, À5.57 to À1.71) ( Figure 5). TSA was taken in the condition of a of 0.05, b of 0.2, and figured out RIS of 851. Although the accrued number of patients did not reach RIS of 851, the cumulative Z-curve cross conventional significance test boundary, RIS-adjusted boundary value, and the effect was prior established ( Figure 6). So, OMI has effect for the LOS on preterm infants and worth clinical use.
Meta-Analysis on Feeding Efficiency
Three trials, which included332 preterm infants, were enrolled in the meta-analysis identifying the feeding efficiency.
There was homogeneity about the 3 studies (P ¼ 0.78, I 2 ¼ 0%). Therefore, a fixed-effects model of analysis was used. Meta- FIGURE 3. Meta-analysis on transition time: OMI effectively reduced the time needed from bottle feeding to total oral feeding compared to route NICU care, random-effect model. OMI ¼ oral motor intervention. analysis result showed OMI can greatly improve the feeding efficiency, with statistical significant difference (MD, 0.81; 95% CI, 0.36-1.27) (Figure 7). TSA was performed in the level of a of 0.05, b of 0.2, and demonstrated RIS of 430. Even though the cumulated number of patients did not reach the value of RIS, the cumulative Z-curve cross conventional statistically significant boundary, TSA-adjusted boundary value and confirmed the result of reliability early (Figure 8). So, OMI has the potential for feeding efficiency on premature infants and worth clinical use.
Meta-Analysis on Intake of Milk
Three trials including 332 premature infants reported the intake of milk. Homogeneity was detected in the incorporated studies (P ¼ 0.25, I 2 ¼ 27%), and then a fixed-effects model of analysis was performed to calculate mean effect size. OMI can validly enhance the intake of milk, with statistical difference (MD, 0.14; 95% CI, 0.06-0.21) (Figure 9). TSA was performed, and RIS of 430 was counted in the condition of a of 0.05, b of 0.2. The pooled result manifested reliability due to Zcurve across conventional statistically significant boundary, TSA-adjusted boundary although the RIS was not accrued ( Figure 10). So, OMI has efficiency for intake of milk on premature infants and worth clinical use.
Meta-Analysis on Weight Gain
Three of all eligible trials, involving 318 patients, were enrolled in the meta-analysis reporting the weight gain. Heterogeneity was checked in eligible studies (P ¼ 0.00, I 2 ¼ 88%). We chose a random-effects model to summarize mean effect size. Meta-analysis result revealed that OMI cannot validly increase the weight gain (MD, À17.54; 95% CI, À151.34-116.26) (Figure 11). RIS and TSA adjusted boundary value cannot be calculated due to limited information size. So whether OMI was effective for weight gain on preterm infants, it still needs more researches to establish.
Publication Bias
We performed a funnel plot to examine the publication bias in all of the included studies. The outcome from the funnel plot analysis is shown in Figure 12, which shows symmetry, thereby indicating that no publication bias possibly exists in the included trials.
DISCUSSION
Advances in reproductive technologies brought out the significant improvements for survival rate of preterm infants in recent years. 39 The underdevelopment of cardiopulmonary system, central nervous system, and oral muscle tissue is responsible for lack of the ability of coordination of SSB for preterm infants, which often results in oral feeding difficulties and oral feeding ineffectiveness, [40][41][42][43][44] prolonged LOS, and increased burden of family and eventually the whole society. 45 Achievement of oral feeding is a primary criterion for discharge for healthy preterm infants. 46 Hence, successfully and effectively transited feeding approach from tube to oral feeding is the important target for health care staffs.
To generate reasonable and reliable pooled results, we selected the Cochrane risk of bias tool to critically assess the methodological quality. As the domain of incomplete outcome data, 4 included studies were judged with high risk of bias. Although reported the dropouts, intention-to-treat (ITT) was not undertook. So, we cannot judge whether the dropouts may impair the pooled results. In blinding of outcome assessment, FIGURE 6. TSA on pooled result of length of hospital stays: although cumulative sample size less than RIS, the Z-curve across the adjusted monitor boundary. RIS ¼ required information size, TSA ¼ trial sequential analysis. FIGURE 7. Meta-analysis on feeding efficiency: the feeding efficiency of preterm infants has been slightly improved after received the OMI approach, fixed-effect model. OMI ¼ oral motor intervention. one research was evaluated high risk of bias, which may result from inappropriate design for blinding, and 7 were unclear risk of bias, obvious information about blinding of outcome assessment cannot be identified in these researches, and may lead to evaluation bias. For blinding of participants and personnel, 7 studies were termed as unclear risk of bias, and specific explanations for this domain were not found. Impertinent blinding may lead to results away from true value and produce measurement bias. Implementing blinding for participants or not will not negatively or positively affect the process of whole study due to the target population included in our study is premature infant. Owing to all outcomes were objective, it cannot influence the outcomes in a large extent for whether implement blinding for personnel. For the domain of allocation blinding, 6 researches were judged unclear risk of bias. Insufficient allocation concealment may cause overestimating effect of intervention. Hence, we should evaluate the reliability of pooled results with prudence. In order to draw a more reliable conclusion, we hope that researchers could emphasize on blinding, allocation concealment, and complete outcome data in further studies. Among 11 studies included in this research, only 1 study was eligible for all domains of quality of methodology. Low-quality researches have greater bias in quality control and will affect the results of this study to some extent.
The meta-analysis results revealed that OMI can effectively improve the condition of transition time, LOS, feeding efficiency, and intake of milk related to route care. But no difference was identified in weight gain between both groups. We summarized the evidence to promote clinical use and further research on this topic in Table 2. The success exploration of OMI and its operation process reflect the limitations of traditional care for preterm infants. We hope that clinical staffs can improve and update their old care manner and make use of the direction of evidence-based nursing theory. Published evidences suggested that early OMI has effects on oral feeding in preterm infants and can shorten the LOS. However, the conclusion is still controversial with some drawbacks, such as small sample size, which can lead to low power regarding the effects of OMI for premature infants. Lack of power may result to get false-negative results, whereas this work combined with TSA to test whether this pooled results were robust and have credibility. And owing to time goes by, techniques of this intervention also improve; so it is needed to explore whether it is effective after about decade years. In addition, this systematic review included studies published between 1960 and 2007, whereas only English literature is eligible for its inclusion criteria 11 and the outcomes may be impaired by selection bias. So it is imperative to do this meta-analysis.
Our meta-analysis has only searched the PubMed, the Web of Science, EMBASE, the Cochrane Library and CNKI, but not SpringerLink, ScienceDirect, Chinese Biomedical Literature Database, and other relevant electronic database and unpublished sources, so there is a risk of incomplete retrieval. In addition, this study included only literature published in English FIGURE 9. Meta-analysis on intake of milk: OMI approach mildly increased the intake of milk of preterm infants compared to route NICU care regime, fixed-effect model. OMI ¼ oral motor intervention. . Meta-analysis on weight gain: no significant difference was detected between OMI and route NICU care regime in terms of weight gain, random-effect model. OMI ¼ oral motor intervention. and Chinese language, and the language restriction may result in selection bias, which affects the credibility of the pooled results of our meta-analysis. OMI is beneficial for oral feeding in preterm infants, but it still has some questions for clinical promotion. For example, the positive function of NNS has formed a consensus, but the application of the time and intensity of NNS not yet formed unified regulations. OMI for preterm infants should be more standardized and systematic. So, more studies need to do to further explore a standard operational approach. In recent years, due to a different mechanism of NNS and oral stimulation, foreign scholars began to explore the influence of the combined use of oral feeding and NNS in premature infants, as well as to determine whether the combination intervention regime is superior to one of all alternatives alone. Some of outcomes for included results were significant. However, they cannot use the pooled analysis for lacking of enough information. But these outcomes have significant value for assessing sucking ability and worth to use in future studies. So, we hope researchers do more studies to provide more standardized, scientific, rationalized approach for clinical use. Although the included studies have differences in the definition of GA for preterm infants, but the sensitivity analyses showed heterogeneity and have little effect on the pooled results. This study uses TSA to calculate the RIS of a of 0.05 and b of 0.2, and it showed credibility for these pooled results.
CONCLUSION
In conclusion, OMI can effectively improve the condition of transition time, LOS, feeding efficiency, and intake of milk, so it is worthy to be used widely in hospitals to improve the clinical outcomes of preterm infants. While RCTs with largescale and high-quality based on RIS are warranted to further investigate the effectiveness of OMI for weight gain and may explore whether it has the potential for other variable on preterm infants such as later growth and development. Linking Evidence to Clinical Use or Further Research OMI including oral stimulation and/or nonnutritive sucking and/or oral support. OMI lead to shorter transition time (ie, from bottle to total oral feeding) for preterm infants. OMI have the efficiency to shorten the length of hospital stays in premature infants. OMI can better the feeding efficiency for preterm infants. OMI have the potential to increase the intake of milk for preterm infants. We cannot get credible conclusion about whether OMI could improve the weight gain for premature infants, and thus more randomized controlled trials with large-scale and high-quality based on RIS are warranted to further investigate the effectiveness of OMI for this problem and may explore whether it has the potential for other variable on preterm infants such as later growth and development, which longer duration need time to observe.
OMI ¼ oral motor intervention, RIS ¼ required information size.
|
2018-04-03T02:54:07.836Z
|
2015-08-01T00:00:00.000
|
{
"year": 2015,
"sha1": "97ebb8d139ad37ca68349506e61991c4f233ffbe",
"oa_license": "CCBYNCSA",
"oa_url": "https://doi.org/10.1097/md.0000000000001310",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "97ebb8d139ad37ca68349506e61991c4f233ffbe",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
249662105
|
pes2o/s2orc
|
v3-fos-license
|
Exposure to gambling promotions and gambling behaviours in Australian secondary school students
Highlights • Adolescent gambling has been associated with a range of harms.• Young people are increasingly exposed to media promotion and advertising of gambling.• No studies have examined the relative impact of different types of advertisements.• We found an association between online gambling ad exposure and gambling behaviours.
Background
Despite legal restrictions on underage gambling, across jurisdictions most adolescents report having gambled at some point during their lifetime (Calado et al., 2017). Review evidence from various countries suggests that between 40% and 80% of youth have gambled in the past year, with 0.2-12.3 % of youth experiencing gambling-related problems (Calado et al., 2017). Adolescent gambling has been associated with a range of harms, including missing or dropping out of school; family disruptions; and substance use (Derevensky and Gupta, 2004;Fisher, 1999;Gupta and Derevensky, 1998;Huang et al., 2007;Yeoman and Griffiths, 1996).
Along with the well-acknowledged influence of family, social and cultural norms on adolescent's gambling, (Delfabbro and Thrupp, 2003) other factors within a young person's environment may also influence gambling behaviours (Messerlian et al., 2005;Shaffer, 2003). One key environmental factor is media promotion and advertising of gambling (Derevensky et al., 2010). Young people are increasingly exposed to messages from a broad range of media which endorse, promote, and glamorise gambling (Parrado-González and León-Jariego, 2020). Gambling advertising takes many forms beyond traditional modalities of television, radio, and print ads (Friend and Ladd, 2009). Gambling industries sponsor professional athletes, sports teams, celebrities, and popular events such as sporting and racing events (Friend and Ladd, 2009). In addition, smartphones, apps and social media have vastly expanded the gambling industry's marketing possibilities and reach (Victorian Responsible Gambling Foundation, 2021). Expenditure on gambling advertising also appears to be increasing. In Australia for example, expenditure on gambling advertising more than tripled between 2011 and 2020 to over $270 million (excluding social media, sponsorships and in-program content) (Victorian Responsible Gambling Foundation, 2021). Advertising expenditure increases are seen internationally including in the UK, Sweden, Canada and Spain (Parrado-González and León-Jariego, 2020; Torrance et al., 2021).
The powerful impact of advertising on children and adolescents' health behaviours has been examined in domains including alcohol, tobacco, and junk food consumption (Russell et al., 2019;Weitzman and Lee, 2020). However, there has been surprisingly little research examining the potential impact of gambling advertisements on young people's gambling behaviour, particularly in the last 10 years when such advertising has proliferated (Labrador et al., 2021). The existing research suggests that adolescent exposure to advertising is associated with an increase in the likelihood of engaging in gambling activities (Abdi et al., 2015;Freund, Noble, Hill, White, Evans, et al., 2022;Hayer et al., 2018;Kristiansen & Severin-Nielsen, 2021;Parrado-González and León-Jariego, 2020). For example, in a 2018 sample of 1,174 Spanish 12 to 20 year-olds, overall exposure to gambling advertising across different media, ranging from low to high exposure, was found to be associated with gambling frequency and problem gambling (Parrado-González and León-Jariego, 2020). Similarly, in a sample of 6377 Australian adolescents aged 12-17 years of age, exposure to one additional type of gambling advertisement was associated with a 6% increase in the odds of gambling in the last month and a 10% increase in the odds of being classified as a problem or at risk gambler (Freund, Noble, Hill, White, Evans, et al., 2022).
Few studies have examined how adolescents are exposed to gambling advertisements (e.g. through television, social media, or at sporting events) (Djohari et al., 2019;Labrador et al., 2021). A 2019 UK study of 99 young people (8-16 years) attending community events such as festivals and football tournaments, found the most often recalled gambling promotion was on television (79%), technology/screens (49%), in association with sports teams (43%), billboards (38%), sports stadiums (36%) and social media (35%) (Djohari et al., 2019). The previously mentioned Australian study found adolescents saw on average four different types of gambling advertisements a month (e.g. ads on television, online, billboards etc.) (Freund, Noble, Hill, White, Evans, et al., 2022). To our knowledge, no study has examined the relative association of exposure to different types of gambling advertisement and adolescent gambling behaviour. This type of information would provide important intelligence to policy and decision makers. Given the expansion of gambling advertising and promotion across multiple platforms, the present study aimed to explore the relative associations between exposure to different types of gambling advertisement and the prevalence of gambling in the last month, problem and at risk gambling, and type of gambling activities among a sample of Australian adolescents.
Methods
Details on the study methodology have been published elsewhere (Freund, Noble, Hill, White, Evans, et al., 2022). A brief overview is presented below.
Study design
Gambling questions were included in the cross-sectional triennial Australian Secondary Students' Alcohol and Drug (ASSAD) Survey for the states of Victoria and Queensland in 2017. A random sample of schools, stratified by education sector, was developed for each participating state. The Australian education sector comprises Government and non-Government (Catholic and Independent) schools. Independent schools include those affiliated with non-Catholic religions (Independent Schools Australia, 2022). Ethics approval was granted by the relevant State and institutional Human Research Ethics Committees (HRECs), including the University of Newcastle HREC (Ref: H-2017-0102).
Sample and procedure
Within participating schools, classes of students in Years 7 to 12 were randomly selected to complete the ASSAD survey. Researchers attended the school to administer the pencil-and-paper questionnaire to selected classes. Further details regarding the ASSAD sample selection process and data collection/analysis procedures (including use of weights to account for over-sampling) have been published elsewhere (Guerin & White, 2018).
Measures
Gambling items were developed through an iterative process including an extensive literature review, advice from experts in adolescent youth gambling and smoking research, and pilot testing of items with a group of adolescents (n = 10) (Freund, Noble, Hill, White, Evans, et al., 2022). Prior to answering the gambling-related questions, students were given the following definition of gambling: 'Gambling is when you pay in your own money knowing that you could lose all of it or, possibly, win back even more than you paid in. There are lots of ways to gamble, for example on the results of races, sports, card games, lotteries, raffles, on machines like "pokies", tipping competitions and sweepstakes.'.
Ever gambled and gambled in the last 30 days
Students were asked 'Have you ever bet any money on any form of gambling?' (yes/no). Students who answered in the affirmative were asked if they had gambled in the past 30 days (yes/no).
Other people's gambling
Students were asked to select any people they knew who had gambled in the last 30 days including: mother/caregiver, father/caregiver, brother or sister, other relative, one of your best friends, someone else you know. A variable called "number of known gamblers" was derived by summing the number of different people a student knew who had gambled in the last 30 days (0, 1, 2, 3, 4 + ).
Types of gambling activities
Students indicated, for each of 13 nominated gambling activities, whether they had gambled on that activity in the last month. Types of gambling activities were categorised based on the perceived level of risk (hard versus soft gambling activities). Hard gambling activities have been defined as those with a potential for a high payout ratio and/or rapid event frequency, (Griffiths, 1999) and included gambling on casino, card or sports games, poker machines, horse or dog racing, personal skill games and two up 1 . Soft gambling activities included tipping competitions, sweeps 2 , bingo, lottery tickets, instant scratch cards, raffles tickets and other types of gambling.
Problem gambling
Students who had ever gambled were screened for problem gambling using the 12 item Diagnostic Statistical Manual IV (Multiple Response format) adapted for Juveniles (DSM-IV-[MR]-J). This tool is frequently used by youth gambling researchers, (Stinchfield, 2011) and has demonstrated reliability and validity (Fisher, 2000;O'Neil et al., 2003;Rossen, 2001). In the current study, response options were revised to a dichotomous scale (yes/no). This is consistent with other Australian studies, (Delfabbro et al., 2005;Delfabbro and Thrupp, 2003) and research suggesting the 'yes/no' response scale is more easily answered than frequency response options for this age range (Purdie et al., 2011). Consistent with Fisher, 1999, respondents were classified as follows: (a) non-problem gamblers (did not endorse any of the diagnostic criteria); (b) at risk gamblers (responded 'yes' to between one and three of the diagnostic criteria); and d) problem gamblers (responded 'yes' to four or more of the diagnostic criteria). Students who had never gambled were included in the non-problem gamblers category.
Exposure to gambling promotion
Exposure to advertising was measured through an adaptation of Hing et al.'s 2014 exposure to sports advertising scale, modified to include non-sports gambling promotions such as promotion on social media. Students were asked to indicate whether they had been aware of a range of advertisements or promotions for gambling in the past 30 days. See Table 1 for the complete list of gambling advertisements and promotions.
Student characteristics
Students self-reported their: postcode; age; gender; main language spoken at home; money to spend on self per week ($AUD); selfconsidered school achievement; and attendance at school on previous school day. Student's home postcode was used to classify their residential location according to the Accessibility and Remoteness Index of Australia (ARIA + ), as either major city or other (inner regional, outer regional, remote, very remote) (Australian Bureau of Statistics, 2018). Level of socioeconomic disadvantage was also based on student postcode using the Socio-Economic Indexes for Areas (SEIFA) Index of Relative Socio-economic Disadvantage (IRSD) decile classifications (Australian Bureau of Statistics, 2017).
Analysis
All statistical analyses were programmed using SAS v9.4 (SAS Institute, Cary, North Carolina, USA). An alpha level of 0.05 was specified for all tests and confidence intervals. Student exposure to gambling promotions in the last 30 days is presented descriptively (as raw N and weighted proportions), by age and gender.
Principal Component Analysis (PCA) was performed on the eleven gambling advertisement exposures using prior communality estimates of one. Components were extracted using the principal axis method and then rotated orthogonally. Gambling advertisement exposures were considered to load on a given component if the absolute loading was greater than 0.40 for that component and less than 0.40 for the other components. Examination of the association of exposure to gambling promotions with student gambling in the last 30 days, type of gambling activity (hard versus soft), and problem or at risk gambling, was undertaken using logistic mixed-models. Gambling advertisement exposure was included in univariate and multivariable logistic mixed-models using component outcomes of the PCA. The multivariable regression analyses included fixed effects for gender, age, money to spend on self, number of known gamblers, socioeconomic disadvantage, perceived school achievement, attended school yesterday, and a random effect for school ID. Available money per week was categorised as "None", "$1-$40", "$41-$80" and "$81+" for analysis. Socio-economic disadvantage was categorised as high (SEIFA IRSD deciles 1-6) or low (SEIFA IRSD deciles 7-10). "Age", "available money per week", "number of known gamblers", "perceived school achievement" were found to be non-linear and were assessed categorically and presented with an overall Wald type-3p-value.
Results
A total of 93 schools participated in the ASSAD survey in 2017 (57 schools from Victoria and 36 schools from Queensland). Details of the school sample have been published elsewhere (Freund, Noble, Hill, White, Evans, et al., 2022). The sample was made up of Government schools (68%), Catholic schools (15%) and Independent schools (17%).
Over 7,000 students took part in the survey. Students who were missing responses to core gambling module questions (n = 707) and those who did not answer the first gambling question (have you ever gambled?; n = 112) were removed from the dataset, resulting in a final sample size of 6377 students for analysis. Students who responded 'yes' to the first gambling question (have you ever gambled?) but were missing a response to the gambling in the last month question were assumed not to have gambled in the last month (n = 272).
The demographics of participants have also been reported in detail elsewhere (Freund, Noble, Hill, White, Evans, et al., 2022). Over half of the sample were female (56%), and the largest age group was those aged 16 years (23%). The majority of the sample were from major cities (65%) and inner regional areas (22%). Just over half of the students (54%) were classified as being disadvantaged, based on SEIFA IRSD deciles (deciles 1-6). Half of the students reported having between $1-$40 available to spend on themselves per week.
Gambling behaviours and types of gambling activities
The prevalence of gambling in the last month and at risk and problem gambling, and the types of gambling activities for this student sample has been described previously (Freund, Noble, Hill, White, Evans, et al., 2022). Briefly, 6% of students reported gambling in the previous month and 10% of these were classified as at risk or problem gamblers (Freund, Noble, Hill, White, Leigh, et al., 2022). The most common type of gambling in the last month was betting on horse or dog races. Approximately 4% of all students reported gambling on any hard modality (e.g. card games, casino games, sports betting and poker machines), or any soft modality (e.g. lottery tickets, raffles, sweeps etc.) activity in the last month (Freund, Noble, Hill, White, Leigh, et al., 2022).
Exposure to gambling promotions
Results of exposure to gambling promotions in the last 30 days by age and gender are shown in Table 1. Across the whole sample, 81% (n = 5165) of students reported being exposed to any form of gambling promotion or advertisement in the last 30 days (data not shown). Of these 5165 students who reported being exposed to any type of advertisements, advertising exposure was most commonly via TV (85%), followed by social media (46%) and then at sporting events (40%).
PCA of exposure to gambling promotions
Results of the PCA are shown in Appendix 1. PCA identified three 1 Two Up is an Australian/New Zealand gambling game in which two coins are tossed in the air and bets are laid as to whether both will fall heads or tails uppermost. 2 Sweeps involve participants paying to randomly receive the name of a competitor (e.g. horse, team etc.), and winning money if their competitor wins.
components which accounted for 55% of the total variance. Five gambling promotions loaded onto the first component: ads at shops or newsagencies, pubs or clubs, on websites, on social media, and pop-ups on websites. The latter two promotions (websites and social media) loaded more heavily on component 1 than the others. Three advertising exposures: ads at sporting events, live studio crosses, and celebrities promoting gambling; loaded on the second component, and three (ads on TV, radio and billboards) on the third component. Due to a specific interest in exposure to online advertisements, the first component was split into two. Based on the relative component loadings, gambling promotion exposures were grouped into the following four binary exposures (exposed/not exposed): Online ads (including websites, pop-ups on websites, social media); Retail, pubs & clubs ads (shops, newsagencies, pubs/clubs); Sports/celebrity ads (sporting events, live studio crosses, celebrity promotions); and Traditional media ads (TV, radio, billboards).
Association of exposure to gambling promotions with student gambling behaviours
Univariate associations between exposure to gambling promotions and student gambling behaviours are shown in Table 2. Results of the adjusted logistic regressions are presented in Table 3. All students with data available were included in the regression analyses (including those who had never gambled or had not gambled in the last month).
Gambled in the last month
In the unadjusted model (Table 2), exposure to Online ads, Retail, pubs & clubs ads, and Sports/celebrity ads were all significantly associated with students gambling in the last month. However, after adjusting for the other characteristics (gender, age, money to spend on self, number of known gamblers, socioeconomic disadvantage, perceived school achievement, attended school yesterday, and school ID), only Online gambling exposure remained significant. Students were more likely to have gambled in the last month if they reported being aware of Online ads in the last month (OR 1.37; 95% CI: 1.01, 1.85); Table 3). As has been previously reported, gender and the number of known gamblers remained significant in the adjusted regression model (Freund, Noble, Hill, White, Leigh, et al., 2022).
Participated in any hard gambling activity in the last 30 days
In the unadjusted model (Table 2), exposure to Online ads, Retail, pubs & clubs ads, and Sports/celebrity ads were all significantly associated with students participation in a hard gambling activity in the last month. After adjusting for the other characteristics (gender, age, money to spend on self, number of known gamblers, socioeconomic disadvantage, perceived school achievement, attended school yesterday, and school ID), there were no significant associations between exposure to gambling promotions in the last month and student engagement in hard gambling activities in the last month (Table 3). As previously reported, gender and the number of known gamblers remained significant in the regression model (Freund, Noble, Hill, White, Leigh, et al., 2022).
At risk or problem gambler
In the unadjusted model (Table 2), exposure to Online ads, Retail, pubs & clubs ads, and Sports/celebrity ads were all significantly associated with students at risk or problem gambling classification. After adjusting for the other characteristics (gender, age, money to spend on self, number of known gamblers, socioeconomic disadvantage, perceived school achievement, attended school yesterday, and school ID), only Online gambling remained significant. Students were more likely to be classified as at risk or problem gamblers if they reported being aware of online gambling ads in the last month (OR 1.84; 95% CI: 1.41, 2.38; Table 3). As previously reported, gender, money to spend on self, and the number of known gamblers remained significant in the regression model (Freund, Noble, Hill, White, Evans, et al., 2022).
Discussion
In this large two-state sample of Australian secondary school students, exposure to gambling promotions was common, with the majority of students being aware of any type of gambling advertising in the last month (81%). Of those students, 85% reported being aware of advertisements for gambling on TV in the last month, 46% of advertisements for gambling on social media, and 40% of ads at sporting events. These results are in line with previous reviews which indicate the majority of adolescents and young people are exposed to gambling advertising on TV, the internet and at sports events (Labrador et al., 2021).
With the exception of Traditional media advertising (TV, radio and Table 1 Exposure to gambling promotion types in the last 30 days among students who reported exposure to any type of advertisements (n = 5165), by age and gender.
Gambling promotion type Gender n (%)* Age n (%)* All n (%)* billboards), exposure to other types of gambling advertising (Online; Retail, pubs & clubs; and Sports/celebrity ads) were associated with each of the gambling outcomes in the unadjusted analyses. However, after adjusting for sociodemographic variables and the range of gambling exposures, only exposure to online gambling ads (websites, pop-ups on websites, and social media) in the last month was significantly associated with adolescent gambling in the last month, and being classified as an at risk or problem gambler. Students who reported being aware of online gambling ads were 37% more likely to have gambled in the last month, and 84% more likely to be classified as an at risk or problem gambler, than those who did not report exposure to online gambling ads. Online gambling exposure was not associated with engaging in a hard gambling activity in the last month. Results suggest that while exposure to most forms of gambling are associated with adolescent gambling behaviours, the role of online advertising is particularly important. Interestingly, exposure to ads on traditional media did not appear to be linked to gambling behaviours, perhaps suggesting a decline in the influence of such avenues of advertising for younger Australians.
Only limited previous research has examined whether gambling advertising exposure is associated with youth gambling behaviours, (Clemens et al., 2017) with five studies having been conducted across Germany, (Clemens et al., 2017;Hayer et al., 2018) Canada, (Derevensky et al., 2010) Spain, (Parrado-González andLeón-Jariego, 2020) and Israel (Gavriel Fried et al., 2010). Similar to findings for adults, (Bouguettaya et al., 2020) exposure to gambling promotions was significantly associated with youth gambling behaviours including gambling frequency in the last year, (Clemens et al., 2017;Hayer et al., 2018;Parrado-González and León-Jariego, 2020) month, (Clemens et al., 2017) or week, (Clemens et al., 2017;Gavriel Fried et al., 2010) Univariate association between exposure to gambling promotion and gambled last 30 days, engagement in hard gambling activities in the last 30 days, and at risk/ problem gambling (unadjusted crude results). *Wald type-3p-value. The number of students included in the unadjusted analyses ranged from n = 4993 to 6377. and with being classified as a problem gambler (Derevensky et al., 2010;Gavriel Fried et al., 2010). However, all five studies used an overall measure of exposure to gambling advertising (i.e. assessed exposure across a range of mediums, including the internet), rather than investigating different types of gambling advertising exposure. The current study provides the first evidence of a specific association between online gambling advertising exposure and youth gambling behaviours. Current findings highlight the need to further explore the impacts of online gambling advertising on young people's gambling behaviours, particularly in the Australian context. The link between online advertising and youth gambling is concerning given the rapid expansion of gambling advertising into the digital sphere via online and social media marketing, (Torrance et al., 2021) and that young people tend to have the highest use of the internet and social media (Gainsbury et al., 2016;Kristiansen & Severin-Nielsen, 2021). Young people can access gambling via betting or related apps, (King et al., 2020) where online advertising may be targeted towards the types of gambling activities that the young person engages in. In contrast to traditional media, social media outlets remain largely unregulated (Gainsbury et al., 2016;O'Loughlin & Blaszczynski, 2018). A 2016 review found that most gambling operators did not incorporate responsible gambling messaging in their use of social media, despite this being a requirement of most advertising codes of conduct (Gainsbury et al., 2016). In addition, advertising via social media may be particularly appealing to young people, because postings might not be recognised as advertising, and content can be shared and promoted by peers, including to underage Table 3 Multivariate association between exposure to gambling promotion and gambled last 30 days, engagement in hard gambling activities in the last 30 days, and at risk/ problem gambling (adjusted results users (Kristiansen & Severin-Nielsen, 2021;O'Loughlin & Blaszczynski, 2018). Restrictions on gambling advertising have been identified as a potentially cost-effective measure for reducing harms associated with gambling (Bouguettaya et al., 2020;Parrado-González and León-Jariego, 2020). The current study highlights the need to continue to regulate all forms of gambling advertising, as well as expand and enforce such restrictions on online gambling advertising (Kristiansen & Severin-Nielsen, 2021).
Limitations
All variables collected in the questionnaire were self-reported, and are subject to potential recall error and social desirability biases. In particular, we asked students to indicate whether they had been aware of a range of advertisements or promotions for gambling in the last month. As such, we measured recall of gambling promotions as a proxy measure of exposure. It is possible that participation in gambling affects recall of exposure to advertisements, rather than the other way around (Gavriel Fried et al., 2010;Newall et al., 2019). In addition, we did not assess the frequency or level of exposure to types of gambling promotions, with students classified only as either exposed or not exposed to each advertising category. There may have been some overlap between the exposure categories -for example, celebrity promotions may have been seen on the internet or TV. It is possible that the lack of an association between exposure to non-online forms of gambling promotions (e.g. TV, radio, during sports events, celebrities promoting gambling etc.) and gambling behaviour was as a result of not assessing the degree of exposure to such promotions among students; and/or due to some overlap between exposure categories.
Conclusion
This large quantitative and representative study is the first to explore the relative role of exposure to gambling advertising via different platforms on adolescent gambling behaviours. We found evidence of an association between youth exposure to gambling advertising and engagement in gambling activities and at risk or problem gambling, especially for exposure to online gambling promotions. Our study suggests the important potential role of social media advertising in influencing the gambling behaviours of Australian youth. More Australian and international research should be conducted to confirm and further the findings, particularly studies using longitudinal designs. Current results point to the need to implement controls on gambling advertising and particularly online gambling promotions, in order to "problematise what advertising normalises" (Parrado-González and León-Jariego, 2020).
Role of Funding Sources
This work was supported by funding from the Victorian Responsible Gambling Foundation and infrastructure funding from the Hunter Medical Research Institute. Dr Megan Freund is supported by a National Health and Medical Research Council Translating Research into Practice Fellowship. ASSAD data collection was funded by state and territory Cancer Councils and health departments, and the Australian Government Department of Health. Cancer Council Victoria managed the national survey. The State Governments of both Victoria and Queensland, Cancer Council Victoria and Cancer Council Queensland funded the conduct of the ASSAD survey in each of their States. The funding sources had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.
Contributors
MF, DH, VW, MS, RSF designed the study and wrote the protocol. NN and MF conducted literature searches and provided summaries of previous research studies. VW, LL and DL conducted the statistical analysis. NN and MF wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
|
2022-06-15T15:06:08.398Z
|
2022-06-01T00:00:00.000
|
{
"year": 2022,
"sha1": "d890a7d34ba4d8c08b6653ffcd7d558043ffe103",
"oa_license": "CCBYNCND",
"oa_url": "https://doi.org/10.1016/j.abrep.2022.100439",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "70205ae1d8dc9af85df6c7bc9ca8a28fd7b4a6e2",
"s2fieldsofstudy": [
"Economics",
"Law"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
237342341
|
pes2o/s2orc
|
v3-fos-license
|
Emerging Applications of Nanotechnology in Healthcare Systems: Grand Challenges and Perspectives
Healthcare, as a basic human right, has often become the focus of the development of innovative technologies. Technological progress has significantly contributed to the provision of high-quality, on-time, acceptable, and affordable healthcare. Advancements in nanoscience have led to the emergence of a new generation of nanostructures. Each of them has a unique set of properties that account for their astonishing applications. Since its inception, nanotechnology has continuously affected healthcare and has exerted a tremendous influence on its transformation, contributing to better outcomes. In the last two decades, the world has seen nanotechnology taking steps towards its omnipresence and the process has been accelerated by extensive research in various healthcare sectors. The inclusion of nanotechnology and its allied nanocarriers/nanosystems in medicine is known as nanomedicine, a field that has brought about numerous benefits in disease prevention, diagnosis, and treatment. Various nanosystems have been found to be better candidates for theranostic purposes, in contrast to conventional ones. This review paper will shed light on medically significant nanosystems, as well as their applications and limitations in areas such as gene therapy, targeted drug delivery, and in the treatment of cancer and various genetic diseases. Although nanotechnology holds immense potential, it is yet to be exploited. More efforts need to be directed to overcome these limitations and make full use of its potential in order to revolutionize the healthcare sector in near future.
Introduction
Nanobiotechnology, a recently coined term, emerged from the blending of molecular biology and nanotechnology. It is a branch of science which revolves around structures or functional materials at the nanoscale, which are produced by employing both physical and chemical methods [1]. In the last thirty years, the discipline of nanotechnology has been a crucial area of research, due to the unique chemical, electrical, optical, biological, and magnetic properties of nanomaterials [2]. Nanotechnology has managed to attract a lot of attention, because it is an established fact that when nanotechnology joins hands with biotechnology, they give birth to a platform which holds immense potential and importance with respect to diversity in applications [3]. Some of these applications include medical imaging, diagnostic kits, diagnostic assays, biological sensors, dentistry, sterilization of medical device surfaces, sunscreens, cosmetics, sports equipment, textiles, environmental cleanup, and gene inactivation [1,4,5]. The development of nanotechnology has provided They have a tunable surface and a large surface area to which drugs and genes can easily conjugate. The fluorescence produced by NDs makes them useful as imaging probes for diagnostic purposes [39,40]. All these properties of NDs are actually due to the combined characteristics of diamonds and NPs [40]. The structure of NDs consists of two major components-(1) the inner diamond core, with carbon atoms in the sp 3 configuration; and (2) the outer graphitic shell (carbon atoms in the sp 2 configuration), with functional groups on the terminal of dangling bonds [41]. Techniques used for the synthesis of NDs include the detonation of explosives, high temperature, high pressure, and the chemical vapor deposition method [22].
Quantum dots are synthetic nanostructures ranging in size between 1.5-10 nm. their semi-conductor nature allows them to transport electrons. When UV light passes through them, the electrons in the QDs are excited, and when these excited electrons move back to their ground state, they emit light. QDs emit light of different colors depending upon their size [42]. QDs made from heavy metals such as cadmium are very toxic and carcinogenic; therefore, they cannot be widely used in the health sector. However, graphene and carbon QDs are safe and stable and have wide scope in the health sector [43].
Nanofilms consist of polymeric sheets with a large surface area and a thickness of relatively few nanometers (10-100 nm) [44]. Multiple oppositely charged layers are assembled together to form multilayered yet ultra-thin biofilms. Layers are deposited one by one for deposition. Various methods are used for the deposition of individual layers, including fluidic assembly, electromagnetic deposition, spin coating, and emersion [45].
Liposomes are spherical vesicles made up of one or more lipid bilayers with an aqueous compartment in between them [42,43]. They are found in a variety of sizes, starting from as small as a few nanometers, and can be as large as several micrometers [44]. They are capable of entrapping various substances, including hydrophilic and lipophilic agents. Therefore, they are also considered to be the most efficient drug delivery system. Another reason for this is because their composition is very similar to the cellular membranes found in the body, which helps with drug delivery in vitro. Their large size also enables them to deliver a high quantity of drugs [45]. The major domains of healthcare in which nanotechnology-mediated nanosystems are playing their positive role are summarized in Figure 1.
Nanostructure Applications in Health Sector References
Nanoparticles Used as antimicrobials and antifungals; used as sensors, as catalysts, and for imaging in diagnostics [5,21] Carbon Nanotubes Used for delivering fibrinogen and bovine protein to cells; serve as vectors for gene delivery; and in the treatment of broken bones, osteoporosis, and breast cancer [26,27,[46][47][48][49] Dendrimers Used for diagnostic applications, for gene delivery, as anti-bacterial agents, as anticancer drugs, to improve vaccine formulations by acting as carriers of antigens, and in treating ocular diseases. [32] Nano-Diamonds Used for the treatment of bone disease by targeted drug delivery (bone regeneration); used in imaging and therapy, in the early detection of cancer, and in the treatment of brain and breast cancers
Role of Nanotechnology in Gene Therapy
Gene therapy is a procedure to replace a defective gene in the DNA (which is responsible for causing a disease) with a normal gene. The gene is usually inserted into the stem cells using a vector [64]. Stem cells have long life and a self-renewal ability; therefore, they are the most suitable targets for gene therapy [65]. The vector used should be highly specific and efficient in releasing the gene or genes of variable sizes. It should not be recognized as an antigen by the host immune system. The vector must have the ability to express the inserted gene throughout the life of that organism [66,67]. When the gene is correctly inserted into the cells, it inhibits and corrects the functions of the mutated gene and induces the normal functioning of cells [68,69]. Viral vectors have been used for years in gene therapy and are still being used. They can take over the host metabolic machinery for the synthesis of proteins that are coded by their DNA. Furthermore, their insertion in the host genome is very stable, and the transduced cells cause the long-term expression of the transgene. These are the properties that make them suitable for gene therapy [67,68]. Some common and efficient viral vectors include lentivirus, retroviruses, adenoviruses, etc. [67,[69][70][71][72][73][74][75][76]. However, there are many risks associated with the use of viral vectors. These include the generation of an immune response, inflammation, and the occurrence of off-target changes in the host body. If the virus triggers the immune response, it not only makes the therapy less efficient but when the same virus enters the body the second time (with the desired gene inserted into its DNA), a secondary immune response occurs, which would rapidly kill the virus, making it impossible to use the same virus for gene therapy [77][78][79][80][81]. Inflammation caused by viral vectors can sometimes be very dangerous, as reported in a recent study in which a leukemic patient died when given a high dosage of adenovirus for gene therapy [82]. Virus virulent genes are deleted prior to therapy, which also compromises the integration and infection ability of viral vectors. Insertional mutagenesis can be life-threatening too, because sometimes these viruses (mostly retroviruses) insert DNA into the tumor-suppressing gene or the oncogenes, activating them to cause tumors in the host body. The selection of appropriate viruses for different body cells is another difficulty in the field of gene therapy. Moreover, viruses can also go through genetic changes with the passage of time, which can lead to other complications in the body [83]. These are some major concerns relating to viral gene therapies, and therefore these methods are not encouraged, and the world is now moving towards the use of nanostructures for gene therapy.
Gene therapy using non-viral nanostructures is safe, as compared to therapy using viral vectors. They are also much less oncogenic and rarely trigger immune responses. Their preparation is much easier than that of viral vectors. There is no risk of virus recombination and no limit on the size of the gene to be loaded. NPs are one of the many nanostructures that are used for non-viral gene delivery. The presence of a positive charge, small size, and high surface-to-volume ratio enables them to penetrate deep into the membranes, thus making them ideal vectors for gene delivery [84][85][86]. The major nanosystems used in gene therapy are shown in Figure 2.
One of the ways in which gene therapy treats many diseases is through gene silencing. Various diseases, such as autoimmune disorders, cancers, and viral infections, can be treated by silencing the expression of genes [87]. RNA interference using small interfering RNA (siRNA) has been used for gene silencing. SiRNA is a 21-25 nucleotide long doublestranded RNA molecule. It forms a complex with RNA-induced silencing complex (RISC) in the cytoplasm and targets the directed complementary mRNA molecule, thus silencing its expression [88,89].
This technique can be very useful if the problem of their stable delivery into the cytosol (they become unstable in physiological fluids) and limited intracellular uptake are resolved [90]. This problem can be resolved by using some vector system. Viral vectors are very risky to use, as mentioned earlier. However, non-viral NPs have been used to overcome these limitations [91]. For example, one of the over-expressed proteins in cancer cells is the RhoA protein. Anti-RhoA siRNA was encapsulated in chitosan-coated polyisohexylcyanoacrylate (PIHCA) NPs. When these NPs were administered to mice infected with breast cancer, they showed 90% tumor inhibition with no toxic effects [92].
In rheumatoid arthritis, tumor necrosis factor-α (TNF-α) plays a role in the release of cytokines and thus causes chronic inflammation. A nanocomplex, thiolated glycol chitosan (TGC) polymer loaded with poly-siRNA, was targeted to TNF-α, which proved to be very efficient in curing rheumatoid arthritis. The inhibition in bone erosion and a reduction in inflammation was also observed in mice in that study [93]. These are just a few examples; there are several other studies in which nanostructure-based complexes have been effectively used to deliver siRNA, thus treating various diseases.
In another way, genetic materials (RNA, DNA, siRNA) can be encapsulated or conjugated with NPs for efficient gene delivery [84,[94][95][96]. The most efficient way to attach genes with NPs is through the formation of DNA-NP complexes. These complexes are formed by means of the electrostatic bonding between them. For this, the surface charge on the NPs is made positive, which then binds strongly with negatively charged nucleic acids. Liposomal and polymeric and many other nanostructures use this mechanism of gene transfer [85,[97][98][99][100]. The encapsulation of genetic material in NPs protects them from enzymatic digestion when they are targeted into the cells. It also protects them from phagocytosis by monocytes [94]. Due to the advantageous aspects of nano-based gene therapy, research is in process on large scale to develop new strategies for its implementation in the healthcare sector.
The Role of Nanotechnology in Targeted Drug Delivery
Nanovectors have great potential in target-specific drug delivery for the treatment of various diseases. Targeted drug delivery is important, especially if the solvents of hydrophobic drugs are toxic. If these solvents are released somewhere else other than the target cell, they may enter the blood stream or other body fluids and contaminate them. Nanostructures allow the continuous controlled release of drugs in desired amounts. Specific and localized drug delivery also reduces drug doses. The small size of NPs allows them to penetrate deep into the tumor cells, and thus they can be useful in improving cancer treatments [94].
The NPs used for drug delivery must contain some important components, including a particle core, an outer biocompatible protective layer and a linking molecule for increased bioactivity (it attaches the core of NPs to bioactive molecules because of the reactive compounds present at both of its ends). Nanovectors are modified before drug delivery and this modification includes coating with ligands such as peptides, folic acid, and antibodies. Ligands are attached to NPs so that they can bind specifically to targeted sites to enhance the specificity even more [16,95,[101][102][103][104][105][106]. It is essential to attach more than one ligand because if only one ligand is attached, there is a possibility that it may bind to receptors present in places other than on the targeted site. In addition, tumor cells are usually overexpress, i.e., they have more than one type of surface receptor [17].
Since nanovectors possess unique properties and various modifications can be performed during drug loading, scientists are now moving towards the implement of nanotechnology-based nanosystems for efficient targeted drug delivery with the aim of curing various serious diseases. Some examples of targeted drug delivery using nanovectors are discussed in the following sections.
Treating Cardiovascular Diseases through Nanosystems
Cardiovascular diseases cause millions of deaths around the world [18]. Various treatments have improved the survival rate of patients with heart diseases but none of them has achieved complete cardiac regeneration, especially for patients after cardiac infarction [107]. Stem cell therapy can be used for therapeutic angiogenesis [108]. Introducing anti-apoptotic and pro-angiogenic genes into the genetically engineered stem cells can prolong their rate of survival and increase their paracrine secretion [109,110]. Viral vectors cannot be used to deliver genes to stem cells as they cannot carry large gene volumes and have immunogenic effects. Bio-compatible NPs are efficient in transferring genes to stem cells. Various nanostructures can be used for delivering genes to stem cells. Liposomes are one of the best contenders for gene delivery as they can prevent the non-specific binding of genes and protect them from degradation [111,112]. Polymers show improved specificity for targets and higher efficiency [113]. In one study, chitosan alginate NPs were used to deliver growth factors to placental cells. The continuous release of growth factors improved the functioning of cardiac tissues at the site of myocardial infarction [114]. NPs also have the potential for tracking and monitoring stem cells. Superparamagnetic iron oxide nanosystems (SPIONs) are made to enter the cells by attaching to cell surfaces. These cells are then internalized by endocytosis [115]. Quantum dots can also be used for monitoring the living cells for a long time [116,117].
Hypertension is a disease that gives rise to many problems, including myocardial infarction, heart failure, stroke, increased blood pressure, and damage to many body organs, including the eyes, kidney, brain, etc. [118]. Many antihypertensive drugs have been used to treat this, but various problems are associated with the use of these drugs, including their short half-life, low bioavailability, poor solubility in water, unwanted side effects, and many more. Targeted drug delivery using nanostems has been effectively performed in order to solve these problems [119]. Nanocarriers that have been used so far for treating hypertension include lipid carrier NPs, solid lipid NPs, polymeric NPs, liposomes, and nanoemulsions [120]. These are just a few examples, but nanotechnology has very promising applications in treating many other cardiovascular diseases through non-viral stem cell-based therapies. Further studies on the effects of nanovectors in the cardiovascular system of a living model need to be performed before they can be safely used in humans.
Nanotechnology in the Treatment of Ocular Diseases
The efficient delivery of drugs in the eye is an enormous challenge because of the presence of complex barriers and elimination mechanisms in the eye. The various barriers present include the tear film, the ocular surface epithelium, and the internal blood-aqueous and blood-retinal barriers. NPs are, however, able to overcome these barriers because of their small size and highly variable surface properties. They can efficiently transport the drug to the targeted site with no toxic effects. Most of the NPs are biodegradable, which means they do not require surgical removal after they have delivered the drug [121,122].
Anterior eye diseases, such as cataracts, conjunctives, keratitis, dry eye, corneal injury, etc., are usually treated using eye drops but the corneal barrier causes drugs to have poor bioavailability. However, nanosystems can increase the bioavailability by prolonging the retention time of the drug on the surface of the eye and improving the penetration of the drug [123]. On the other hand, posterior eye diseases in the choroid and retina include retinoblastoma, glaucoma, choroidal neovascularization, macular degeneration, and posterior uveitis. Eye drops are not usually effective in treating these diseases, so interocular injections are performed, which leads to many unwanted side effects [124]. However, nanosystems have improved the delivery of drugs to the posterior portion of eye and the various nanosystems used for this purpose include nanovesicles, nanoimplants, NPs, and hydrogels [123].
Nanotechnology in the Treatment of Brain Diseases
Brain diseases can be treated efficiently if we can overcome the issue of the bloodbrain barrier (BBB). The BBB is a boundary between circulating blood and the neural tissues of the brain. The presence of the BBB is the major hurdle in the treatment of brain diseases because it does not allow the drugs to enter the central nervous system (CNS) and maintains homeostasis in the brain. Any disturbance to the BBB causes neuro-inflammatory and neurodegenerative diseases such as Parkinson's disease, Alzheimer's disease, etc., but even a damaged BBB does not allow drugs to enter the brain [125,126]. However, various types of NPs can cross the BBB and so can efficiently deliver drugs to damaged areas of the brain. NPs use organic and inorganic materials as a core to penetrate the BBB. Inorganic materials include silica, molybdenum, cerium, iron, and gold, whereas organic materials that can be used include PLA, PLGA, and trehalose. The distinct features by which NPs are able to treat neurodegenerative diseases are their small size, high drug loading ability, and efficient imaging performance (particularly for inorganic NPs). Some NPs themselves show some therapeutic efficacy, i.e., showing antioxidant properties, inhibiting Aβ aggregation, and reducing ROS levels [125].
NPs, when conjugated with ligands, show the best performance by interacting with BBB receptors at low density. NPs can adopt multiple pathways in order to cross the BBB [127]. The proposed pathways which NPs can use to cross the BBB are shown in Figure 3. The main pathways include
•
The paracellular pathway and passive transmembrane diffusion; • Transport proteins: carrier-mediated transport and efflux proteins; Through any of these pathways, NPs can cross the BBB and can be taken in by the neurons or active astrocytes of the brain [127,128]. In receptor-, adsorptive-, and carriermediated penetration of NPs across the BBB, various ligands are involved, such as:
•
Ligands that can adsorb proteins from the bloodstream [129]; • Ligands that can directly interact with BBB transporters or receptors [130][131][132]; • Ligands that can increase the hydrophobicity and charge of NPs [133]; • Ligands that can improve the circulation time of NPs in the blood [133].
The morphology and charge of NPs are also important in this case. Zwitterion and neutral NPs have a greater circulation time compared to positively and negatively charged NPs [134,135]. Overcoming the blood-brain barrier has enabled NPs to be used for the treatment of many diseases such as stroke, Alzheimer's, and Parkinson's disease, and many more, which are discussed below.
Role of Nanotechnology in Cancer Diagnosis and Treatment
Nanomedicine involves the implementation of nanotechnology in the treatment, screening, and diagnosis of various diseases, including cancer, and has the potential to revolutionize public and individual health [134]. In the formulation of various drugs for cancer treatment and in the discovery of cancer biomarkers, nanotechnology plays a vital role [136,137]. Through prediction, personalized therapy, diagnosis, medicine, and the prevention of cancer, it also contributes comprehensive techniques and worthy approaches against cancer [138].
The Utilization of Different Methods Involving Nanotechnology in Cancer Diagnosis
Obstacles in the early detection of different kinds of cancer are expected to be solved with the use of nanotubes, nanocantilevers, NP probes, and nanowire arrays [139]. Without the utilization of radioactive labeling or extrinsic fluorescent dyes, micro-cantilevers were used to determine single nucleotide polymorphisms in a 10-mer DNA target oligonucleotide [140]. Information about biomarkers related to the tumor microenvironment and the distribution, presence, and relative abundance of cancer signatures has been provided by probes with molecularly targeted recognition agents [139]. On the basis of fluorescence resonance energy transfer, nanoprobe systems have shown the potential to detect DNA mutations in tumor cells in clinical samples, as well as to detect the loss of DNA [141]. At the early stage of cancer, the expression of typical biological molecules has been detected using carbon nanotubes [142].
Different Methods Used for Cancer Treatment
To ease the intake of vehicles into target cells, a number of cancer-targeting ligands, such as growth factors or folate, cytokines and antibodies, have been used [143]. Through enhanced permeability and retention effects, caplostatin (TNP-470) was found to aggregate selectively in the tumor vessels, as well as halting the hyper-permeability of cancerous blood vessels [144,145]. By means of the EPR effect, NP-conjugated chemo-therapeutic substances-for example, angiogenic minute molecule inhibitor and doxorubicin-can enter into cancer cells, causing a growth inhibition and particular vascular shutdown [146][147][148]. By enhancing imaging and targeting cancer cells, oligonucleotides perform an important function. Furthermore, coupling them to metallic NPs, e.g., quantum dots and magnetic, ruby-eye doped, and gold nanoparticles, increases their related vasculature [149][150][151][152]. For the direct observation of circulating tumor cells in blood, a new SERS NP system was also introduced recently [153].
Targeting the Cancerous Micro-Environment
The arginine-aspartic acid-glycine motif has been found to display strong selectivity and affinity for the cell surface in many proteins; therefore, it is an appropriate ligand for therapeutic NPs targeting cancer [154]. In an inactive form as a pro-drug or nanoformulation, a drug that has a short half-life or that is extremely cytotoxic in circulation may now be controlled. Through tumor-specific molecules, these drugs are targeted at the tumor micro-environment. The tumor micro-environment eases its transformation to an active state upon arriving at its destination. By attacking both stroma and tumor cells, this tumor-activated nanoformulation therapy performs its function [155].
Consisting of a lower dose of chemotherapeutic medicines, metronomic therapy includes an administered schedule which lacks long periods of rest [156,157]. The novelty of this concept is in targeting the tumor micro-environment, specifically endothelial cells that are highly sensitive to the continual administration of low dosages of medicine as compared to cancerous cells. Thus, tumor angiogenesis is inhibited, causing the inhibition of tumor growth [158]. Poly-base NPs, upon aggregation in the low-P H micro-environment provided by the tumor tissue, become captured in the fenestrated cancerous vasculature and aid in the increased transfer of drugs to cancerous sites [159].
Targeting Drug-Resistant Tumors
Metastatic colon tumor cells that overexpress integrin α5β1 have been found to be targeted by PEGylated liposomes altered with a fibronectin-mimetic peptide [160]. NPs have been made to increase endocytosis when targeting multidrug-resistant tumors, or to bypass or inhibit efflux pumps on the membrane, considering various methods of drug resistance in cancer cells [161]. As an effective inhibitor of P-gp, TPGS 1000 (D-alphatocopheryl polyethylene glycol 1000 succinate) has become one of the dominant surfactants that increases the cytotoxicity of the G-185 cells of colchicines, doxorubicin, paclitaxel, and vinblastine, which can be compared to that of parental cells [162]. Another promising and important example of a modifying substance for P-gb is the pluronic block copolymer (P85). In P85 treatment, membrane fluidization gives rise to an interference with metabolic mechanisms and the inhibition of the P-gb ATPase drug efflux system [163]. The NP system is made up of a lipid shell and a polylactic-co-glycolic acid (PLGA) doxorubicin-conjugated polymer core composed of cholesterol, PEG distearoylphos-phatidylethanolamine, and phosphatidylcholine. To cause vascular disruption in cancer cells, these NPs are filled with a natural phenolic compound called combretastatin [164].
To be overexpressed on the surface of drug-resistant cancer cells, NPs with folate acid ligands that can attach to folate receptors were shown to obtain particular accumulations in cancerous cells. The NP transport system for the comprehensive and synergistic roles of cancer nano-chemotherapy may be improved by means of attempts to co-govern drugs with ultrasound and photodynamic therapy and thermosensitive therapy [165,166].
3.6.5. Personalized Therapy for Cancer ανβ3-targeted para-magnetic NPs have been used to study angiogenesis in great detail, as well as to study nascent melanoma cancers non-intrusively, [167]. The efficacy of treatment, specifically in case of melanomas, can potentially arise through early detection [168]. To target malignant tumors with greater specificity and affinity, NPs associated with bio-targeting ligands, for example, small molecules, peptides and monoclonal anti-bodies, can be employed. In regard to each person's molecular profile, these advancements offer opportunities for the development of tailored oncology, in which tumor therapy, diagnosis and detection are personalized [169].
Targeted towards tumors, tissue, cells, or organs, and governed by an external magnetic field, iron oxide NPs can be used to attach themselves with nucleotides, proteins, antibodies, and drugs. Towards cellular receptors such as urokinase plasminogen activator receptor, magnetic iron oxide NPs are currently being used. With the aim of improving the effectiveness of drug transport conducted by receptor-mediated endocytosis and for the destruction of cancer stromal fibroblasts, this nano-construct allows the immersion of drug-delivering NPs in the endothelial cell layer of the tumor, tissues, or cancerous cells. Due to their spatial imaging ability and increased biocompatibility, these iron oxide NPs may find significant applications in cancer treatment and imaging [170]. Breast cancer cells expressing HER2 receptors can be attached by means of the HER2 antibody (herceptin) trastuzumab, conjugated with iron oxide NPs. When attached to the tumor magnetically, it can exhibit increased antitumor activity with cytotoxic molecules such as doxorubicin [171].
Cancer Treatment through Thermal Ablation
Metallic NPs have shown a capacity for implementation in targeted hyperthermic therapy, particularly in the case of carbon nanotubes, gold silica nanoshells, iron oxide nanoparticles, and solid gold NPs [172,173]. To treat deep tissue cancers, iron oxide NPs have been employed as both therapeutic and diagnostic nanoscale agents [174]. An increased amount of gold NPs could be obtained by marking gold NPs with antibodies in opposition to specific tumor cells. The application of radio-frequency fields to cells caused the confined heat and death of tumor cells, after the incorporation of NPs [172]. Through the utilization of magnetically induced heat, magnetic NPs provide an encouraging method for the minimally intrusive excision of small cancers in the breast [137,175]. This approach has the value of allowing the refined and selective tuning of the degree of energy deposition, permitting sufficient temperature control at the target site, and this method also meets the enhanced need for breast-conserving therapies [176]. The anti-human epidermal growth factor receptor 2 antibody can be employed in transporting drugs to human epidermal growth factor receptor 2-overexpressing tumors, and can initiate an anticancer response [175]. A pegylated colloidal GNP (gold nanoparticle) consisting of tumor necrosis factor-α attached to its surface called CYT-6091 has been shown to increase thermal therapies and has been broadly investigated as an adjuvant [177].
Furthermore, to treat tumors via the induction of hyperthermia, superparamagnetic NPs show attractive properties [178]. There is a transformation of magnetic energy into thermal energy when these NPs are subjected to an alternating magnetic field of adequate frequency and strength. The heat produced is then transported to the cells revolving around the NPs. Once the protein denatures and the restricted temperature increases more than 40 • C, this can cause tumor cell death via apoptosis [179,180].
Alzheimer's Disease
Alzheimer's disease (AD), the most prevalent form of dementia, is a neurodegenerative disorder [181,182], the initial symptoms of which include impaired memory and declining cognitive abilities, which lead to damage to the motor system [183]. A lot of literature supports the positive relationship between the concentration of the soluble aggregates of Aβ peptide and the degree of dementia in AD patients [184][185][186][187]. They accumulate to form insoluble fibrils, which further aggregate to form characteristic plaques [188,189]. Hence, most of the current research is centered towards the prevention of their aggregation. For this purpose, nanomaterials are exploited, owing to their exceptionally small size and fit for crossing the BBB.
Grape resveratrol (a neuroprotective, anti-inflammatory compound [190]) and OX26 mAB-conjugated solid lipid nanoparticles (SLNs) can inhibit Aβ aggregation [189]. SLNs have a hydrophobic lipid core, which allows the dispersion of the drug, thereby increasing its bioavailability [191,192]. Moreover, they are rapidly opsonized and clarified from the blood stream, which presents a convincing argument that these SLN do not accumulate in the blood stream unnecessarily, hence cutting down on the associated threats [193]. Similarly, the monoclonal antibody against fibrillary human amyloid β42 was conjugated with iron oxide NPs that successfully targeted aggregates in the arterioles of mice [194].
Curcumin and water-soluble PLGA-NPs conjugated with Tet-1 peptide can also destroy amyloid conjugates [195]. Curcumin, which has anti-mutagenic, anti-inflammatory, antioxidant, anti-cholesterol, anti-tau hyperphosphorylation, and anti-amyloid properties, is an excellent candidate for AD treatment [196][197][198][199][200][201][202]. B6 peptide was conjugated with curcumin-loaded PLGA-NPs [203], which decreased the diameter of curcumin, enhancing its cellular uptake. Additionally, they prevented tau hyperphosphorylation and deposition and boosted learning and memory in mice. Memantine, a neuronal death-preventing drug [204], has also been loaded into PEG-coated PLGA-NPs, which reduced β amyloid plaques and the characteristic inflammation of AD [205]. Negatively-charged gold nanoparticles (AuNPs) have also been proven to be effective against amyloids, since in their bare form they inhibited Aβ fibrillization and dissociated fibrils [206]. This finding marks them as potential carriers for anti-AD drugs. Protein capped cadium sulphate (PC-CdS) and iron oxide NPs exhibit anti-tau aggregation properties, while keeping the viability of neuroblastoma cells intact. Moreover, PC-CdS NPs can also disaggregate Tau cells [194,207].
Parkinson's Disease
Parkinson's disease (PD) is the second most common neurodegenerative disease [208]. Dopaminergic (DA) neurons in the substantia nigra pars compacta selectively die and α-syn Lewy bodies start forming [209,210]. Six mutations are found to be behind familial PD-SNCA, DJ-1, parkin, PINK 1, ATP13A2, and LRKK2 [211]. Since α-syn's connection to PD emergence has been established, scientists are now searching for new methods to interfere with its expression [212]. In a recent study, an N-isopropylacrylamide derivative was immobilized on oleic acid along with short hairpin RNA (shRNA) and loaded in magnetic iron oxide NPs. Nerve growth factor was also added to N-isopropylacrylamide. ShRNA interfered with α-syn synthesis successfully, thereby making it a potential tool for treating PD.
Retinoic acid (a neuroprotective chemical) NPs have also been used and found to be therapeutic for DA neurons. They also induced the production of mRNA and transcription factor proteins that make the survival of DA neurons possible, namely, Nurr 1 and PitX. This makes them suitable for use in the prevention of PD onset [213]. The co-loading of curcumin and piperine, which have extraordinary cognitive and antioxidant properties, on glycerly monoleate NPs has been performed recently. These NPs were coated with several surfactants that increased the bioavailability of loaded compounds. According to in vivo results, they can inhibit α-syn and reduce oxidative stress, apoptosis, toxicity induced by rotenone, and restrain DA's neuronal degeneration process [214].
RNA interference (RNAi) can knockdown specific genes. In a study, α-syn-targeting RNAi and polyethylenimine NPs were used to treat incurable neurodegenerative disorders such as PD [215]. In a matter of 5 days, the α-syn level was reduced by almost 50% and no side effects (such as the induction of toxicity) were observed. PEG has also been exploited for use in the treatment of PD. PLGA-loaded NPs were coated with lactoferrin, which enhanced the delivery of loaded rotigotine. The results indicated that the cells that came in contact with it did not compromise their viability. In fact, free rotigotine was toxic. It was also reported that a high amount of rotigotine was heterogeneously distributed to the striatum, which is a primary affected region in PD [216].
Amyotrophic Lateral Sclerosis
Amyotrophic lateral sclerosis (ALS) is another neurodegenerative disorder which is characterized by affected upper and lower motor neurons, the spinal cord, and the motor cortex region. Abnormal amounts of mutant superoxide dismutase (SOD) are also observed in ALS patients. The orderly progression of the disease has been explained by misfolded SOD 1. Hence, research is largely channeled towards reducing the levels of SOD [217]. Antisense oligonucleotides (ASOs) can effectively silence the proteins but their inability to cross the BBB rendered them useless. However, to overcome this problem, ASOs were loaded onto calcium phosphate lipid-coated NPs. When these were negatively charged, they successfully delivered ASO into a neuron-like cell line [218].
Oxidative stress and damage have also been reported to contribute to ALS. Reactive oxygen species (ROS) and reactive nitrogen species (RNS) damage DNA, RNA, and other molecules [219]. Cerium NPs can take part in coupled redox reactions that neutralize ROS and RNS. This makes them suitable for use in antioxidant therapy for a number of neurodegenerative disorders, including ALS [220].
Huntington's Disease
Huntington's disease (HD), an autosomal-dominant neurodegenerative disorder, is characterized by anxiety, involuntary movements, and chorea [221,222]. It has been observed that biological materials obtained from patients with neurodegenerative diseases such as Huntington's exhibit oxidative stress and mitochondrial defects. HD brains have faulty electron transport chains as well [223]. 3-nitroproponoic acid (3-NP) is a neurotoxin which leads to the generation of ROS; therefore, researchers are looking for new agents that can potentially inhibit the production of 3-NP.
Curcumin has also been used to treat Huntington's disease, along with SLN. They can ameliorate 3-NP induced in HD mice by decreasing the amount of intermediate complex II activity. The signs observed include reduced swelling of the mitochondria, lipid peroxidation, and ROS production. Neuromotor coordination was also enhanced [224]. SLNs have been loaded with rosmarinic acid and introduced into 3-NO-induced mice and this also showed promising results [225]. β cyclodextrin (CD) NPs have also been used to carry siRNA, which can silence or modify the expression of mutant HTT. CDNPs decreased the amount of mutant gene mRNA dramatically and also showed partial toxicity, but the overall toxicity profile was satisfying [226].
Trehalose-loaded zwitterion NPs inhibit amyloid and polyglutamine aggregation in HD mice brains [227]. Their zwitterionic shell enhances cell uptake without inducing cytotoxicity. It hindered aggregation by forming multivalent bonds. This method requires trehalose in micro amounts; however, when used in molecular form it is needed in millimolar concentrations. Trehalose and zwitterion are other anti-amyloidogenic molecules.
Cystic Fibrosis
Cystic fibrosis is a fatal genetic disorder caused by a mutation in the cystic fibrosis transmembrane conductance (CFTR) gene. It is characterized by abnormal transportation in endothelium cells of a number of tissues. This leads to abnormally thick and sticky mucus, which blocks organs. The chief blocked organ is the lung. This translates into the emergence of recurrent bacterial infections, which destroy the lung tissues progressively, and as a result, pulmonary disease grows large enough to cause mortality [228,229]. To treat this disease, correction of the CFTR gene seems like an attractive option.
Chemically altered mRNA was loaded onto lipid NPs. It was reported that this allowed an increase in the number of membrane localized CFTR, restoring its primary function of acting as a chloride channel. Additionally, the nasal application of these NPs restarted chloride transport to nasal airway epithelium cells. This proved to be an effective tool for CFTR treatment that can be manipulated in the future.
PLGA-NPs coated with PEG have been employed for almost every genetic/neurodegenerative disorder, including cystic fibrosis as well. In this case, they effectively delivered anti-inflammatory compounds and CFTR correctors [230]. Another interesting approach employed for the treatment of CFTR infections was loading biodegradable NPs with antibiotics, such as ciprofloxacin complex. This approach was adopted to fight the infections that are characteristic of the disease itself. The bacteria which were targeted by these antibiotic-loaded NPs were Pseudomonas aeruginosa. Mucus was checked and analyzed for results after treatment. It was reported that colloidal stability was proven and that the mucus became noticeably less turbid, showing a decrease in pathogenic bacteria [231]. Table 2 summarizes the outcomes of some nanomaterials employed to treat various genetic disorders.
Huntington's Disease
Curcumin SLN Reduced the activity of intermediate complex II [224,227] Trehalose-loaded zwitterion NPs Inhibited amyloid and polyglutamine aggregation
Cystic Fibrosis
Lipid NPs Increased the amount of membrane-localized CFTR [230,231] PLGA NPs coated with PEG Effectively delivered anti-inflammatory compounds
Nanotechnology in the Treatment of Nervous System Diseases
Fortunately, NPs are not only valuable for genetic or neurodegenerative disorders; they have also been manipulated to treat other severe neurological traumas, such as in post-stroke neuroprotection and spinal cord injuries. These two areas might sound very complex, but studies have proven over time that nanotechnology can help to fight these severe diseases. Regeneration or repair in the central nervous system is another large problem. This is a prevalent problem because the damaged axons lack the ability to regenerate and regrow. The obstacles in the way of the regeneration of these axons are extrinsic inhibitory molecules and an age-dependent drop in intrinsic regenerative capacity, along with some other factors [232,233]. Presently, researchers are investing their efforts in looking for alternative ways to inhibit the action of factors that do not promote the growth and regeneration of damaged axons and neural cells. In this regard, nanotechnology is found to be a highly effective potential tool to treat central nervous system disorders [234].
For the treatment of spinal cord injuries, conventional drugs are used, which have become unpopular due to drawbacks associated with them. These drugs, when systematically administered, were found to be highly inefficient since they were metabolized rapidly before reaching the target and were cleared from the bloodstream. Now, the aim is to modify them in such a way that their bioavailability can be enhanced. For this purpose, adenosine was conjugated with lipid squalene into nano-assemblies. This method showed astonishing results, since the neurologic deficit score was improved and early motor recovery of the hind limbs was also observed [235].
Macrophages have been observed to perform a key role in the entire inflammation process in microglia and macrophages and they contribute to the chronic phase of neurodegeneration; hence, they have been established as a therapeutic target [236]. Polymethyl methacrylate NPs have been used to target specific cell populations of macrophages in order to decrease inflammation without exhibiting toxicity [237]. A charged surface and surface PEGylation enhance this process, allowing cellular uptake. This is a different approach for the treatment of such diseases. Table 3 summarizes the outcomes of some nanomaterials employed with and without drugs in an effort to treat some major neurodegenerative disorders.
Conclusions and Future Perspectives
Nanotechnology research has grown exponentially within the last few decades, and the focus on healthcare sectors has increased in parallel. Theranostic development has led to a significant amount of understanding of some of the complex etiologies involved, as well as increasing the chances of early diagnosis and therapeutic potential with the help of nanomedicine. Various nanosystems have been exploited and integrated at a limited scale but have proven to be efficient in solving various bottlenecks in various healthcare sectors. However, nanomedicines and nanodevices are still at an early developmental stage and one way to accelerate this process is to direct research studies so that researchers work towards developing new methods to overcome the associated limitations. The gradual development of nanotechnology-based methods has given rise to a hope that soon lifethreatening and disabling disorders will be effectively treated. The gaps due to inadequate efficacy and preclinical safety studies need to be filled on a priority basis so that we can make full and timely use of the great potential of nenotechnology, which is yet to be realized. Nanotechnology has a solution to many problems, but this does not mean in any way that there are no challenges or limitations associated with it.
One of the major obstacles in the implementation of nano-based products in living systems for healthcare services is toxicity. Various nanomaterials have triggered unwanted allergic and other reactions that can be potentially harmful to the body. Toxicity is a very complex concept in itself because it is dependent on a diverse range of factors such as morphology, size, dose, surface area, route, and duration of administration [246]. This directs our attention towards another area, which is the need to standardize or personalize the use of nanomaterials. Furthermore, the reliability and reproducibility of experiments involving NPs remains another area that needs to be worked on. Since these are extremely small entities, controlling their activity in sensitive environments is also hard. Some other limitations are their high cost, the presence of impurities, their environmental impacts, etc. [247]. If these dangers are not dealt with carefully, they may have seriously lethal repercussions.
Since nanotechnology is a relatively new area, it remains relatively underexplored. In fact, we cannot say that the physiochemical behavior of these NPs is fully known in vivo and in vitro. This is why we might not judge accurately that which type of nanomaterial will be used precisely for what purpose. There are some NPs that might be very useful in one system, but in other systems they might be entirely toxic. One example of such a case is PEI, which is an excellent transporter of nucleic acids, but it shows cytotoxic traits [237]. Another factor that we tend to ignore is that the NPs have varying compositions, sizes, and shapes and each of them has different impacts on living systems. Moreover, the duration exposure, as well as the coating, aggregation, charge and solubility of nanomaterials, also influence their performance [237].
Despite the sophisticated instruments and tools developed in recent times, we still need to develop modern tools that can quickly characterize synthetic nanomaterials, separately from existing analytical tools. We are also in dire need of establishing standardized protocols to synthesize these nanomaterials, not only ensuring high yields, stability, and purity, but also complying with the issued security guidelines. An efficient in vivo monitoring system can considerably boost biomedical processes such as the treatment and diagnosis of various serious diseases [248]. There is also a need to come up with mechanisms that help us to thoroughly understand the fate of NPs once they have been used. These questions include how long they stay in the body, what conditions impact the duration of degradation, how to make them stay for longer and shorter periods, what are their long-term and short-term impacts, how exactly does the body behave towards these outsider entities on a micro and a macro level, what are their characteristics and their mechanisms of action, and how we can standardize these particles to ensure the reproducibility of experiments. These should be addressed before the implementation of nanotechnologies in healthcare sectors. Apart from these, there are also many questions that require well-studied and well-experimented answers. We also need to identify the potential hazards associated with these nanomaterials in order to avoid any unforeseen circumstances. Furthermore, the different nanomedicines and nanoformulations targeting various diseases must be meticulously designed in order to achieve the safest and most efficacious therapeutic regi-men. We conclude with the vision that nanotechnology will push forward to develop more promising therapies to cope with various severe diseases, and will also provide researchers with effective tools to solve the various bottlenecks in healthcare sectors.
|
2021-08-29T06:16:17.950Z
|
2021-07-21T00:00:00.000
|
{
"year": 2021,
"sha1": "edf5505e13b69c51384c4fb811b5889020dc64bf",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/1424-8247/14/8/707/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "fb20f97277793ca8f4789c15abceaa14fe7f08d1",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
149828049
|
pes2o/s2orc
|
v3-fos-license
|
Theoretical Orientation of Mental Health Workers in Turkey
The aim of the study is to evaluate the primary theoretical orientation of mental health workers (psychiatrist, psychologist, psychological counselor and social worker) in Turkey. The study was conducted on 1073 (572 women, 501 men) mental health workers, including 133 psychiatrists, 312 psychologists, 430 psychological counselors, and 198 social workers. Mental health workers between the ages of 22 and 73, from every city in Turkey participated in the study. Personal Information Form designed by the researchers was used as data collection tool in the study. Personal Information Form was emailed to the professionals and the data was collected online. Collected data were analyzed by frequency and percentage calculations. The findings of the study demonstrated that the primary theoretical orientation of psychiatrists and psychologists were cognitive behaviorist therapy, of psychological counselors were constructivist therapy and of social workers were system/family systems therapies. Furthermore, it was observed that the least preferred psychotherapy approach by the professionals was the multi-cultural therapy. The results of the study were discussed within the realm of the literature, and recommendations were presented for mental health workers and researchers.
Psychotherapy theories aim to help and change the individual by enabling the individual's different thoughts, feelings and behavior (Corsini & Wedding, 2008;Sharf, 2015).Mental health workers, while fulfilling this objective, differ among each other due to their individual traits, their target population, their education and their theoretical orientation (Brammer & MacDonald, 2003;Corey, 2013;Gladding, 2009;Hackney & Cormier, 1996;Heffler & Sandell, 2009;Welfel & Patterson, 2005).Theoretical orientation makes it possible to assume hypotheses related to the behavior and experiences of the client/patient, to formulate specific therapy responses and to evaluate the therapeutic process (Poznanski & McLennan, 1995).Theoretical orientation affects several variables including the therapeutic process, the approach to human nature, therapeutic methods and techniques, as well as the function and the role of the therapist and display an intrinsic systematic structure (Gladding 2000;Peterson & Nisenholz, 1999).Thus, it could be stated that the theoretical orientation has an extremely critical effects on the therapist, the therapeutic process and the client (Buckman & Barker, 2010;Tartakovsky & Kovardinsky, 2013).Mental health workers, independent of their education, area of work and target population, have to consult a psychotherapy theory or theories for an efficient therapy/consulting/interview process.Sharf (2015) stressed that when the therapist would not utilize any theories in the therapeutic process, the therapist would have to proceed using only instincts and experiences, which would in turn limit the therapeutic process.
Psychotherapy theories differ in terms of the function and role of the therapist, psychotherapeutic goal and process, psychotherapeutic methods and techniques (Corsini & Wedding, 2008).There is no "right" or "wrong" psychotherapy.Each theory contributes to the understanding of individual behavior and therapeutic applications (Corey, 2013).Certain psychotherapy theories are centered on the individual (Rogers, 1963), while others consider the individual within the framework of the family and society (Goldenberg, 2012); some focus on the past experiences of the patient/client (Freud, 1914), while others focus on the present time (Perls, 1992).Corsini and Wedding (2008) reported that there are over 400 psychotherapeutic approaches.However, these approaches, rather than being individual psychotherapy theories, are the derivatives of psychodynamic, cognitive/behavioral, humanistic and transpersonal approaches (Peterson & Nisenholz, 1998).
Psychoanalytical therapy attempts to explain psychological disorders via subconscious processes, impulses and instincts and tends to treat the source of the disorder (Joannidis, 2006), while the constructivist approaches, instead of focusing on the causes of the problem, focus on the strengths and exceptional moments to adopt an approach centered on the solution (Sklare, 2005).Therapists that adopt the family systems approach focus on the understanding the family/relationship in initiating the change in the patient/client and necessity of the implementation of the therapeutic process on the whole family members (Nichols & Schwartz, 2004), while the therapists that adopt the client-centered approach put the individual at the center and believe that the individual is capable of solving his/her problems (Rogers, 1965).Existentialist therapists are against implementing a therapy using pre-established techniques and they focus on subjects such as freedom and responsibility, identity formation, evaluation of the meanings and objectives, awareness of death and extinction (Yalom, 1980).Therapists that adopt behavioral approaches, within the boundaries of scientific methodology, focus on the alleviation of the behavior that causes problems of adaptation in daily life based on certain methods and techniques (Kazdin, 2012).
Cognitive behaviorist therapists study focused on symptoms, attempting to exchange the negative schemes in the emotions and thoughts of the client with positive schemes (Boterhoven De Haan & Lee, 2014;Butler, Chapman, Forman, & Beck, 2006;Foreman & Pollard, 2011;Wollburg & Braukhaus, 2010).On the other hand, psychotherapists oriented towards multi-cultural therapy suggest that most psychotherapy theories were based on the basic values of European-Americal culture, thus it would be dangerous to conceive these as universal values (Corey, 2013).Psychotherapists oriented towards multi-cultural approach evaluate the patient/client within the context of the culture he or she belongs to, and utilize methods and techniques secondary to the culture of the patient/ client (Corsini & Wedding, 2008).Psychotherapists with an eclectic approach, instead of formulating a new theory on the personality of the patient/client, aim to harmonize the psychotherapy with the personality of patient/client.Eclectic psychotherapists assume that there are no universal mechanisms of change and try to benefit from every theory (Corsini & Wedding, 2008).Sue, Ivey and Pedersen (1996) argued that psychotherapy theories bear the traces of the culture/society they evolved in, one single theory would not be valid in all cultures/ societies, and thus there are different theoretical orientation in different cultures.Theoretical orientations of the psychotherapists were analyzed in counties such as the USA (Bike, Norcross, & Schatz, 2009;Fleischman & Shorey, 2014;Scandell, Wlazelek, & Scandell, 1997), Argentina (Castañeiras, García, Lo Bianco, & Fernández-Alvarez, 2006), Romania (Ciorbea & Nedelcea, 2012), Israel (Tartakovsky, 2014), Denmark (Jacobsen, Nielsen, & Orlinsky, 2012), Sweden (Larsson, Kaldo & Broberg, 2010), Portugal (Vasco, Garcia-Marques, & Dryden, 1993), the UK (Fear & Woolfe, 1999), and Australia (Byrne & Reinhart, 1990).However, there are no previous studies conducted on the theoretical orientation of the mental health workers in Turkey.It is considered that the investigation of the theoretical orientation of mental health workers in Turkey would provide information on which psychotherapy theory/theories are preferred, in which frequency, and by whom and what types of therapy service the clients/patients receive.The objective of the study is to examine the primary theoretical tendency of the mental health workers in Turkey.
Determining the preferred therapy in Turkey is critical.This may indicate what type of therapy the client receives.Indeed, the way in which each therapeutic approach looks at human nature, the methods and techniques used, and the therapeutic process differ from each other.There are limited therapeutic approaches specific to the culture of Turkey.Therefore, this research can raise awareness about cultural sensitivity.In this context, it is thought that this research is important in terms of mental health services.
The Process
In this research, firstly data collection tool was prepared.Personal Information Form was prepared for this purpose.Then E-mail was sent to mental health workers in order to participate in the study.Information form designed by the researchers after expert opinion was obtained, and similar studies and scientific books on psychotherapy theories that were reviewed, was used in the study.The form contains personal information on the professionals (gender, age, professional seniority, educational background, and the province that the professional works at) and questions to determine the primary theoretical tendency of the professional.Form was based on the selection of one of the 11 basic psychotherapy theories or expression of other theories if needed.The forms completed by 1130 professionals were examined and 57 faulty forms were discarded and analyses were conducted on remaining 1073 forms.The data collected were analyzed by calculating frequencies and percentage.
Findings
Table 2 show that the approaches that psychiatrists adopt most are cognitive behavioral approach (45.1%) eclectic/ holistic approach (23.3%) and psychodynamic approach (10.5%) and psychoanalytical approach (7.5%).The approaches that psychologists adopt most are cognitive behavioral approach (29.7%), eclectic/holistic approach (20.8%), psychodynamic approach (7.4%) and system/ family systems approach (7.1%).The approaches that psychological counselors adopt most are constructivist approach (23.5%), cognitive behavioral approach (21.6%), eclectic/holistic approach (15.6%), and client-centered approach (10.9%).The approaches that social workers adopt most are system/family systems approach (33.8%), constructivist approach (13.6%), client-centered approach (13.1%), cognitive behavioral approach (10.6%), and eclectic/holistic approach (9.1%)When the results of the research are evaluated in a more systematic way, cognitive behavioral therapy is the most preferred approach for both psychiatrists and psychologists.Indeed, one of the two psychiatrists sees the cognitive behavioral approach as the primary theoretical orientation.Similarly, one of the three psychologists sees the cognitive behavioral approach as the primary theoretical orientation.However, the constructivist approach is more prevalent in psychological counselors.Indeed, one of the four psychological counselors sees the constructivist approach as the primary theoretical orientation.Finally, one out of three social workers sees the system/ family systems approach as the primary theoretical orientation.Furthermore, it was observed that the least preferred psychotherapy approach by the professionals was the multi-cultural therapy.This finding of the research can be considered as interesting.
RESULT, DISCUSSION AND RECOMMENDATIONS
Since there is no mental health legislation in Turkey, it is not clear who could conduct psychotherapy.However, it could be stated that usually psychiatrists, psychologists and psychological counselors, and partly the social workers apply psychotherapy in Turkey.The education that psychiatrists, psychologists, psychological counselors and social workers receive, their target populations and fields of study differ.These differences make them utilize different theories in psychotherapy.The scholars explained the differences in theoretical orientation of the psychotherapists by multi-dimensional reasons.In these studies the differences in theoretical orientation were analyzed based on the therapist (attachment style, personal traits, demographic features, attitude, values and expectations) (Castañeiras et al. . 2006;Jacobsen et al., 2012;Larsson et al., 2010;Scandell et al., 1997;Vasco et al., 1993) and the variables of psychotherapy theory (applicability and efficiency of the theory on the case) (Negreiros, 1994;Oliver, Lightfoot, Searight, & Katz, 1990;Vasco & Dryden, 1994).Thus, the differences in theoretical orientation of the psychotherapists should be explained by multiple causality, instead of a single factor.It was determined in this study that approximately one of the two psychiatrists, and one of the three psychologists adopted cognitive behavioral approach as their primary theoretical orientation.The fact that cognitive behavioral approach was the primary theoretical orientation could be evaluated with relation to the therapist and the theory.
An evaluation based on theory would suggest that cognitive behavioral theory was preferred the most due to its symptom oriented structure, its ability to produce solutions to psychological problems within a short period of time (Boterhoven De Haan & Lee, 2014;Butler, Chapman, Forman, & Beck, 2006;Wollburg & Braukhaus, 2010), its structured and eclectic nature, its applicability in different cultures (Corey, 2013), its ability to examine the private experiences of the client/patient using a scientific approach (Weishaar, 1993), the fact that it was researched the most, its functional nature, its ability to provide a practical and efficient treatment, and its clear and intelligible characteristics (Leahy, 2002).In fact, Moyer, Sohl, Knapp-Oliver, and Schneider (2009) determined that in one third of all psychosocial responses conducted during the last 25 years utilized primarily cognitive, behavioral or cognitive-behavioral methods.Thus, it could be argued that the reasons for the cognitive behavioral approach to be the primary theoretical tendency of psychiatrists and psychologists are the sensitivity of the theory for different cultures, its ability to provide results in shortterm, its ability to allow for scientific assessments, it's efficiency in therapy, it's clear and intelligible nature for the client/patient, it's structured nature and its ability to allow for the utilization of different therapeutic techniques.
On an analysis based on the therapist; in Turkey, psychiatrists receive an undergraduate education in medicine, and they are specialized in psychiatry in graduate school of medicine.Because of the nature of the education, they receive (medicine and education of biological psychiatry) (MacCluskie & Ingersoll, 2001), they work mostly in psychiatric clinics in hospitals and often conduct adjunct psychotherapy and medicine therapy.In Turkey, psychiatrists deal with the examination, indication, classification and treatment of cognitive, emotional and behavioral disorders of adaptability to the environment of the clients/patients.When performing these actions, psychiatrists frequently resort to drug treatment since they lack sufficient training in applied psychotherapy (Öztürk & Uluşahin, 2016).On the other hand, psychologists graduate from psychology departments and usually work at hospitals, psychological counseling centers, and partly at schools providing individual psychological counseling and group psychological counseling (Sümer, Helvacı, & Mısırlısoy, 2013).
The educational programs for both psychiatrists and psychologists almost lack any formal training on the application of a particular theory.Furthermore, both psychiatrists and psychologists provide services to a vast number of clients/patients every day.This fact increases the workload of psychiatrists and psychologists immensely.Therefore, it could be deducted that psychiatrists and psychologists are in need of a structured, symptom-centered, efficient in treatment, open and intelligible approach like the cognitive behavioral therapy.Thus, it could be stated that their education and workload were influential factors in psychiatrists and psychologists adopting cognitive behavioral therapy as their primary theoretical tendency.Furthermore, the fact that Turkish people are externally controlled and dependent to authority figures is considered, it was stressed that cognitive behavioral therapy could easily be implemented on Turkish clients/ patients, and mental health workers are inclined to use the cognitive behavioral approach (Mocan-Aydın, 2000).Although in the past, psychiatrists and psychologists were inclined to psychoanalytic therapy in Turkey (Mocan-Aydın, 2000), it was determined because of this study that a serious change has occurred.Thus, it was observed that only a few psychiatrists and psychologists had the psychoanalytical approach as their primary theoretical tendency.
The primary theoretical tendency of psychological counselors in Turkey was the constructivist approach.Psychological counselors in Turkey graduate from the psychological guidance and counseling/psychological programs in departments of educational sciences of faculties of education in the universities and usually perform psychological counseling and guidance services for students, parents and school personnel.Psychological counselors, in addition to guidance services such as adaptive, guidance, moderating, developing and preventive services in educational institutions, perform psychological counseling activities as well.Although they often concentrate on educational, occupational and personal problems of the students, psychological counselors also perform protective and developmental activities in these areas.Psychological counselors deal with mild problems when compared to psychiatrists and psychologists.Psychological counselors, although aim to fortify the psychological health of the individuals, attempt to help them by acting upon their normal and healthy aspects even when they provide services to individuals with serious psychological problems.Hence, preventive and developmental principles are considered in the education of psychological counselors in Turkey (Turkish Psychological Counseling and Guidance Association, 2006).Ültanır (2005) stressed that psychological counselors working in educational institutions should help train students with strong traits, who could solve their problems in an effective and convenient manner.When the education, target population and field of study of psychological counselors are concerned, it could be stated that they are inclined towards constructivist therapies.Thus, constructivist therapies concentrate on the exceptional moments where the client is successful and discovery of strong traits (Sklare, 2005).In the past, psychological counselors had the tendency towards client-centered therapy; it was observed in the study that there has been a significant change in the theoretical orientation of psychological counselors.One out of four psychological counselors identified his or her primary theoretical tendency as constructivist approach in this study.On the other hand, client-centered therapy was the primary theoretical tendency of one in ten psychological counselors.
Another finding of the study was that one in every three social workers adopted system/family systems approach as his or her primary theoretical tendency.Social workers in Turkey graduate from social services undergraduate programs and work on the psychosocial effects, and results of issues such as psychiatric care, child protective services, domestic violence, poverty, racism and immigration (AASW, 2008).Social workers deal with the emotional and social problems that restrain the individual in these fields of study.They perform case studies and provide counseling services (Turkish Psychological Counseling and Guidance Association, 2006).
Family is at the center of social services applications.Social workers take into account also the family, family members and their problems while working with individuals with problems (Early & GlenMaye, 2000).
It could be considered that social workers are inclined towards system/family systems therapy because of that attention paid by social workers to the family.Thus, system/family systems therapy evaluates the individual within the context of the family and the society.System therapists put emphasis on the importance of individual within the systems of family and society, and scrutinize the behavior and expectations of the individual within the system (Corey, 2013;Goldenberg & Goldenberg, 2012).Although social workers utilize humanitarian occupational values such as autonomy of the individual, respect in human relations, protection of the pride and value of the individual, the findings of the study determined that they had the tendency to adopt system/family systems therapy.
Finally, it was determined that the mental health workers in Turkey adopted the multi-culture approach the least.Although there are no theories developed within the Turkish culture, the fact that mental health workers inclined towards the theories based on the western culture instead of multi-culture psychotherapy could be considered as very interesting.Thus, Corey (2013) stresses that most psychotherapeutic approaches were based on European-American culture, and it is dangerous to consider these as correct values and as universal.On the other hand, multi-cultural psychotherapy advocates cultural sensibility, awareness of cultural differences, respect and recognition for cultural differences.Multi-cultural psychotherapists consider differences of power such as ethnic origins, gender, social class, sexual orientation, age, and religion (Corsini & Wedding, 2008).It was considered that this finding was a result of little education on cultural sensibility (Bektaş, 2006), and the non-multi-cultural structure of Turkish system of education (Kaya & Aydin, 2014).
When the results of this study are considered; further studies could scrutinize mental health workers based on theoretical orientation.In addition, multi-cultural therapies should be prioritized in training of mental health workers.Finally, Turkish mental health workers should develop psychotherapy theories particular to culture and implement existing therapy theories by adapting them to the Turkish culture.Performing of this research with percentage and frequency can be considered as a limitation.
In subsequent studies, the theoretical orientation can be examined together with the causes.In addition, the relationship between the theoretical orientation and demographic variables can be examined.
|
2019-05-12T14:22:42.084Z
|
2018-01-01T00:00:00.000
|
{
"year": 2018,
"sha1": "e7185920392a878f83645359effc6fd72ceb4399",
"oa_license": "CCBYNCSA",
"oa_url": "https://doi.org/10.5455/jcbpr.278015",
"oa_status": "GOLD",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "e7185920392a878f83645359effc6fd72ceb4399",
"s2fieldsofstudy": [
"Psychology"
],
"extfieldsofstudy": [
"Psychology"
]
}
|
67752652
|
pes2o/s2orc
|
v3-fos-license
|
Comparison of ammonia emissions related to nitrogen use ef fi ciency of livestock production in Europe
for the effects of losses in the the for ef fi cient use of (N). Of in agricultural 80% to Because the largest atmospheric loss of reactive nitrogen from production systems is (NH
a b s t r a c t
The increasing global demand for food and the environmental effects of reactive nitrogen losses in the food production chain, increase the need for efficient use of nitrogen (N). Of N harvested in agricultural plant products, 80% is used to feed livestock. Because the largest atmospheric loss of reactive nitrogen from livestock production systems is ammonia (NH 3 ), the focus of this paper is on N lost as NH 3 during the production of animal protein. The focus of this paper is to understand the key factors explaining differences in Nitrogen Use Efficiency (NUE) of animal production among various European countries. Therefore we developed a conceptual framework to describe the NUE defined as the amount of animalprotein N per N in feed and NH 3 eN losses in the production of milk, beef, pork, chicken meat and eggs in The Netherlands, Switzerland, United Kingdom, Germany, Austria and Denmark. The framework describes how manure management and animal-related parameters (feed, metabolism) relate to NH 3 emissions and NUE. The results showed that the animal product with the lowest NUE had the largest NH 3 emissions and vice versa, which agrees with the reciprocal relationship between NUE and NH 3 within the conceptual framework. Across animal products for the countries considered, about 20% of the N in feed is lost as NH 3 . The significant smallest proportion (12%) of NH 3 eN per unit of Nfeed is from chicken production. The proportions for other products are 17%, 19%, 20% and 22% for milk, pork, eggs and beef respectively. These differences were not significantly different due to the differences among countries. For all countries, NUE was lowest for beef and highest for chicken. The production of 1 kg N in beef required about 5 kg N in feed, of which 1 kg N was lost as NH 3 eN. For the production of 1 kg N in chicken meat, 2 kg N in feed was required and 0.2 kg was lost as NH 3 . The production of 1 kg N in milk required 4 kg N in feed with 0.6 kg NH 3 eN loss, the same as pork and eggs, but those needed 3 and 3.5 kg N in feed per kg N in product respectively. Except for beef, the differences among these European countries were mainly caused by differences in manure management practices and their emission factors, rather than by animal-related factors including feed and digestibility influencing the excreted amount of ammoniacal N (TAN). For beef, both aspects caused important differences. Based on the results, we encourage the expression of N losses as per N in feed or per N in product, in addition to per animal place, when comparing production efficiency and NUE. We consider that disaggregating emission factors into a diet/
Introduction
Nitrogen (N) as a nutrient is an important contributor towards food security. With increasing world population, the demand for food increases thereby increasing the demand for reactive N. The intensification of agriculture over the last century has led to an increase in N recovery in livestock but also an increased N surplus . Inefficiencies in the production chain of food protein mean that N is lost, both as unreactive N 2 and as reactive N compounds (N r ), contributing the majority of the N r pollution of the global environment . Ammonia (NH 3 ), nitrous oxide (N 2 O), nitrogen oxides (NO x ) and nitrate (NO 3 À ), contribute to acidification, eutrophication and climate change, threating biodiversity, water-, air-and soil quality . When deposited from the atmosphere, N r cascades through a number of ecosystems (Galloway et al., 2003), with negative effects on N poor natural ecosystems in particular. Ammonia also contributes to the formation of secondary atmospheric particulate matter, with the smaller of these (PM 2.5 ) implicated in a range of adverse impacts on human health. About 80% of the N harvested in agricultural crops is used to feed livestock , Managing N flows in livestock systems is therefore of critical importance when seeking to reduce N r pollution. Several indicators are used to assess the efficiency of agricultural production and hence its likely contribution to environmental pollution. Nitrogen Use Efficiency (NUE), the ratio of output-N (in products) to input-N, is an indicator of the efficiency with which N input into an agricultural production system is converted to N in agricultural products. NUE can be calculated for a whole production system or for individual components (Gerber et al., 2014;Uwizeye et al., 2016). The NUE of feed (ratio of N in livestock products to the N input in feed) is one such component and is one indicator of agricultural sustainability and potential for improvement (Powell et al., 2010). Another indicator commonly used in relation to agricultural products is the N footprint (Leach et al., 2012), which is the total amount of N lost to the environment resulting from the production of a unit weight of product i.e. a measure of emission intensity. This indicator has the advantage that it includes all N inputs and losses in the food production and processing system, thereby enabling an integrated comparison for different production chains for the same product. However, it has the disadvantage that comparisons among products are difficult.
The main focus of this paper is on N r lost as NH 3 because NH 3 is the largest atmospheric loss of N r from livestock production systems . In 1999, the Gothenburg Protocol of the Convention of Long Range Transboundary Air Pollution (CLRTAP) established national emission ceilings for a range of gases, including NO x and NH 3 (UNECE, 2013). This was followed in 2001 by its EU equivalent, the National Emission Ceilings Directive 2001/ 81/EC Directive (NECD). Ammonia emissions, of which ca. 90% originates from livestock excreta, fell by 23% between 1990 and 2015. However, the emissions of other major atmospheric pollutants fell by 50e80% in the same period. Ammonia even increased by 1.8% in the EU 28 between 2013 and 2015 (EEA, 2017), so more effort is needed to reduce these emissions.
Understanding the cause of the variation in NH 3 emission among countries helps build confidence in the validity of national emission inventories and contributes to understanding how emission reductions may be achieved. This is an important part of periodic inventory reviews that are undertaken within the scope of the CLTRAP and NECD. The large number of parameters that influence NH 3 emission from livestock production including animal breed, livestock diet, housing, manure management, production parameters, climatic conditions etc., makes direct comparisons of NH 3 emissions among countries difficult. This means that at present, review teams can only calculate apparent emission factors (total NH 3 emissions per animal category), which can identify differences among emissions reported in different national inventories but not their origin. The main livestock products consumed in Europe are dairy (cheese, milk), pork, poultry meat, beef and eggs (EEA, 2017), so the categories for the production are dairy and beef cattle, pigs, laying hens and broilers. Common to these is the conversion of feed protein into animal protein.
Ammonia-N expressed per unit product or per N in product enables a direct comparison among different animal products, provides a better insight regarding the differences among countries and thus insight regarding the mechanisms of efficiency parameters of protein turnover. Therefore this paper compares NH 3 eN emissions among various European countries in relation to the production of animal protein from feed protein. The methodology is developed and implemented for the production of milk, beef, pork, chicken meat and eggs in The Netherlands, Switzerland, United Kingdom, Germany, Austria and Denmark. Using a conceptual framework of the N-flow in livestock systems, this paper aims to establish which parameters we need to focus on, distinguishing animal-related factors (quality and quantity of the feed, protein turnover and consequent NUE) on the one hand and manure management, explaining the fraction of total ammoniacal N (TAN) that emits as NH 3 eN, on the other.
Conceptual framework
A conceptual framework was developed of NH 3 emissions from livestock and manure management (Fig. 1). This framework represents a simplification, since manure management is considered as a single entity, whereas in practice, it comprises several elements (e.g. livestock housing, manure storage). The source of the volatile NH 3 in manure is TAN. The production of TAN from N in feed is shown in Fig. 1. A proportion d of the feed N (Nfeed; kg day À1 ) is digested and absorbed by the livestock, yielding an amount of metabolisable protein. A proportion p of the metabolisable protein is then converted to N in milk, meat or eggs. For milk, the whole of the product is of value for human nutrition and this is essentially also true for eggs, since the shell forms a small part of the product. For beef and pork, the inedible part of the carcass (mainly bone) is significant. However, this study was a comparison of the environmental impacts of raising different types of livestock, rather than of individual livestock end products. Moreover, since bones etc. are needed to produce the edible portion of the carcass, we considered it appropriate for this study that the N in the whole carcass should represent the product N. The N retained in animal protein (Nproduct; kg day À1 ) is therefore: And the NUE of feed (NUEfeed) is: The remainder (1-p) of the N in metabolisable protein is excreted to the urine (milk, beef and pork production) or in droppings (poultry), and can be considered to be TAN (TANex). A part of the feed protein will normally be indigestible and a proportion of the digestible feed protein is used to form microbial protein in the gut. These are excreted as organic N in the faeces. Organic N also enters manure through spilled feed, defined as f. A proportion (m) of the N in manure can be mineralized (TANmin), and thus also contributes to the amount of TAN. The TAN produced (TAN; kg day À1 ) is therefore: A proportion of the TAN (e) volatilizes as NH 3 eN from manure in livestock buildings and manure stores, following manure application to land and from excreta deposited during grazing. The masses of N 2 OeN and N 2 eN emitted are much less than that of NH 3 eN hence they are omitted here. The value of e depends on numerous factors, of which livestock type, temperature, air velocity, pH and concentration of TAN are the most important. The NH 3 eN emission is therefore: The NH 3 emission intensity (AEI), which is the NH 3 eN emission per unit of N in the product (kg kg À1 ) can be calculated from (Eq. (5)): From Equation (2), Equation (4) can be written as Equation (5): This equation shows that the AEI will be related to the inverse of the NUEfeed (further referred to as NUE) and that the nature of that relationship will depend on a combination of feed/animal characteristics (d, p; i.e. TAN excreted) and manure management system characteristics (f, m, e). Fig. 2 represents protein deposition given sufficient energy supply. When the metabolisable protein (MP) supplied in feed is insufficient to satisfy the animal's maintenance protein requirement, protein will be remobilised from body tissues, the animal will lose weight and no protein deposition in egg, milk or meat will occur. With increasing MP intake, protein deposition increases until the maximum (potential) protein deposition (PPD) is obtained (Whittemore and Fawcett, 1976). The efficiency with which the MP in excess of the maintenance requirement is used for protein deposition varies, depending on the quality of the MP (in particular, the balance of essential amino acids) and on the type of protein (egg, milk, meat) being deposited. The animal's genetic capacity limits PPD. Ideally this occurs with a feed intake supplying the minimum protein needed for potential protein deposition (FPP). Above FPP, surplus feed protein N is all lost as TAN in urine, illustrated by the linear relationship of NH 3 and NUE with increasing feed protein content in Fig. 2. The figure illustrates firstly that by decreasing FPP by maximizing p while avoiding protein in feed beyond FPP minimizes NH 3 eN emission per kg product-N. Secondly, NH 3 emission will decline if e is reduced. This will lower the NH 3 curve parallel to the y-axis. Finally, m should be minimized during storage. In practice, digestibility of Nfeed in intensive livestock systems only varies between about 0.7 and 0.9. However, for cattle on extensive grassland, lower digestibility feeds are encountered, especially outside the growing season (Armstrong et al., 1986).
If e was equal, differences among countries would arise through: genetic potential of maintenance requirement (MR); genetic potential of PPD; FPP; actual feed protein. The amount of protein in the feed, FPM and FPP depend on inherent genetic properties, but also on feeding management, including protein quality and essential amino acids (Fig. 2).
The main manure-related factors influencing m and e are: surface area; temperature; air velocity; pH (Webb et al., 2012). Ammonia emission is also related to manure management: how manure is stored; where it is stored; for how long. Some countries have introduced emission reduction techniques. Variation of these aspects within the theoretical framework will move the red line up parallel to the Y-axis (high e), or down (low e) (dotted red lines).
Data analysis
The national emission inventories of The Netherlands, Switzerland, United Kingdom, Germany, Austria and Denmark were used to estimate NH 3 emission from livestock production. These models and data regularly undergo a detailed external review for quality control and quality assurance and were tested for congruency by Reidy et al. (2008Reidy et al. ( , 2009). The principal mass flow approach compared well among the national models when all use the same input values for TAN excretion (following d and p) and NH 3 emission factor (e).
Data were collected from the six countries for dairy, beef, fattening pigs, laying hens and broilers. For Austria poultry production is of minor importance and was excluded from the analyses. The data used for each livestock category were: N intake and excretion; percentage of TAN in excreted N; the NH 3 emission factors (% of TAN) at housing, manure storage, during grazing and from manure after field application. To express in relation to N in product, data were collected on growth, milk production and egg production as well as the N-content of livestock animals, eggs and milk. The N contained in meat was defined as the N-content of the ground carcass. For dairy, only milk production was included, not the meat of the cull cow after milk production or of the calves produced during the cow's lifetime. For eggs the shell was included but the meat at the end of the laying cycle was not. Parental animals or young replacement animals were not considered. The only Ninput therefore was the N in feed. The N-content of the products was calculated as the mean of the national data, apart from Austria, for which there were no national data and Swiss values were used.
The data were used to calculate NUE as Nproduct/Nfeed. The NH 3 eN lost was expressed per N deposited in product. The implied emission factor (IEF) was calculated as the sum of all NH 3 eNemissions over the manure management chain, expressed as % of TAN.
Annual NH 3 emission can be considered as the product of the processes determining the amount of TAN that is susceptible to volatilisation and the IEF, with the former described in Equation (2). The component d(1-p) describes the processes that take place within the animal while the component m(1-d) describes those taking place in the manure management chain (Fig. 1). The factor m can take a positive or negative sign; in some manures, mineralisation of organic N leads to the formation of TAN whereas in others, there is an immobilisation of TAN into organic N (Jensen and Sommer, 2013). Not visualised in Fig. 1, but influencing m, is the amount of bedding material, affecting m through availability for microbial energy needs, but also adding a small amount of N to the manure. Not all countries take m into consideration in their NH 3inventory model (Reidy et al., 2009). Some countries (e.g. DK, NL) have made more effort to reduce NH 3 emission, resulting in a variety of abatement techniques for housing, storage and field application, further contributing to variation in manure management. NH 3 eN losses per kg N in feed for all animal products were assessed and compared using a double sided two-sample Student's t-test with P < 0.05.
The origin of the variation in emission intensity within a product can be explored by separating out the contribution TAN excretion makes to NH 3 emission (NH3 TANex ) from that made by the manure management system (MMS) (NH3 MMS ). For any dataset containing estimates of NH 3 eN emission and TAN excretion, we can calculate the mean NH 3 eN emission (NH3 ave ) and TAN excretion (TANex ave ). Using the subscript i to identify an individual element in the dataset, then: The sum of NH3 TANexi and NH3 MMSi expresses the extent to which the NH 3 eN emission from element i deviates from the mean of the dataset. If plotted, a quadrant is drawn with clockwise a: low NH3 TANex , high NH3 MMS ; b: high NH3 TANex , high NH3 MMS ; c: high NH3 TANex , low NH3 MMS ; d: low TANex, low NH3 MMS Table 1 presents for each country per animal place and year the total amount of N fed, the excretion of TAN by the animal, the IEF as percentage of excreted TAN and the production of N in the animal product (egg, milk or meat). NUE of the feed did not differ as much as TAN and IEF due to less variability of Nfeed and Nproduct, with a coefficient of variation (cv) of 5e14%. The IEF had the largest cv varying between 27 and 39% with the smallest variation for milk and pork, and the largest for chicken.
Results
The Nproduct of milk and egg are easy assessable and frequently monitored, especially for milk because the market price depends on the protein content, so nationally-derived statistics were used for these data ( Table 2). The Nproduct of meat is costly to assess and not routinely measured. When measured, it is the N-content of the carcass, which for this study is considered the same as the N-content of the meat. The mean of values given by the NL, CH, UK, DE and DK were used for these data rather than national-specific values ( Table 2). The variation coefficients were relatively low except for the protein content of the chicken due to the low Ncontent of UK chicken. Fig. 3 shows the reciprocal effect between NUE and NH 3 eN emission per Nproduct. In terms of NUE the order from low to high was: beef 0.22; milk 0.26; eggs 0.29; pork 0.35 and chicken 0.55. For NH 3 eN this was chicken 0.27, pork 0.53, milk 0.65, egg 0.70 and beef 0.92. The product with the lowest NUE had the largest NH 3 emissions and vice versa, which agrees with the reciprocal relationship within the theoretical framework shown in Figs. 1 and 2. For all countries, NUE was lowest for beef and highest for chicken, with the production of 1 kg N in beef requiring about 5 kg N in feed, of which about 1 kg N was lost as NH 3 eN. In contrast, the Fig. 2. Protein deposition, nitrogen use efficiency (NUE) and ammonia (NH 3 eN) emission related to digestible protein in feed. FPM and FPP are digestible feed protein at maintenance level and potential production level respectively. The solid red line is NH 3 eN emission per kg protein deposition (EI), the solid green line is NUE (no dimension), and dotted red lines are lower or higher NH 3 eN emission due to factors related to manure management. production of 1 kg N in chicken meat required about 2 kg Nfeed and 0.2 kg was lost as NH 3 . The production of N in milk required ca 4 kg Nfeed with ca 0.6 kg NH 3 eN loss, the same as with eggs and pork, but those needed only 3 kg Nfeed per kg N in product. From Equation (6), we note that the slope of this relationship will be e (d(1-p)þm((1-d þ f)) and if extrapolated, it will intersect with the x-axis where NUE ¼ 1 (theoretical point where all the Nfeed is deposited in the product and no NH 3 emission occurs). Furthermore, we note that since Nproduct appears in the definition of both AEI and NUE, the relationship AEI versus 1/NUE is identical to NH 3 eN versus Nfeed. Table 3 gives the NH 3 eN for every kg N that is fed to the animal. It shows that the NL chicken meat has the lowest environmental impact because only 8% of the N-input is lost as NH 3 eN. Swiss pork is highest because 35% of the Nfeed is lost as NH 3 eN. Variation among countries is large, the data do not provide enough information to significantly discriminate among animal products, except for the difference between beef and chicken. Fig. 4 plots NH3 MMSi against NH3 TANex from which the relative contribution to NH 3 eN emission can be seen. For all animal products, the differences among countries in MMS are larger than in TANex. For chicken, the differences in TANex are smallest and for beef largest. Livestock products from the NL are always in quadrant c or d indicating the lower emission factors by MMS. The results for DK are also in c and d except for chicken, which are in quadrant b (relatively high for both TANex and MMS). The converse is the case for CH and DE, they are always in a or b, with highest MMS emissions. Noticeable is the exceptional position of UK beef in quadrant c with very high TANex and at the same time low MMS emissions. The opposite is the case for CH in quadrant a. Secondly it is noticeable that UK has quite low MMS emissions for chicken. The small MMS emissions for UK beef is due to the animals grazing all day for ca 6 months of the year and depositing ca 55% of their annual N excretion to grassland. Ammonia emissions from excreta deposited on grazed pastures are only about 20% of those from excreta deposited in and around buildings and handled as manure. This is because the N in urine is predominantly in an organic form (mainly urea) and most of this infiltrates rapidly into the soil, before hydrolysis to TAN can occur. The small MMS emissions for broilers is because emissions from UK broiler buildings, estimated as 10% of TAN excreted, are less than those estimated for the other countries.
Discussion
The reciprocal of NUE reflected the differences in AEI (Fig. 3). Although differences among countries were large, beef, with the lowest NUE had the highest NH 3 eN loss per kg Nproduct and chicken meat with the highest NUE, the lowest. Since Nproduct appears in the definition of both AEI and NUE and the relationship AEI versus 1/NUE is identical to NH 3 eN versus Nfeed we encourage Table 1 Feed-N, TAN excretion, implied emission factor of NH3eN (IEF), N production in kg/y per animal place and Nitrogen Use Efficiency of the feed (NUE) defined as N production/ Feed-N, CV is coefficient of variation. the expression of N losses per N in feed or N in product, in addition to the loss per animal place, when comparing production efficiency and NUE. The large losses from beef production and the small losses from chicken production correspond with the results of Leip et al. (2014) who calculated the N footprint with the Nitrogen Investment Factor (NIF) of different food products, defined as kg N-input per kg N product at the farm level. For beef, they calculated 15e20, compared with 3e5 kg N/kg N for chicken, pork, milk and eggs. This is equivalent to 1/NUE in this study which was 5 for beef, and 2e4 kg/kg for chicken, pork, eggs and milk. The differences in between our findings and those of Leip et al. (2014) occur primarily because we limited this study to the N flows in the animals and MMS, so did not include soil processes. However, it is possible the gaseous products associated with the denitrification process (N 2 O, NO and N 2 ) were relatively more important for beef systems. This would be consistent with the greater N 2 O emission factor (2.0%) used for IPCC for N deposited during grazing than for N applied to soil in manure (IPCC, 2006). We also limited our study to the N input in feed for adult animals and did not include the animals reared for replacement. An additional factor is that the protein contents in Table 2 are measured as that of the ground carcasses. The greater proportion of non-edible product in beef production compared with poultry meat means that our approach increases the proportion of feed N captured in the beef. This study did not take inputs from litter into account, which are an additional N source for dairy and beef especially. For dairy cows, for example, 1.5 kg of straw per cow per day would increase the organic N input compared with Nfeed by ca 5%. A fraction would be converted into TAN, depending on the net mineralization, being the sum of mineralization and immobilization of organic N. As Webb et al. (2012) describe, this depends on various factors including temperature and the availability of oxygen, which depend on manure management. Overall, it was considered that omitting the litter N from the calculations had little impact on subsequent results and interpretation. Leip et al. (2014) also considered crop growth in their meta-analysis. Yet the similarity of the results for chicken, pork, milk and eggs is remarkable, indicating that NH 3 is the major N component of the N footprint. Finally, the fact that we did not take into account the meat production of the milked cow, can affect balances among the environmental impact of products. In terms of the ranking of livestock products according to the environmental impact per unit of product, our findings are consistent with LCA studies in which eutrophication impacts are expressed per kg of the most economically important fraction(s) of the carcass; i.e. beef has the greatest impact per kg and poultry meat the smallest (e.g. De De Boer, 2010, Williams et al., 2006).
The breakdown of feed-protein into NH 3 eN as shown in Fig. 1 gives the NH 3 -emission per kg Nproduct as presented by the red solid line in Fig. 2. From Equation (6) we know the reciprocal relation between NH 3 and NUE, this is visualized in Fig. 2 with the solid green line. When feed protein > FPP, NUE decreases because p Table 3 NH 3 eN losses per kg N in feed for all countries and animal products, the mean per product and the variation coefficient (%). For the last column, different letters mean significant difference between rows based on two-sample t-tests. decreases and with increasing amount of digestible protein (d), TAN increases linearly. With equal urine volume, this means an increased concentration of TAN, resulting in a linear increase of NH 3 emission. If the animals are fed to supply their specific protein demand, they will be fed at FPP. With phase feeding and the addition of supplementary amino acids, farmers try to feed closer to FPP. However, this is expensive and to be safe and ensure the genetic potential of production is realised, farmers will often feed above this level, using lower cost formulations that ensure essential amino acid requirements are met but others may be exceeded. Table 3 gives the NH 3 eN for every kg N that is fed to the animal. With livestock being the main source of NH 3 eN and NH 3 eN being the main atmospheric Nr of livestock production, this factor gives a good impression of the acidifying impact of a system within the identified boundaries, in this case from the N in the feed to the N on the field. The results of the LCA review of De Vries and De Boer (2010) showed less impact on acidification (i.e. NH 3 emissions) for milk and egg production than for meat products. However, they expressed the impact per kg product which in the case of milk and eggs contain more water and are therefore hard to compare with meat. In the case of this study the impact of milk would then have been ca. 5 times less, and of eggs, ca 1.5 times (Table 2). Expressing N-losses per Nproduct is a better measure for N efficiency or N recovery when comparing across products. The variation in this study was much smaller than the 80% observed by De Vries and De Boer (2010), nevertheless, the data did not provide enough information to significantly discriminate among products, except for chicken and beef (Table 3). To explain differences among countries we have to observe the different processes as described theoretically in Figs. 1 and 2 and the calculated results in Fig. 3. These results indicate that to explain differences among countries for NH 3 emissions due to production of pork, chicken, milk and eggs, priority should be focussed on the differences in MMS among countries, rather than on the TANex. Fig. 4 visualizes this more closely. For beef, TANex might be an interesting factor to observe more closely, although given the high IEF (Table 1) the emission factors should not be neglected. A low NUE for grazing livestock is not necessarily an environmental problem: i. because these livestock convert plant material that is inedible by humans into meat and milk, and ii. because the NH 3 eN emission per Nfeed is still acceptable, since the NH 3 emissions from excreta N deposited during grazing are much lower than from excreta N flowing in manure management systems. Understanding the differences brings us closer to reducing the environmental impact of livestock.
The results in Fig. 4 imply that, with the exception of beef production, the differences in NH 3 emissions among countries are mainly a result of manure management factors, rather than due to differences in TAN excretion. With beef, large differences are caused by TANex as well. Theoretically variation in TANex depends on genetic potential of maintenance requirement (MR), genetic potential of protein deposition (PPD), the amount of feed protein at maximum protein deposition (FPP) and the actual feed protein (Fig. 2). For the countries considered in this study, the differences in animal genetics and in diet formulation are probably less than differences in manure management practices (particularly considering the uptake of ammonia emission mitigation techniques). A more global study including a much greater diversity in systems may reveal larger differences in genetics and diets.
A low NH3 MMS (quadrant d and c) in Fig. 4 reflects efforts made to achieve emission reductions. In the NL and DK reduced emission manure application is mandatory and reduced emission housing for pigs is standard procedure. Also, reduced emission poultry housing is standard in NL, and only a small amount of the poultry manure is applied to land, because it is burned for electricity or exported. Denmark has a high NH3 MMS for poultry (quadrant b) because it is just a small market and the effort to invest in reduced emission buildings would have little impact at the national scale. It should be noted that differences among countries are not implying good or bad in terms of agricultural sustainable performance. For instance, animal welfare considerations in CH demand that livestock have access to a floor area larger than in other countries and this leads to larger emissions and values end up in quadrant a or b. In contrast, the NH 3 eN emission for beef in the UK in quadrant c, is below the mean of the dataset. Here, beef production is mainly pasture-based, which results in the TAN excretion being above the mean of the dataset. However, this is outweighed by the much lower NH 3 emissions from the excreta deposited on pasture than from housed animals. Secondly, as Norton et al. (2015) noted, interpretation of NUE gives an idea of potential for improvement. Another context might give another view on efficiency as illustrated by OECD (2008) using gross N-balances as agro-environmental indicators referring input of N (volatile N compounds, fertilizer-N, manure-N, biological N-fixation and atmospheric N-deposition) and output of N (arable crop-N, fodder crop-N and pasture-N) resulting in the highest NUE for CH (0.6) and the lowest for DK and NL (0.4e0.45) in 2008, giving another view on improvement. In this context, reducing NH 3 emission would not improve efficiency, because the lower N input of NH 3 eN would be completely compensated by the higher manure-N input. In our study NH 3 eN reduction is a key factor to improve NUE and the environmental impact of livestock production.
Conclusions
Across animal products for the countries included in the study, about 20% of the N in feed is lost as NH 3 . For chicken meat NH 3 eN/ Nfeed is significantly lowest at 12%. For the other products NH 3 eN/ Nfeed are 17%, 19%, 20% and 22% for milk, pork, eggs and beef respectively, these were however not significant due to differences among countries. To produce 1 kg of N in chicken, pork, eggs, milk and beef, 2, 3, 3.5, 4 and 5 kg of N in feed are needed respectively. NH 3 eN/kg Nproduct (Ammonia Emission Intensity) and NH 3 eN/ Nfeed are considered better metrics to compare the environmental impact of livestock than NH 3 eN emission per animal place or per kg product and are good indicators for assessing the efficiency of potential mitigation measures. At the same time, high emission intensities may reflect trade-offs with animal welfare and with the conversion of human-inedible protein sources (e.g. forages) to edible animal products by ruminants.
On the basis of the conceptual framework presented for NH 3 emissions from livestock production systems, it appeared that the larger part of the differences among countries were caused by differences in manure management practices and their emission factors, rather than by TANex and feed digestibility parameters, except for beef where both aspects are of importance.
The expression of N losses from animal production is assisted by using NUE as an indicator. Furthermore, the disaggregation of the emissions into a TAN effect and an effect of the manure management system is a useful way to help understand differences in emissions among national NH 3 emission inventories and form the basis of a discussion during the periodic NH 3 inventory review.
|
2019-03-11T17:20:17.458Z
|
2019-02-01T00:00:00.000
|
{
"year": 2019,
"sha1": "74925bfe4ebbb9997b193aacad8d8aa523752ae4",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1016/j.jclepro.2018.11.143",
"oa_status": "HYBRID",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "5ea49bd4501412154a78b0b25795ea7d686b6e1f",
"s2fieldsofstudy": [
"Agricultural And Food Sciences"
],
"extfieldsofstudy": [
"Medicine",
"Biology"
]
}
|
133488039
|
pes2o/s2orc
|
v3-fos-license
|
Estimating the Effect of Attending a Public versus a Private University in Colombia on Academic Achievement
We evaluate the impacts of attending a public university in Colombia on the academic achievement of graduates from higher education. Our measurement of academic achievement represents the progress made between the college entrance and graduation standardized test scores. We find that public Higher Education Institutions (HEI) improve student test scores in 11 of the 12 programs analyzed. The superiority of public HEIs relative to private ones suggests the need to promote greater regulation of the latter, and a review their current standards to help bridge the gap that currently exists in terms of the value added public HEIs have in comparison to the private ones. It also suggests that, at least in the short run, it could be socially beneficial to expand the public provision of some of the higher education public programs that added more value.
Resumen
Se evalúa el impacto de asistir a una universidad pública sobre el rendimiento académico de los egresados de la educación superior en Colombia.El rendimiento académico se define como el progreso entre el examen de entrada a la universidad y los resultados de las pruebas estandarizadas justo antes de la graduación.Se encuentra que los programas de Instituciones de Educación Superior (IES) públicas mejoran los resultados de los estudiantes en 11 de los 12 programas analizados.La mayoría de los resultados son robustos una vez la muestra se condiciona a programas de mayor calidad (por ejemplo, programas acreditados o las mejores universidades del país).La superioridad de las IES públicas en relación con las privadas sugiere la necesidad de promover una mayor regulación de estas últimas y una revisión de sus estándares actuales para ayudar a superar la brecha existente en términos de valor agregado de las IES públicas en comparación con las privadas.También sugiere que, al menos en el corto plazo, podría ser socialmente beneficioso ampliar la provisión pública en algunos de los programas públicos de educación superior que agregaban más valor.
Códigos JEL: A22, I2, J3, D3, O3 We would like to express our appreciation for the helpful comments of participants at the Economics Seminar of Banco de la Republica (Central Bank) of Colombia in Bogota, at the workshop for the Regional Study on Higher Education in Latin America and the Caribbean in the World Bank, and in particular, those of Maria Paula Gerardino and Maria Marta Ferreyra.We also thank Daniel Velez, Andrea Franco and Gabriel Suárez for their helpful comments.We also appreciate the access to data provided by ICFES.The works published are provisional, and their authors are fully responsible for the opinions expressed in them, as well as for possible mistakes.The contents of the works published do not compromise Banco de la República or its Board of Directors.
We would like to express our appreciation for the helpful comments of participants at the Economics Seminar of Banco de la Republica (Central Bank) of Colombia in Bogota, at the workshop for the Regional Study on Higher Education in Latin America and the Caribbean in the World Bank, and in particular, those of Maria Paula Gerardino and Maria Marta Ferreyra.We also thank Daniel Velez, Andrea Franco and Gabriel Suárez for their helpful comments.We also appreciate the access to data provided by ICFES.The works published are provisional, and their authors are fully responsible for the opinions expressed in them, as well as for possible mistakes.The contents of the works published do not compromise Banco de la República or its Board of Directors.
I. Introduction
For a developing economy like the Colombian one, which has higher education enrollment rates that reflect its level of development, its challenges lie in continuing to increase its enrollment rates with equity while focusing its efforts on improving quality.These objectives are essential to improving efficiency in allocating labor in the economy and the productivity of that labor, preparing the population to take advantage of technological innovations, creating new markets, and developing a much more formal and equitable labor market.
Higher education enrollment rates in Colombia grew significantly in the period.Since 1996, in particular, a major change has occurred in the distribution of workers across the different educational levels.The share in this distribution has grown substantially for those with postsecondary education (See Lopez, 2010).By 2012, total expenditure on tertiary education in Colombia as a percentage of GDP was 2%, higher than Germany, 1.4%, and France, 1.5%, although still lower than the United States and South Korea, 2.7% (OECD, 2015).
Even though the total expenditure on higher education in Colombia as a percentage of GDP was not very different from that of these developed economies, the share of total expenditure coming from the public sector is smaller than in all countries but South Korea.
While the percentage of expenditure on tertiary education paid for with public funds is 45% in Colombia, it is 55% in the United States, 83% in France, 96% in Germany but only 34% in South Korea (OECD, 2015).Differences in the share of publicly provided tertiary education as well as their respective institutional differences might contribute to explaining the differences in educational achievement between Colombia and these countries.
Privately provided higher education can contribute to improving the quality of higher education on several of the desired dimensions outlined by Barr (2009).First, it can increase competition, and thus generate better incentives for universities to react to what students and firms require.According to the international literature, policies that correctly align incentives for schools have a greater probability of providing high quality education than those focused on expanding their resources, a framework largely valid in the case of universities (Hanushek, 2003).Secondly, private higher education also serves those students who are able to bear the cost of education and expect to be compensated for their investment by the market afterwards.
Furthermore, the existence of private options reduces the chance they will compete with students with limited resources at public universities.
In the case of the United States, Eide et al. (1998) found that graduates from elite private undergraduate colleges are more likely to attend a graduate school than graduates either from lower quality private undergraduate colleges, or from public colleges of any quality.
Furthermore, Brewer et al. (1999) found that annual earnings are consistently higher for graduates from private undergraduate colleges than they are for graduates from any public college regardless of its level of quality.In addition, Black et al. (2005) provided evidence that, regardless of whether the universities were public or private, increases in college quality pass a simple social cost-benefit analysis.
Providing high quality tertiary education requires supervising and regulating both public and private universities.Lack of regulation and adequate incentives, regardless of whether there is an increase in higher education enrollment rates, is likely to lead to much lower quality education (See Rau et al., 2013).
The debate about the lack of quality higher education in Colombia, its broad heterogeneity in some academic programs fields, and barriers to access for certain segments of the population has been extensive and has generated several studies, discussions and protests by students, teachers, and other stakeholders. 1A key point of contention in this debate is the difference in the quality of education provided by public versus private institutions and the differences in the subsequent results they generate in the labor market.These results provide further evidence of the differences in the quality of secondary education provided by public and private institutions, previously documented for Colombia by Guarín, Medina and Posso (2016), and also to the public funding of education in private institutions, previously documented by Bettinger et al. (2016), Angrist et al. (2002), Angrist, Bettinger, Kremer (2006) and Bettinger, Kremer and Saavedra (2010) in the case of secondary education, and by Attanasio et al.
This document is part of recent literature that estimates the value added that higher education gives academic performance in Colombia in different contexts. 2We estimate the differential effect of studying at a public university versus a private one on the value added.
Given the endogenous character of attending a public or private university, we require an identification strategy that recognizes this selection issue.Like Balcazar and Ñopo (2014), who studied the effect of enrolling in one education program rather than another, this study obtains matching estimates to identify the causal relationship between the type of school and the value added to academic performance.
1 See studies such as World Bank (2003), OECD and World Bank (2012), Saavedra and Medina (2012), etc.An example of public statements on the issue was registered with the discussions and marches were held in Colombia in the months of October and November 2011 regarding the amendment to Law 30 of 1992 proposed by the government.
Our academic achievement measure measures student's progress from the time he or she started college, as measured by the score on the ICFES test or SABER 11, to the time of graduation, measured by the score on the ECAES test or SABER Pro. 3 Given the limited comparability between the subjects evaluated in the ECAES tests within the different programs, the analysis in this document is done at the university program level.This effect can be understood as a proxy for the value added generated by higher education, in particular, the existing differences between the generated value in public and private institutions.
To calculate the differential effect on academic achievement of studying at a public or a private higher educational institution (HEI), differences in differences (DID) matching estimators are obtained using the method proposed by Abadie andImbens (2006, 2011).We match the ECAES and ICFES test results of students that took the ICFES tests between 2000and 2004, and the ECAES tests between 2004and 2008. .We found that the value added of public universities is positive and significant for both men and women in 11 of the 12 programs tested.The positive effects of public universities on engineering programs are robust to analysis within different samples based on different criteria for quality (those at accredited universities, in selective programs which received students with good ICFES test scores, or the ones that have been internationally well-ranked) for both men and women.It is important to notice that in the case of men, much of the effect in engineering, despite being positive, disappears once the sample is restricted to selective programs.However, we find positive and robust effects for both men and women in education, accounting, and dentistry programs.Most of the results obtained in the case of public higher educational institutions are maintained when attendance at a private university is the treatment of interest.
The overwhelming superiority of the public versus private HEI, suggests the need for greater regulation over the latter, to review its current standards, and possibly, of the accredited status that some of these institutions hold so as to help bridge the gap that currently exists in terms of the value added of the public versus private HEI.
The article contains seven sections including this introduction.In the second section, the factors that may differentiate public and private institutions of higher education are discussed.These are the possible mechanisms by which universities could affect academic performance.In the third section, the construction of our data set is described and the main descriptive statistics are provided.Subsequently, some stylized facts about the Colombian labor market and system of higher education are shown the fourth section.In the fifth section, the econometric methodology used to estimate the effect of public/private higher education on academic achievement is presented.The main results of our exercise are described in the sixth section.The main findings and conclusions are highlighted in the last section.There are some studies that show that public and private universities generate different results in society, although the evidence for Colombia and Latin America is scarce.Looking across countries Glomm and Ravikumar (1992) studied the effect of human capital accumulation financed through public and private institutions on the evolution of income inequality.The authors found that although public education quickly reduces income inequality, private education generates higher per capita income.Winkler (1995) found that, in the case of Latin America, the public university budget was insufficient during the 80's and 90's and that these institutions used to pay teachers low salaries, which ended up generating negative incentives for research quality.He suggested improving the efficiency of the system by allocating the higher education budget based on performance criteria.
II. Background and
More recently, Del Bello (2002) has shown that the growth of private universities in Latin America has been significant 5 , although that growth has taken place at the cost of lower quality.That reduction in quality has been partially due to a poor institutional approach which does not allow appropriate regulation and supervision mechanisms to be generated.According to Del Bello (2002) private university growth has been concentrated in countries like Brazil, Mexico and Colombia.In the case of Chile, Bellei (2007) evaluated the educational reforms through the academic achievements of public and private universities.The author finds that, in general, there were no differences between public and sponsored private universities, and where there are differences, they favor the public university.Moreover, Bellei (2007) argued that, in Chile, those reforms had increased segregation and inequality in education while there was no evidence of overall improvement in the educational system.
Finally, referring to the case of Colombia, Rodriguez and Ramos (2014) found that the private institutions in the Caribbean region of Colombia had significantly lower academic achievement with respect to the public universities in 2009, a fact which is particularly important for law, medical, and engineering programs.Even though they include an aggregate value concept, their framework and empirical application are very different from the ones we will consider next.
In the literature where the value added of higher education in academic performance in Colombia is estimated, Saavedra and Saavedra (2011) studied the value added of higher education on critical thinking, and on skills to solve problems and communicate, by comparing the results on cohort standardized tests of students in their first and last year at the university.In a recent article, Riehl, Saavedra and Urquiola (2016) explore the effect on the academic achievement and earnings of studying at a public or private HEI.Similar to this document, their measure of value added is each student's percentile relative to all other students in their sample in the same exam field and cohort.According to the authors, colleges' value added on earnings and on academic performance are not perfectly correlated, with private HEI adding more value on earnings and the public ones on academic achievement.The authors nonetheless, do not attempt to identify a causal relationship that would allow them to determine whether a specific individual would do better by enrolling in a public versus a private HEI. 6 6 Here we provide a clear identification strategy (see section III).There are also significantly differences between theirs and our sample.While Riehl, Saavedra and Urquiola (2016) study a sample with approximately 81,000 graduates from 157 colleges, here we study 129,387 students from 201 HEI, and 14 academic programs.These differences are partly explained by their interest in focusing on graduated students, with earnings observed in their graduation year.In addition, they study the period 2009-2011
b. Mechanisms
The mechanisms by which public or private universities may affect academic achievement are primarily concentrated in three main areas.First, there are differences associated with endowments to institutions of public and private education such as course materials, infrastructure, and teacher quality, etc., which are important for their performance.
Second, there is the institutional approach, which provides different incentives for teachers and administrators of public and private educational institutions.Third, there is the role played by classmates in public and private educational institutions.In short, the social environment of the two is very different and this can have an important influence on the students' academic performance.
Among the differences in the types of input used by public and private institutions, one of the most significant ones is related to the quality of their teachersand the effect this has on a student's performance.Hanushek (2002) suggests that teachers with better academic performance and better standardized tests perform better in the classroom.Rivkin et al. (2005) showed that a low-income student can achieve academic results that are comparable to those of a high-income student, if she or he has a teacher whose quality is above average.Hanushek (1992) showed that a student with a "good" teacher could have an academic performance equivalent to a student who is one academic year ahead but has a "bad" teacher.In addition, the academic programs at public universities have (on average) greater recognition in terms of quality than private universities.10According to the National Accreditation Committee (CNA),11 a university or high quality program in Colombia must focus its work "towards an ideal of excellence" and must be able to "show high quality by specific results, consolidated tradition, impact and social recognition."Currently, 31% of programs offered by public universities are highly ranked by the ministry of education.In the case of programs at private universities, only 21% are equally recognized.12However, according to the Ministry of Education of Colombia, 30% of the public universities are accredited as high quality institutions while, in the case of private universities, 31% are.Similarly, of the total research groups recognized by the Administrative Department of Science, Technology and Innovation (COLCIENCIAS) 13 , 51.4% belong to public universities while 48.6% belong to private universities.14Yet, when the top 50 universities in Colombia are analyzed using the methodology of the QS University Rankings Latin America, 31 out of the 50 universities are private. 15 second factor that generates differences between public and private institutions of higher education is the institutional approach, one of the main determinants of incentives for teachers and students.Although public and private universities have, on average, similar levels of experience measured in years of operation, there are fundamental differences in key areas such as budgeting, teacher incentives, curricula, and accountability. 16he economic literature states that the identification, retention and motivation of capable teachers are key tasks in the administration of educational institutions.The compensation systems in the public sector tend to be rigid and bureaucratic, with wages that often do not distinguish by grades or courses, nor are they based on historical performance (Neal, 2002).Private universities are not always restricted in these ways.For example, a private university in Colombia could make special offers to teachers who are hard to find for certain subjects while a public university would not have the same flexibility.To a large extent, this is due to the high degree to which the sources of funds for public institutions are centralized.Jaramillo (2003) showed evidence that private universities in Colombia are more decentralized in funding and have greater sources of income.Melo et al. ( 2014) estimated that the Colombian government invested an annual average of 0.93% of GDP in higher education between 2000 and 2012, but funding for private universities comes primarily from the collection of fees.
According to the Observatorio de la Universidad Colombiana, a typical private university in Colombia charged between 2 and 28 times the monthly minimum wage for tuition in 2014, with an average of about 6.5 million pesos per semester.
The institutional framework is also reflected in the curriculum.Jaramillo (2003) indicated that in Colombia, private universities are more likely to offer degrees in social and administrative sciences, engineering, and technologies while public universities are more likely to offer degrees in education, humanities, agricultural sciences, and natural and exact sciences.Jaramillo (2003) concluded that this is due to the internationalization approach of the private sector.
Finally, it is important to consider teacher incentives and mechanisms for accountability.Literature on secondary education in the United States provides some examples.
Neal ( 2002) stated that ideally, in an institution of public secondary education, administrators should be allowed to compensate the best teachers better, but there is the risk that variability in teacher's compensation might become a tool for administrators to favor the teachers who share their own political affiliation, or to practice nepotism.Hanushek (2002) claims that, although a way to make public school teachers in the United States accountable has been to evaluate the performance of their students through standardized tests at the level of each State, that approach is limited to reaching the proposed accountability goals.Indeed, Neal (2002) found that some standardized tests at the state level do not produce results that are consistent with test results at the national level.The same thing was found by Koretz (2002), andKlein et al. (2002).Jacob and Levitt (2003) also presented evidence of cheating on standardized tests and cheating that is specifically associated with changes in teacher incentives.
There is no indication of how institutional differences affect university teachers in Colombia.What is evident, as previously mentioned, is that teachers in public universities are, on average, more educated and have more attractive contracts than private university teachers.
A final factor to consider is the role played by peers in public and private educational institutions.In general, public universities have higher thresholds for admission and, on average, receive better students.For the 2007-2012 period, using figures from the Ministry of Education, the rate of entry to public universities, measured as the ratio between the number admitted and the number enrolled was 25%, while for the private universities it was 78%.In the same period, the average ICFES exam score at public universities was 0.24 standard deviations higher than at private universities.
However, students entering private universities come from households with higher income, better educated parents, and smaller families.For the 2007-2012 period, 6.9% of the families of students who entered public universities had incomes that were above 3 minimum wages, unlike private university students, for whom the same ratio was 18.0%.Also, 6.5% of the students who entered public universities during this period reported that their mothers' educational level was tertiary university or higher, while this figure was 15.6% for students who entered private universities.Finally, 8.3% of the students who entered public universities lived in households with 4 or fewer members, and 17.7% of private university students had households with similar sizes.
On the other hand, authors like Eide et al. (1998) have provided evidence that elite universities, most of these private, increase the odds of attending graduate school.This may be linked to social network dynamics within private universities.
III. Data
The main information source we used was the databases provided by the Colombian Institute for the Evaluation of Higher Education (ICFES), 17 in particular, the information collected when students take the SABER 11 (or ICFES) and SABER Pro (or ECAES) tests.We also used other sources of information provided by the Ministry of National Education (MEN) 18 and the CNA, 17 ICFES is the acronym in Spanish for el Instituto Colombiano para la Evaluación de la Educación Superior. 18MEN is the acronym in Spanish for Ministerio de Educacion Nacional.
which provide most of the information on supply and demand of education at the secondary and tertiary levels.
Based on this information, we built a database containing information about students, their families, and the school they attended at the moment they took the ICFES tests as well as their respective ICEFS and ECAES tests scores.It also included the type of higher education institution they attended.We used the ICFES scores as each student's baseline outcome, and the ECAES as his follow-up outcome.
The ICFES test provided our baseline information for about 2,120,797 students who took it between 2000 and 2004.In addition to the score obtained by each individual in each of the areas assessed, the database includes information related to the high school from which each individual graduated as well as their socioeconomic background at that moment. 19 The ECAES test provided follow-up information for about 421,423 students who took it between 2004 and 2008.We focused on the 2004-2008 period because the structure of the ECAES test was relatively stable during these years, with field-specific questions. 20Beginning in 2009, the ECAES examination went through a process of continuous changes, due to Decree 3963/2009, that made it impossible for us to compare those years to the previous ones.In particular, during the 2009-2011 period, the test included generic skills in written communication, problem solving, critical thinking, interpersonal understanding, reading comprehension and English which were applied to all fields.Also, starting in the second half of 2011, field-specific questions were eliminated, and the different fields were put together into 30 reference groups, so that 30 tests were constructed.Finally, even though the ECAES test was not mandatory in the 2004-2008 period, the number of people who took the ECAES at that time was approximately equal to the total number of people who graduated from the universities considered in our empirical exercise. 21The ECAES dataset also contained information about the institution of Higher Education the student attended when he took the test. 22 19 The test assesses 8 areas of knowledge: Biology, Mathematics, Philosophy, Physics, Chemistry, History, Language and Geography. 20It is important to note that in the 2004-2008 period all programs were not assessed, therefore, in our exercise only those programs that were evaluated during the period of interest and had enough information are included. 21Our empirical exercise considers 201 Colombian universities, and a total of 14 academic programs.On average and including all academic programs at 201 universities analyzed between 2004 and 2008, 73,327 students graduated while 79,290 students took the ECAES test. 22Information regarding the nature and origin of the HEI was obtained by cross-checking the institution's code with information provided by the MEN, which contains these features for 333 HEI nationwide.The HEI can be by nature: a technological school, technological institution, university, professional and technical university; while the origin may be: a corporation (unofficial) foundation (unofficial) governmentally established at the state (department), municipal, and national level, and special regime.
The master database with all the students' baseline and follow-up information, contains a total of 129,387 students (50,637 men and 78,750 women).The database was complemented with additional information obtained from the MEN and CAN datasets, containing characteristics of the schools, and the accreditation status of each higher education program.
The school information is obtained from the 166 MEN Resolution, which contains its nature (public or private), type of schedule, and whether it was all-male, all female or co-educational.
In order to check the robustness of our exercises, the data contain information that allows us to filter students, institutions or programs based on their academic quality.(i) academic reputation, (ii) the reputation of employers of graduates, (iii) ratio of the number of teachers to students, (iv) citations of articles published by its researchers, (v) articles per teacher, (vi) proportion of teachers with doctorates, and (vii) the importance of their presence on the web.contain 14,234 men and 21,339 women who were identified as students of programs that are accredited by the CNA (representing 42% and 40% of their total population respectively).The fifth and sixth databases contain 16,207 men and 24,756 women who studied in selective programs (representing 48% and 46%, of their total population respectively).Finally, the seventh and eighth databases contain 11,323 men and 15,293 women who studied in the top 20 universities in the country (representing 33% and 28%, of its total population respectively).
Among the selected programs (see Table 2), Bogotá represents the largest share of students, especially when one considers the private universities.Cali, Medellin, Barranquilla and Bucaramanga account for 25% of the male students in public universities and 33% in private universities; and for 20% and 34% of the women in public and private institutions respectively.For the national total, there is an average of 1.1 male and 1.7 female matched students in private universities that took the ICFES and ECAES tests for every 1,000 inhabitants while for the five main cities that average varies between 1.7 and 5.5 in the case of males, and between 2.5 and 8.8 in that of female students.Table 2 also illustrates that in all areas assessed in the ICFES test, there is positive selection of students in public HEI, that is, higher average scores in HEI, and negative selection regarding social variables such as parents' education, household size and family income.This is discussed in greater detail with respect to the math and language results for each of the analyzed programs in Tables 3 and 4. In them, we present the average scores in math and language in each program for public and private HEI, and for both men and women in accredited and non-accredited programs in each case.In most of the cases, public universities accept students with higher average results in both mathematics and language.
For men in non-accredited programs, the average results in mathematics and language for public HEI are at least two additional points above private HEI in 6 and 7 of the analyzed programs respectively (Table 3, columns xxvii and xxix), while private HEIs have an advantage of that magnitude in only the administration program (in mathematics) for accredited programs.
In the case of women in non-accredited programs, Table 4 shows that the average ICFES scores in math and language for those in public HEI are at least two additional points above the private HEI in 3 and 6 programs respectively (Table 4 columns xxvii and xxix).In no case do private institutions have an advantage of that magnitude.Within the accredited HEI, the results are even better for the public institutions.
The academic achievement variable measures the difference between ECAES and the ICFES standardized scores, where standardization is calculated for individuals who studied in the same program and took the ECAES. 24he ICFES variable is measured as the average score in three areas of general knowledge: mathematics, language and biology.The results are robust to other definitions.The difference between public and private in the average ICFES score is lower when only accredited programs are considered, and the variation is greater in private universities.The difference between the ICFES test results for students at public and private HEIs brings with it implicit differences in other areas.Since our work attempts to identify the students' gains in academic achievement in public versus private undergraduate programs, the variables we control for in the baseline were selected while bearing in mind the need to comply with the assumption of conditional independence.The variables considered are meant to allow us to minimize the possibility that matched treated and untreated students were not comparable.
The control variables for this exercise can be grouped under four categories: individual, home, school, and place where the ICFES test was taken (see
N. of Individ.
N. of Individ.
Math
Lang.Note: In the column for women's participation, the results of four programs appear in gray to highlight that these are 100% men or 100% women given the previously explained selection program criteria by gender.Source: ICFES, ECAES Tests; authors' calculations.
a. Stylized Facts
As was mentioned above, the labor market and the higher educational system in Colombia are deeply linked.The basic labor market indicators -Participation Rate, Employment and Unemployment -for the last 25 years show that there have been significant differences based on educational level.For example, individuals with higher education have usually had higher share of the market and employment rates as well as lower unemployment rates (see Medina and Posso, 2010).
It should be noted that between the second quarter of 1994 and 1999, there was an increase of about 10 percentage points in the unemployment of unskilled labor from levels close to 6% to levels of around 17%.For skilled labor, the increase in unemployment was approximately 11 percentage points.However, it remained at lower levels than those recorded for the least skilled workers, going from a rate of 4% to 15% by mid-1999 when the Colombian economic crisis reached its peak.
In turn, the participation rate and employment of individuals with higher education between 1984 and 2006 were significantly higher compared to those without higher education.
Between 1994 and 1999, when unemployment increased for both the skilled and the unskilled laborers, the participation rate of trained men was higher than that of the unskilled.This pattern is similar to the one observed for the employment rate which, although it declined for the analyzed period, remained higher for the skilled workers.
There are also important differences in labor income by educational level.Medina and Posso (2010) showed that, based on the unconditional earnings associated with higher education, finishing college has substantial effects on earned income.Posso (2008) showed that earnings related to secondary and basic education are much lower than those obtained with any higher education degree.
However, despite the better income associated with higher education, it must be noted that there are deep differences in wages among individuals with higher education.Medina and Posso (2010) used the standard deviation of log hourly wage for individuals with and without higher education to illustrate that a significant proportion of the growing wage inequality in Colombia has occurred within the most highly educated group.This may be associated with the significant growth of higher education enrollment rates in Colombia that began in the midnineties as documented by Posso (2008).This unprecedented growth in higher education enrollment rates, especially in private institutions, produced a significant effect on the formal and informal labor markets.As of 2010, while the employment of people with primary education remained at 1985 levels, that of people with a secondary education had doubled, and employment of those with at least some university education was about four times that of 1985.25The growth experienced by institutions of higher education and professional and technical programs brought with it deep changes in the labor market.Table 6 shows that the main departments (regions) and, probably, the major metropolitan areas saw the most growth in higher education.The departments (regions) with more university students in private HEIs per capita are Bogotá, Atlántico, Santander, Norte de Santander, Valle del Cauca, Caldas and Antioquia.Note also that technical and technological education is mainly offered by public institutions.As is the case in our data, the census shows that about 72% of higher educational institutions are concentrated in the main urban centers around the country.
IV. Methodology
To estimate the differential effect of studying in a public versus private higher education program on academic achievement, we obtained differences in differences (DID) matching estimators (Heckman, Ichimura andTodd, 1997, 1998).The DID matching estimator allows us to obtain unbiased estimates when we are able to control for the baseline variables that determine whether the individual will be selected into the treatment group.In addition, the DID allows us to control for unobserved characteristics that are fixed over time.
Note that standardization also allows the individual's progress with respect to the distribution of their peers in the program to be observed. 26The treatment variable, , is an indicator variable that takes the value of one if the individual was treated and 0 otherwise.
Treatment depends on the nature of the higher educational institution (HEI) where the individual finished his studies, which can be public or private. 27The parameter of interest is the impact on the standardized test scores of the value added of studying in a higher education public/private program, i.e., the average treatment on the treated individual, ATT. 28For example, if we wanted to estimate the ATT associated with attending a public versus a private 26 Alternatively, a linear regression could be estimated where the outcome variable is standardized ECAES scores, and the ICFES score is an additional control on the right side of the equation.Overall, the findings of our exercise are not modified by this specification change.However, it should be noted that our identification strategy allows us to control for unobservable factors that are constant over time, while the alternative specification would require additional assumptions. 27The implicit assumption is that individuals do not move between universities or programs. 28ATT is the abbreviation for the average treatment of the treated individual.
HEI, takes the value of 1 if the individual studied at a public HEI and 0 otherwise.If the ATT was meant to assess the effect of studying at a private versus a public HIE, would take the value of 1 if the individual studied at a private HEI and 0 otherwise.In this paper, both results are presented.
Let us define as the outcome under treatment at time , where = 0, 1, with = 0 representing the baseline given by the time immediately before enrollment in the higher education program, and = 1 the follow up, given by the end of the program.A vector X of individual characteristics associated with the individual, his family, and the school he attended before going to the university is observed at the baseline.The ATT is defined as: Component E[ 11 − 00 |, = 1] is the expected value added to the score between the moment students graduated from high school, 00 , and the moment they graduated from the public higher education program, 11 given their observable characteristics, .E[ 01 − 00 |, = 0] is similarly defined for those who graduated from a private HEI.The DID matching estimator of the ATT is robust to the presence of unobservable components that are persistent over time (separable) and that could bias our estimate. 29Identification of the ATT parameter requires the estimating E[ 01 − 00 |, = 1] = E[ 01 − 00 |, = 0], i.e., it requires that the treatment, , conditional on , not be used to predict changes in 01 − 00 . 30The estimator DID matching ATT, ̂ , is given by where is the total number of individuals who were treated.To estimate ( ̂01 − ̂00 ), matching estimators are used.In the literature, there are different methods that adjust to the above conditions and are analyzed in detail by Imbens (2004). 31In this exercise, 29 For example, cognitive and non-cognitive skills that are fixed over time after finishing high school but differ between test subjects and controls.Likewise, it controlled by geographical elements that, although fixed in time, may vary between test subjects and controls. 30It also requires that 0 < Pr( = 1|) < 1. 31 According to Imbens (2004) these methods can be grouped into 5 categories: (1) based on the estimate of regression functions of the outcome variable depending on the control variable methods, (2) matching with covariates, (3) based on the propensity score or probability of participation in the program methods, (4) combinations of methods (1), ( 2) and (3), and (5) Bayesian methods.
the matching estimator proposed by Abadie andImbens (2006, 2011), which makes it possible to obtain the Bias-Corrected Matching (BCM), is used.This estimate assumes that the selection of students for treatment is exclusively based on the observable characteristics of students at the baseline, and it is robust to the presence of unobservable components that are persistent over time. 32able A.2 in Appendix 2 presents the definitions of our control variables.However, even after controlling for selection based on these variables, the matching estimators might still exhibit some forms of biases which the BCM estimator allows us to control for (see Abadie andImbens 2006, 2011).Some generalities of BCM method are presented here.
a. Bias-Corrected Matching: general aspects
The main objective of this methodology is to find a consistent estimator of the counterfactual scenario ( − ) for the treated individuals.The estimator of the counterfactual scenario ( − ) is given by the equation where = { , , … , } is the set of nearest neighbors for individual , such that ≤ and = ∑ ( − ) = .
The choice of each best neighbor, ,, must meet the following conditions: (i) () = 1 − , where is equal to one if the individual was exposed to treatment.
where 1 { } is an indicator function equal to 1 if the expression in brackets is true and 0 otherwise.‖‖ = ( ′ ) 1 2 ⁄ and corresponds to the Mahalanobis distance. 33 Just as in Abadie and Imbens (2006), the Abadie and Imbens (2011) estimator uses matching with replacement, i.e., it allows each individual within the control group to be used more than once as a match for the treated individual.Abadie and Imbens (2006) showed that estimators matching with neighbors generate a bias in finite samples that causes the estimator ( ̂01 − ̂00 ) in equation ( 3) to be an inconsistent estimator for ( 01 − 00 ) in general.This bias is associated with the fact that the number of neighbors is fixed.
Intuitively, the matching estimator will be biased in finite samples when the pairings are not exact, which is particularly important when a large number of covariates are included.
The BCM estimator proposed by Abadie and Imbens (2011) adjusts the differences between different pairings using the differences between the values of their respective covariates.To make adjustments feasible, the dimensions of covariates are reduced using regression methods.
Thus, the DID matching estimator with bias correction, ̂ ,
𝐴𝑇𝑇
, would be given by where ̂ is the bias correction which is given by the following function 33 Mahalanobis distance is defined as . 34 Abadie and Imbens (2011) showed that the ECM estimator is robust even when ̂0() regression is incorrectly specified.
V. Results
Before illustrating our estimates of the parameter of interest, ̂ , , it is important to show whether the matching procedure leads to balance in the covariates in .Abadie and Imbens (2011) proposed a simple informal balancing test which is done in four steps: 1.All covariates of vector are normalized so that they have a mean equal to 0 and a variance equal to 1.
2. Differences in the unconditional mean for each covariate between all treated and control individuals are calculated before the matching procedure.
3. Using the matching method described in the previous section, the best matches for each individual are found.We fix = 4.
4. Finally, the differences in the means of the covariates between each individual and his matched peers are calculated.Once these differences are obtained, the mean and standard deviation of the differences are calculated.
A first group of binary variables that control for the socioeconomic conditions of individuals in the baseline are restricted so that the matching is exact. 35The test results are focused on balancing other variables not restricted to exact matching.All results are presented for four neighbors (M = 4) although our results are robust to other numbers of neighbors. 36Also, the first row at the top shows the results before matching (step 2) while the second and third rows show the results after matching for the case in which treatment is defined as having attended a public university and private university respectively ( step 4).
In both figures, it is clear that before matching, differences between treated individuals and controls were significant for an important number of covariates (step 2).These differences were evident in the range and interquartile range and in the presence of outliers.For example, before the matching, multiple covariates differed by more than 0.5 standard deviations.After comparing only the closest neighbors, both the rank and the difference in the interquartile 35 Variables with exact matching are age (measured in quartiles), ICFES score in mathematics, language, and biology (measured in quartiles), binary variable of strata (= 1 if strata > = 3), nature and character of school, variable binary for the 13 major cities. 36 Estimates with M=1 y M=16 do not differ significantly from those presented with M = 4.
range were significantly diminished, and now there was no program in which any individual presented differences of more than 0.5 standard deviations in any variable.Finally, in the vast majority of programs, the outliers with differences of more than 0.5 standard deviations disappear.It is important to note that the balance is better when treatment is defined as having attended a private HEI.
To ensure that our estimates are robust to the matching method used, our estimates also include the results from using Propensity Score Matching (PSM) with and without bias correction (Rosenbaum and Rubin, 1983).Since the PSM estimator is the difference in the outcome variable for those included in the common support, once the treated individuals have been adequately weighted by their probability of participating, then the ATT estimator is defined only in the region of the common support.Hence it is necessary to ensure the condition of common support.parameter puts more weight on those individuals most likely to attend a private university. 37If the selection on observable characteristics differs between public and private universities, then these two parameters should not be symmetrical.In our exercise, the results for these two types of treatments are similar with the exception of a few cases.Thus, in section 5.1 the results in the case of public universities are presented, both for the total sample and the sample restricted to selective programs, accredited, or the best universities in the country.At the end of this section the focus is on the differences found when the treatment definition is attending a private university.
Before presenting our results, it is important to remember that they were obtained from the universe of individuals that took the EACES test which, as we previously showed, are largely representative of the graduates from the programs assessed.The fact that our results are not meant to take into account self-selection of individuals due to different graduation rates between public and private HEI, nor with respect to any difference arising from the length of time students took to graduate must be kept in mind.
a. Average public university effect on those who attended public university
Our parameter estimates ̂ , are obtained for each academic program (e.g.Economics).
The programs that are analyzed for both male and female include law, education, civil engineering, industrial engineering, computer engineering, management, accounting, economics, medicine and dentistry.For the specific case of men, the analysis also includes electrical engineering and mechanical engineering while in the case of women, it includes nursing and psychology (see section 2. data).For each program, five estimators are calculated using the definition of the outcome variable described in the methodology section: In general, the value added of public universities relative to private is found to be positive and significant although the effect is not observed in all programs.Likewise, the results are robust for the different matching methods, especially in the male sample.The heterogeneity in the effect on men and women is remarkable, especially when the total sample is compared to the sample that is restricted to accredited programs, targeted programs or the top 20 universities.For the total sample, public universities produce improvements in academic performance in 9 to 11 programs out of 12 considered in the case of men, and in 7 to 11 programs out of the 12 analyzed in the case of women (see figures 1 and 2).The differences lie in the matching method used.While the ̂ , estimator finds positive effects in 11 of the programs, the ̂ ,− estimator finds an effect on 9 programs in the case of men and 7 programs in the case of women.The greatest effect is observed in the civil engineering program (both women and men) with a magnitude of about 0.6 standard deviations (SD).In most programs, the effect is positive, and the magnitude of the effect is between 0.2 and 0.4 SD.In the case of men, public universities have a negative effect of around -0.38 SD on the management program while, in the case of women, the effect is negative for the psychology program with a magnitude of -0.1 SD.Likewise, the results of the different estimators have the same sign in most cases although it is clear that the bias-corrected estimators substantially correct the bias associated with the covariates in finite samples in every program.In all cases ̂ , > ̂ , , which is evidence of the importance of selection bias in the observable characteristics associated with the decision to attend a public HIE.It is interesting to note that for law and medical programs in the case of men, and law, computer engineering, management and medical programs in the case of women (in the unconditional difference between public and private universities) the sign for variable result is opposite the ̂ , estimator.This demonstrates the importance of controlling for a large number of observable characteristics.
However, the above results are sensitive to the quality of the academic program in both cases whether dealing with the sample that considers only accredited programs, the one that is restricted to targeted programs, or the one that is limited to the top 20 universities in the country.
In the case of the accredited program sample (see Figures 1 and 2), positive effects are observed in 8 to 10 of the 12 academic programs analyzed in the case of men, and in 8 to 9 of the 12 programs analyzed for women.In particular, it should be noted that the effect on medical programs is not significant for either men or women, and the effect on law programs is not robust to matching methods for either men or women.Similarly, the effect on psychology and management programs is not statistically significant in the case of women.
Selective programs
When the sample is conditioned to only selective programs, the results change significantly.In the case of men, the positive effects on all engineering (civil engineering, electronics, industrial, mechanical and computer), education, accounting, and dentistry programs are preserved.In the case of dentistry, accounting and engineering, the magnitude of the effect is reduced.In engineering programs in particular, the effect declines to half of the effect found with the total sample or less.For both the total sample and the sample restricted to accredited programs, the effect of the public university on management program is negative.Finally, the effect on the economics, law, and medical programs is no longer statistically significant.
In the case of women, positive effects on engineering (civil, industrial, computer), education, accounting, dentistry and psychology programs are observed although, in the first case, the effect is not robust for the matching methods.The positive effects of public universities are magnified in education and civil engineering programs but are significantly reduced in other cases.The exception is psychology which had a negative effect on the total sample.Unlike the total sample, the effects on law, nursing, and medical programs are not statistically significant with the ̂ , estimator, and the effect on the economics and business administration programs turned out to be negative.Negative effects are also observed in medicine and law when considering only the PSM estimator bias correction.
The 20 best universities in Colombia
When the sample considered includes only programs in the 20 best universities in Colombia and only in the case of men, negative effects are observed for the law and medicine programs.
At the same time, there is no effect on management and positive effects on the remaining programs.In the case of women, negative effects on law and positive effects on education, civil and industrial engineering, accounting, economics, and dentistry are observed.No effects on other programs are found.
In the case of men, positive and robust effects of public university on academic performance are found in accounting, computer engineering, mechanical engineering, industrial engineering, electrical engineering, and civil engineering programs.Likewise, the negative effect of public universities on the management program is robust for most of the cases considered except for the sample that includes only the best universities in the country.The effects on law, economics, and medical programs are not robust to different samples.
In the case of women, the effects are positive for most programs considered when the entire sample is included.However, the effect ceases to be positive in multiple programs when only accredited programs, selective programs, or the best universities in the country are included.There seems to be a strong positive effect on industrial and civil engineering, dentistry, and accounting programs in all samples except for the selective program sample.The effects on law, economics, medical, nursing, and psychology programs are not robust to the different samples considered in particular but are sensitive to the quality of programs considered.
b. Average Effect of private university
The differences between the people attending public universities, and those attending private universities make an evaluation of the effect of these types of institutions on their respective populations worthwhile.In the previous section, we assessed whether the people who attended public universities did better because they attended them, or if they would have done better if
VI. Conclusions
The results for the total sample that includes all programs show that public HEI allow male and female students to get better scores in 11 of the 12 programs tested.The positive effects of public universities are robust to different samples analyzed in engineering programs for both men and women.It should be noted that in the case of men, much of the effect in engineering, despite being positive, declines once only the most selective programs are considered.In addition, positive and robust effects for both men and women in the fields of education, accounting, and dentistry programs are observed.
For the sample of men, the negative effect of public universities on management is robust for all the cases considered.However, the effects on law, economics, medicine, nursing, and psychology, are not robust across the different samples considered, and in particular, they are sensitive to samples that are conditioned to program quality.In the case of the economics and psychology programs, the effect is ambiguous for the different samples analyzed.
Most of the results obtained in the case of public higher educational institutions are consistent when the treatment of interest is defined as attending a private HEI.In general, the effects are very much the opposite of those found in the case of public HIEs.However, some important differences are observed.First, when the total sample is used, private HEIs have no statistically significant effect on the psychology program for male or female students.Second, when the sample was restricted to the most selective programs in the case of men, private HEIs have a positive effect on law, and have no significant effect on electrical or industrial engineering, nor on computer engineering.Finally, when the sample was restricted to selective programs in the case of women, negative effects are observed on nursing and economics, and no effects were found on industrial engineering.
The overwhelming aggregate superiority of public HEIs versus private ones, suggests the need to promote greater regulation of the latter, review their current standards, and possibly, the accreditation status some of them hold to help bridge the gap that currently exists in terms of the value added public HEIs have in comparison to private ones.It also suggests that, at least in the short run, it could be profitable to expand the public provision of some of the higher education public programs that added more value than their private alternatives did.
It is important however, to note that no cost-effectiveness estimates were presented since only the benefit of studying at a public versus private HEI was estimated.Their relative cost was not included given that it would have required us to have information on costs for each institution which we did not have access to.
VII. References
Several studies of the Colombian labor market show evidence of significantly better performance in productivity by the skilled versus the unskilled labor force.The skilled have more access to well-paid, quality employment, which may be a reflection of different factors such as technical change, labor market polarization, or effects associated with the quality of higher education, etc.4 Melguizo et al. (2015) estimated the effect of the combination of programs and universities chosen by students on the value added of higher education.Gomez (2015) estimated the value added of higher education on academic performance for different regions of the country.These researchers used standard value added models like those used by McCaffrey et al. (2004), Kane and Staiger (2008), Rivkin et al. (2005), Chetty et al. (2014, 2014b), Cunha and Miller (2014), etc.
5)and ̂1() is a consistent estimator of 1 () = [ (1)| = ]. 34Note that the adjustment is the average difference between the predicted regression value of ̂1() for each individual treated with respect to their nearest neighbors.All our exercises are based on the proposed equation estimator (4).The robust to heteroskedasticity standard errors are estimated following the proposal ofAbadie and Imbens (2011, p.10).
Figures A. 1
Figures A.1.1 and A.1.2 in Appendix 1 illustrate this exercise for both the total sample (Figure A.1.1),which includes all programs, and for programs restricted to selective samples (Figure A.1.2).Each figure includes six boxplots which provide information on the minimum and maximum value, quartiles 1 and 3 as well as the median and outliers.Note that the left column in each figure presents the results for men, while the right column shows the results for women.
(i) the DID matching estimator, ̂ , , (ii) the OLS conditional estimator in covariates, ̂ , , (iii) the PSM estimator, ̂ , , (iv) the PSM estimator with bias correction, ̂ ,− , and (v) the difference between standardized ECAES and ICFES scores, which are standardized for individuals who studied in the same program, is calculated and used to find the unconditioned difference between public and private universities.The results are summarized in Figures1 and 2. The tables corresponding to these graphics are presented in Appendix 4. The results are presented for both men and women using all the universities considered in this exercise with their accredited programs, selective studies programs, and the top 20 universities in Colombia (see definition in section 2).All results used the case of four neighbors ( = 4) for matching procedures although the results are robust for other values of (e.g. = 16).Likewise, all the results are expressed in terms of standard deviations in order to facilitate the interpretation of the magnitudes of the coefficients in terms of the variability associated with each program.
Figure 3
Figure 3 Treatment Results: Private, men
Figure 4
Figure 4 Treatment Results: Private, women
Figure A. 1
Figure A. 1.2.Covariate balancing test for the sample restricted to selective programs Total differences: Men and Women
Una Estimación del Efecto sobre el Rendimiento Académico de Asistir a una Universidad Pública o privada en Colombia
Arlen Guarín, Sebastián Londoño, Carlos Medina, Julieth Parra, Christian Posso * y Carlos Eduardo Vélez † The first filter considers only individuals who were attending accredited programs at the moment they took the EACES.The second filter considers only individuals in programs limited to students who scored in the 25th percentile of the ICFES test -a score that was better than the average for all those taking that test.This allowed us to eliminate programs that admitted students with poor results on the ICFES tests and retain only those in selective programs.The third is restricted to programs that belong to the top 20 universities in Colombia based on the QS University Rankings Latin America, a ranking based on criteria that is slightly and indirectly associated with the ICFES test.
23Given the low comparability between the different subjects assessed in the ECAES tests for different programs, we measure value added by program.To ensure a minimum robustness in our results, only those programs that had a minimum of 100 students of either gender per program were analyzed in each group of public and private universities.As a result, for both men and women, we analyzed academic programs in medicine, dentistry, civil engineering, industrial engineering, computer engineering, law, education, management, accounting and economics.Nursing and psychology were analyzed only for women, and electrical and mechanical engineering programs were analyzed only for men.Overall, our exercise focused on and analyzed a total of 14 programs of which 12 academic programs included both genders.
Table 3 .
Table in Appendix 1 with the definition of the different variables).ICFES Test Results for mathematics and language by program for men in accredited and non-accredited programs at public and private
Table 4 .
ICFES Test Results for mathematics and language by program for women in accredited and non-accredited programs at public and
Table 5
illustrates, by gender, the most important features of the data for selected programs.Except for programs related to engineering, the selected programs are mainly composed of women.Additionally, public funding is particularly high for the education program.In the remaining programs, there are more students in the private institutions.There
Table 5 .
Individuals by program, for each gender and city
Table 6 .
Public and private university attendance by department (state), 2005 Note: The percentage of university students per capita is estimated based on the population of those 16-24 years of age.Source: Population Census 2005.Authors' calculations.
(Caliendo and Kopeining, 2005)sents the common support for the different exercises and programs considered in the exercise.It is important to emphasize that in most of the exercises, the common support includes much of the unit interval, which means that almost any combination of features observed in the treated group is also observed in the control group(Caliendo and Kopeining, 2005).
|
2019-04-26T14:25:05.527Z
|
2016-10-01T00:00:00.000
|
{
"year": 2016,
"sha1": "212cf48d9ad1f5fe784d3a2e07018f4605f815cc",
"oa_license": "CCBYNCSA",
"oa_url": "http://repositorio.banrep.gov.co/bitstream/20.500.12134/6279/1/be_968.pdf",
"oa_status": "HYBRID",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "e4ae6ed6a49258b39953cfc55cff5abc19cd4f76",
"s2fieldsofstudy": [],
"extfieldsofstudy": [
"Geography",
"Political Science"
]
}
|
267219388
|
pes2o/s2orc
|
v3-fos-license
|
Antimicrobial and Synergistic Effects of Syzygium cumini, Moringa oleifera, and Tinospora cordifolia Against Different Candida Infections
Introduction The burden of multiple drug resistance in human pathogens has necessitated the search for and development of antimicrobial agents with a wide range of structural classes and potentials to selectively act on the several mechanisms of actions exhibited by the pathogens. However, most synthetic antimicrobial agents have been linked with adverse side effects and high costs, furthering the need to explore more options. Syzygium cumini, Moringa oleifera, and Tinospora cordifolia are three medicinal plants used in traditional medicine systems for various infectious diseases. They contain various phytochemicals that exhibit antimicrobial activities against various bacteria, fungi, and parasites. The mechanisms of their antimicrobial action may involve the disruption of microbial cell walls and membranes, the inhibition of microbial enzyme and biofilm formation, the modulation of microbial gene expression and quorum sensing, and the induction of microbial cell death. Therefore, the present study evaluated the potentials of aqueous and ethanol extracts of S. cumini, M. oleifera, and T. cordifolia in managing infections as measured by their inhibitory effects on species. Materials and method Syzygium cumini, M. oleifera, and T. cordifolia were obtained and authenticated, and their aqueous and ethanol extracts were prepared. The antibacterial properties of the aqueous and ethanol extracts were examined. In addition to broth microdilution and biofilm development experiments, we also employed disk diffusion and agar-well diffusion techniques. The inocula of various species, including krusei, parapsilosis, utilis, albicans, and glabrata, were prepared for these assays. The synergistic effect of plant extracts with fluconazole was also evaluated. Results Syzygium cumini, M. oleifera, and T. cordifolia emerge as promising sources for the development of effective and sustainable antimicrobial interventions. Interestingly, the aqueous and ethanol extracts were effective against the selected species. Also, the synergistic combination of plant extracts with fluconazole was observed to triple the potency of the extracts. Furthermore, the potency of the plant extract as an antifungal and synergistic agent was ranked as S. cumini > M. oleifera > T. cordifolia. Conclusively, the plant extracts are effective in the management of opportunistic fungal infections.
Introduction
Opportunistic fungal pathogens are major causes of infection in immunocompromised individuals such as surgical patients or critically ill patients [1].Of these opportunistic fungal pathogens, the most common are Candida species, which are the primary sources of systemic and mucosal infection in humans [2].Although this fungus occurs as a normal flora in animals and humans, it may become opportunistic and produce lethal and disabling infections [3].Candida infection is characterized by high death rates and exorbitant healthcare costs for patients and governments [4].The prevalent mortality rate as a result of Candida infection has been attributed to the ever-growing rate of invasive systemic infections as well as septicemia cases, particularly in immunocompromised individuals [5].Interestingly, 90% of invasive infections have been linked to different species of opportunistic Candida, making it the fourth principal cause of nosocomial bloodstream infections [6].
Analytically, the last decade has recorded as high as 20% of Candida infections attributed to these non-albicans Candida species [7].Of the non-albicans Candida species, Candida glabrata has the highest frequency of infection [8].It is not only the least sensitive Candida species to antifungal drugs but also the most difficult to eliminate.Conversely, Candida parapsilosis is mostly considered the least virulent Candida species, and it is capable of frequently causing candidemia in health workers with poor hand hygiene [9].Candida krusei is prevalent in individuals who are diagnosed with Down syndrome [10].
Resistance of Candida species to various antifungal drugs has resulted in immunosuppressed patients being subjected to extensive stays in intensive care units, surgery, and broad-spectrum antibiotic treatments.All of these have been observed to further elevate the possibility of disseminating candidiasis [4].Regrettably, multidrug resistance (MDR) in human pathogenic microorganisms has developed due to the indiscriminate use of commercial antimicrobial drugs commonly used in treating infectious diseases.This situation has forced scientists to search for new antimicrobial substances from various sources as novel antimicrobial chemotherapeutic agents.
Having considered various possible approaches, exploring medicinal plants containing diverse bioactive compounds that have been employed for antifungal activity in folklore could be more productive in treating and managing candidiasis.This is because natural products from these medicinal plants have been explored and observed to exhibit unlimited opportunities in the treatment of candidiasis owing to their bioavailability of chemical diversity [11].Syzygium cumini, Moringa oleifera, and Tinospora cordifolia are three medicinal plants rooted in traditional medicine that exhibit broad-spectrum antimicrobial activity against bacteria, fungi, and viruses [11].The bioactive compounds within each plant, including ellagic acid, quercetin, and tannins in S. cumini, isothiocyanates and chlorogenic acid in M. oleifera, and alkaloids and glycosides in T. cordifolia, contribute to their antimicrobial efficacy.Abdelgadir et al. reported that S. cumini extracts possess in vitro antibacterial activities and could be used not only in traditional medicine.He suggested that further research work should be carried out on this plant to determine the toxicity as well as the optimum dose.The result should then be compared with standard antibiotics [12].
The mechanisms of their antimicrobial action may involve the disruption of microbial cell walls and membranes, the inhibition of microbial enzyme and biofilm formation, the modulation of microbial gene expression and quorum sensing, and the induction of microbial cell death [11].Beyond their direct antimicrobial effects, these botanicals hold promise for therapeutic applications in infectious diseases, wound healing, and immune support.Further research is warranted to elucidate specific molecular mechanisms, optimize formulations, and validate their clinical efficacy.Therefore, this study seeks to comparatively evaluate edible and safe-for-consumption plants, such as S. cumini, M. oleifera, and T. cordifolia, in the management of Candida infections, as measured by their inhibitory effects on Candida species.
Sample collection and preparation
Fresh leaves of S. cumini, M. oleifera, and T. cordifolia were obtained from local farmers in Akure South Local Government, Ondo State, Nigeria.The leaves were authenticated with voucher number (0347) obtained at the Centre for Research and Development (CERAD), Federal University of Technology, Akure, Nigeria.The leaves were then sorted out of debris, washed, air-dried, and pulverized into fine textured powder using Warring Commercial Heavy-Duty Blender (Model 37BL18; 24ØCB6) to ensure better surface contact with extraction solvents.The pulverized samples were then prepared according to modified methods [13].Typically, 20 g of the powdered samples were weighed with a weighing balance (Sartorius BP61S) and dissolved in 100 mL of distilled water (for aqueous extract) and 70% ethanol (for ethanol extract).This was macerated for 24 hours and filtered into a conical flask with Whatman's No. 1 filter paper to get a clear filtrate.Subsequently, each extract was shaken on a vortex mixer to facilitate dilution.The filtrate was kept in the refrigerator at 4°C for further use.This was repeated every week to get a fresh filtrate.The concentrated extracts were further diluted by mixing with appropriate phosphate buffer saline (PBS) and dimethyl sulfoxide (DMSO).Stock solutions of aqueous/ethanolic extracts using the appropriate aseptic technique were further diluted by dissolving 100 μL of the aqueous/ethanolic extracts in 200 μL of PBS/5% of DMSO to get a 1:2 dilution and 100 μL of the aqueous/ethanolic extracts in 400 μL of PBS/5% of DMSO to get a 1:4 dilution, respectively.
Standard drug preparation
Fluconazole, used as a positive control, was dissolved in 5% DMSO to give a stock concentration of 10 mg/mL.Fluconazole, a well-established antifungal drug, is often used as a positive control in antimicrobial assays at varying concentrations than those used for the plant extracts being studied.This is a result of certain reasons that include its known efficacy, which negates the need for higher concentrations such as those used for experimental extracts.It serves as a benchmark for making comparisons and needs to be distinctly identifiable in its effects.Fluconazole's potential toxicity at higher concentrations also necessitates its use at lower concentrations.Moreover, differences in solubility and modes of action between fluconazole and plant extracts may require disparate concentrations to effectively demonstrate their respective antimicrobial effects.
Inoculum preparation
Reference strains of Candida (C.krusei American Type Culture Collection [ATCC] 14243, C. parapsilosis ATCC 22019, C. utilis and C. albicans ATCC 76615, and C. glabrata ATCC 66032) were obtained from Coventry University Super Laboratory.These were then grown in Sabouraud (SAB) dextrose agar for 18 hours at 37°C as previously described by Vasanthi et al. [14].The handling and use of the Candida were approved by the CERAD Ethical Committee of the Federal University of Technology with the ethical number (FUTA/ETH/2020/016).The C. krusei, C. parapsilosis, C. utilis, C. albicans, and C. glabrata isolates were maintained in the SAB dextrose (Thermo Scientific™ Oxoid) agar all throughout the study duration.Each of the Candida species (C.krusei, C. parapsilosis, C. utilis, C. albicans, and C. glabrata) from their stock cultures was streaked on agar plates and then incubated at 37°C for 24 hours.Two yeast colonies from each of the agar plates were emulsified in sterile SAB broth (Oxoid, CM0147) as inocula [15].It was incubated at 37°C for 24 hours, and the resulting solution was vortexed for 10 seconds.Afterward, the turbidity was adjusted with a spectrophotometer (Biochrom Ltd, Cambridge, England) to adjust it to 0.5 McFarland standard at 530 nm wavelength.Exactly 0.5 McFarland gives an equivalent approximate density of yeast cells 1.5 × 10 8 CFU mL − 1 [16].
Disk Diffusion Method
The evaluation of S. cumini, M. oleifera, and T. cordifolia activities was carried out on C. krusei, C. parapsilosis, C. utilis, C. albicans, and C. glabrata using the disk diffusion method as reported by Okla et al. [15].SAB dextrose agar (SDA) was poured into the Petri dishes, and each of the Petri dishes was then loaded with 100 µL of the inocula (C.krusei, C. parapsilosis, C. utilis, C. albicans, and C. glabrata).Typically, these 6 mm diameter sterile filter paper disks were obtained and inserted into Petri dishes containing 20 µL of plant extracts.The different concentrations (100%, 50%, and 25%) of S. cumini, M. oleifera, and T. cordifolia were loaded via a 0.22 mm Millipore filter over the filter paper disks.Distilled water and ethanol only (without plant extracts) served as the negative controls, while fluconazole (10 mg/mL) was the positive control.DMSO was used as the solvent for the ethanol extracts, and PBS was used as the solvent for the aqueous extracts.All the Petri dishes used in the disk diffusion method were maintained in a 5°C environment for 2 hours during the diffusion of plant extract and then incubated for 24 hours at 37°C anaerobically.The incubation results on each Petri dish were documented after the zone of inhibition had been measured.Each test was carried out in triplicate, and the average values were recorded.
Agar-Well Diffusion Method
As employed by Hirsch et al. [17] with a slight modification, inoculum susceptibility tests were carried out on fresh agar plate cultures provided using a sterile swab stick to spread the yeast culture throughout the plates and allowed to dry.The agar plate was divided into sections and labeled accordingly.Five wells were made using a sterile cork borer (6 mm in diameter) into agar plates containing inocula and were filled with 100 µL of plant extract of 100%, 50%, and 25% concentration separately; 50 µL of three concentrations (100%, 50%, and 25%) of S. cumini, M. oleifera, and T. cordifolia were added to the wells.It was allowed to stay for 1 hour at room temperature before incubating.Fluconazole (10 mg/mL) served as a positive control in another well.With the lids uppermost, the plates were incubated for 24 hours at 37°C.Inhibition zones of the extracts against the fungi were measured in millimeters (mm) and interpreted by comparing them against the Clinical and Laboratory Standards Institute (CLSI).The values were presented means of three independent experiments.
Broth Microdilution Assay
Microdilution was performed as recommended by the CLSI [18].Therefore, the minimum inhibitory concentration (MIC) of S. cumini, M. oleifera, and T. cordifolia plant extracts that exhibited strong antifungal activity against selected species of Candida (C.krusei, C. parapsilosis, C. utilis, C. albicans, and C. glabrata) was investigated.Various concentrations of the stock, 0.78 to 200 mg/mL, were assayed against the test fungi.For the broth microdilution test, 100 µL of each fungal suspension in the suitable growth medium was added to the wells of a sterile 96-well microtiter plate already containing 100 µL of two-fold serially diluted plant extract in Mueller Hinton agar.Column 1 was a control to monitor sterility, while columns 2-11 contained 100 µL each of inoculated broth and plant extracts.A multichannel pipette was then used to transfer and mix biosurfactants from columns 2-11, resulting in 100 µL per well, and 100 µL taken from the last well was discarded.The final volume in each well was 100 µL.Control wells were prepared with culture medium, fresh broth suspension, and plant extracts only in amounts corresponding to the highest quantity present.The plates were incubated at 37°C for 48 hours.All measurements of MIC values were repeated in triplicate.The MIC was determined by checking for turbidity using a spectrophotometer [19].The inhibitory potentials of the undiluted concentrations of the plant extracts against the biofilm formation of Candida species were evaluated using the resazurin reduction method as described by Gómez-Casanova et al. [20]; 0.015g of resazurin powder (R7017-1G Sigma-Aldrich, Inc, St. Louis, MO, USA) was dissolved in 100 mL of diluted PBS X10 to get a final concentration of 0.015%.It was sterilized by filtration (Whatman's filter paper) and stored at 4°C until use.The preparation procedures were performed in the dark, and the resazurin solution was then kept in a foil-wrapped bottle to prevent exposure to light due to its sensitivity to light.The inoculum was prepared per the CLSI recommendation.Under aseptic conditions, with a 96-well plate, 100 µL of SAB broth and plant extract were dispensed in each well of column 1 as a control to monitor sterility, while columns 2-11 contained 100 µL each of inoculated broth and plant extracts.A multichannel pipette was then used to transfer and mix biosurfactants from columns 2 to 11, resulting in 100 µL per well, and 100 µL taken from the last well was discarded.Column 12 was used as a positive control with inoculum only and incubated for 48 hours at 37°C.According to the protocol used by Elshikh et al. [21], 0.015% resazurin was added to all wells (30 µL per well) and further incubated for 3 to 4 hours to observe color change.The reduction of resazurin, therefore, correlates with the number of live cells.The transference of electrons from NADPH to resazurin is expected to reduce the blue resazurin to a pink, fluorescent counterpart, resorufin.On completion of the incubation, columns were read spectrophotometrically at 595 nm.
Synergistic effect of plant extracts with antibiotics
Synergistic antifungal activity was measured using a good diffusion method according to the National Committee for Clinical Laboratory Standards [22,23].Under aseptic conditions, Petri dishes containing approximately 20-25 mL of SDA medium were inoculated using a cotton swab.The plants, separately powdered, were extracted by adding 24 hours of old culture of the fungal strains.Wells (6 mm diameter) were punched in the agar and filled with 30 µL of plant extracts or antibiotics, but in case of synergism effect, 30 µL of each extract and antibiotic was added into the wells.Replicates of each plate were done.The plates were incubated at 37°C for 24 hours.The antifungal activity was assessed by measuring the inhibition zone diameter (mm) around the well.The average of three replicates for each extract, antibiotic, and combination was calculated.The synergism effect was considered when combinations were exhibited with fluconazole (10 mg/mL).
Statistical analysis
The results from the present study were collected and graphed for statistical analysis using Microsoft Excel and Statistical Package for Social Science (SPSS) Software Version 26 (IBM Corp., Armonk, NY, USA).The effects of the different extractants at the different selected dilution concentrations (100%, 50%, and 25%) were evaluated against the Candida species.Liquid dilution, as well as biofilm formation, was evaluated via univariate one-way analysis of variance (ANOVA), and the significance differences were obtained at p < 0.05.The inhibition zones were calculated as means ± SD of three replicates.The whole procedure was repeated thrice to ascertain the validity of the obtained results.
Disc diffusion assay
The aqueous and ethanol extracts of S. cumini proved to be effective against all Candida species tested, while these Candida species did not show any effect toward T. cordifolia and M. oleifera even with the highest concentrations.The results showed maximum activity against C. parapsilosis at 100% concentrations of water and ethanol by S. cumini with the mean zone of inhibition (mm) measuring 15.00 ± 1.00 and 16.33 ± 0.57, respectively, which was lower than the control value (Table 2).However, the inhibition zones of S. cumini, when treated with 50% and 25% concentrations on the same isolate, were 10.33 ± 0.57 and 9.33 ± 0.57 for water and 13.00 ± 1.15 and 10.33 ± 0.57 for ethanol, respectively.Nevertheless, T. cordifolia and M. oleifera showed no activity against C. parapsilosis.Only S. cumini exhibited an inhibitory effect on C. albicans with a mean zone of inhibition (mm) measuring 11.00 ± 1.00, 9.00 ± 0.00, and 9.00 ± 1.00 for water extracts, while water extracts show 15.00 ± 1.00, 12.00 ± 1.00, and 8.00 ± 1.00 for 100%, 50%, and 25% concentrations, respectively.Not all the isolates used in the study were inhibited.However, the antifungal activity varied with species.Candida krusei and C. parapsilosis were the most inhibited species, and C. albicans was the least inhibited species.Aqueous extracts were a bit less active than the ethanolic extracts in all concentrations.The zone of inhibition recorded at 100% concentration was higher than that of 50% and 25% concentration for all the extracts.As the amount of the extract increased, the inhibitory effect also increased.Fluconazole (10 mg/mL), used as the positive control, produced a zone of inhibition of 10.00 ± 0.57, 23.00 ± 1.00, and 15.00 ± 0.00 only on C. glabrata, C. parapsilosis, and C. krusei, respectively.However, there was no significant activity against other Candida species; 5% of DMSO and PBS negative control was used and did not have any inhibitory effect on any of the five isolates of Candida species tested, demonstrating a total lack of antimicrobial activity.The results showed that both aqueous and ethanol extracts of S. cumini were found to be more effective against all the fungi tested (p < 0.05) compared to other extracts.
TABLE 2: Mean zones of inhibition of plant extracts against Candida species
Note: The mean value (n = 3) is presented as mean ± SD.The one-way ANOVA was used to determine the average zones of inhibition, which were reported as mean ± SD.
Key: The negative control is 5% DMSO for ethanol and PBS for water extracts; the positive control used was fluconazole.
NI, no inhibition; Conc, concentration
In summary, the analysis shows a significant difference in the MIC among the groups being studied, as shown in Figure 2. Given the p-value of 0.000, the results are highly significant at the 0.05 level.The use of fluconazole as a positive control provides a reference point for the comparison.
Broth microdilution assay
The effectiveness of the plant extracts in the Candida strains was confirmed by measuring the MICs.MICs of plant extracts for the organisms were determined using the broth dilution method.Results indicated different MIC levels based on the fungal strains being tested.MICs of aqueous and ethanol extracts ranged from 0 to 100 mg/mL.MICs of aqueous and ethanol extracts ranged from 0 to 100 mg/mL.MIC values ranged from 0.15 to 100 mg/mL for C. glabrata, 0 to 50 mg/mL for C. utilis, and 0 to 100 mg/mL for C. albicans, C. parapsilosis, and C. krusei.
In ethanolic extracts of the three plants, MIC ranged from 0.39 to 100 mg/mL, and the range for aqueous extracts was the same.Syzygium cumini recorded the lowest MICs for all tested fungal strains with aqueous and ethanolic extracts.For ethanolic extracts of S. cumini, the MICs were 3.
TABLE 3: MIC values of aqueous and ethanol extracts (mg/mL) of T. cordifolia, M. oleifera, and S. cumini against Candida species
Note: There was a two-fold serial dilution with 100 mg/mL at initial concentration.Fluconazole was used as a positive control.
0, no inhibition activity
The presented tables offer insights into the antifungal activities of three plants -T.cordifolia, M. oleifera, and S. cumini -against several Candida species.The MIC values highlight the effectiveness of the plant extracts in inhibiting the growth of Candida species.A lower MIC value indicates a stronger antifungal effect.Ethanol extracts generally seem more effective than water extracts, especially in the case of T. cordifolia against C. albicans and C. parapsilosis.Interestingly, the water extract of M. oleifera shows no inhibitory activity against C. albicans and C. krusei, whereas its ethanol counterpart demonstrates significant activity.
Fluconazole, a standard antifungal drug, is included as a positive control, and its varying MIC values against different Candida species offer a benchmark for comparison.For instance, S. cumini's ethanol extract displays notable potency against C. glabrata with an MIC of 0.15 mg/mL, which is notably lower than that of fluconazole for the same species.
The subsequent ANOVA analysis underscores a statistically significant difference in MIC values among the groups studied, as shown in Figure 3.This supports the notion that the effectiveness of these plant extractsand their method of extraction -varies when it comes to inhibiting different Candida species.The statistical significance emphasizes the potential therapeutic promise and warrants further investigation into the active constituents responsible for the antifungal activities of these plants.
Resazurin-based turbidometric assay
In the resazurin-based turbidometric (RBTB) assay, all sterility control wells for all tested fungi remained blue in color after an overnight incubation followed by 4 hours of incubation with resazurin.In contrast, all wells in the growth test columns (containing growth medium and fungi) of all tested fungi changed color from blue to pink or from blue to pale pink.However, there was no consistency in the growth.There was an inconsistent increase and decrease in all samples.The growth percentage was calculated using univariate ANOVA for each Candida species based on the extracts.
Synergistic effect
The synergistic effect of plants and fluconazole was evaluated using an agar-well diffusion assay.The results of the synergistic activity of extracts with antibiotics were determined by the diameters of inhibition zones, which are presented in Figure 4.All combinations of fluconazole with the plant extracts (water and ethanol) showed a synergistic effect on all fungal strains used.Of all plants tested, C. glabrata showed the greatest susceptibility with a mean zone of inhibition (mm) of 36.00 ± 0.00 and 45.00 ± 0.00 for T. cordifolia and fluconazole, respectively, 39.00 ± 0.00 and 43.00 ± 0.00 for M. oleifera and fluconazole, respectively, and 38.00 ± 0.00 and 42.00 ± 0.00 for S. cumini and fluconazole for water and ethanol extracts, respectively.In water extraction, the combinations of T. cordifolia and fluconazole had the highest combination against C. albicans, with a mean zone of inhibition (mm) of 33.33 ± 0.57.However, in ethanol extraction against C. albicans, M. oleifera and fluconazole combination showed more synergy effect with a mean zone of inhibition (mm) of 37.33 ± 0.57.The ethanolic extracts showed a more synergistic effect with fluconazole against fungal strains than the aqueous extracts, although a lesser effect was shown in the ethanolic extracts against C. krusei for the three plants.Among the five yeasts used in this research, the most resistant was C. utilis.Conversely, C. parapsilosis and C. krusei were the most sensitive strains to the tested extracts, as shown in Table 4.There was no significant difference in the zones of inhibition (p < 0.05 ).However, all combinations of extracts with fluconazole were found to be fungicidal against the five tested fungi compared to extracts alone and fluconazole alone, which had lesser zones of inhibition.This shows that the extracts had the ability to increase the action of fluconazole in these strains through synergism.Moreover, there is no significant difference in the mean zone of inhibition of the plants (Table 4).
Combinations
Fungal strain mean zone of inhibition The mean values are expressed as mean ± SD (n = 3).Inhibition zones were measured in mm.Zones of inhibition include 6 mm for disk diameter.
Discussion
The ability of medicinal plants to act as antifungal agents is attributed to their phytochemical constituents, such as saponins, alkaloids, cyanogenetic glycosides, steroids, terpenoids, and flavonoids [15].Alkaloids are therapeutically useful due to their antispasmodic, bacterial, and analgesic effects [13].Flavonoid is a hydroxylated phenolic substance that is commonly synthesized by plants in response to microbial infection and, therefore, has been evaluated to elucidate potent antimicrobial activities against a wide range of microorganisms.Its therapeutic activity has been linked to its ability to form complexes with soluble and extracellular proteins as well as bacterial cell walls [23].Saponins are established antimicrobial agents as they can mediate the leakage of proteins and certain enzymes from invading pathogenic cells [24].Besides, steroids possess antibacterial attributes as there is a relationship between membrane lipids and their sensitivity to steroidal compounds, which establishes the proposed mechanism of steroids' antibacterial action via their association with membrane lipids to cause leakages from liposomes [25].Interestingly, some bioactive phytochemicals are bioavailable in S. cumini, M. oleifera, and T. cordifolia plants [25][26][27].
The agar-well diffusion method has been widely employed to determine the antifungal activity of plant extracts.The aqueous extract of T. cordifolia did not show any activity against all five target fungal species for all concentrations used.Inferentially, the aqueous extracted phytochemical constituents of T. cordifolia lack sufficient or appropriate bioactive antifungal constituents that can be evaluated using the agar-well diffusion method [28].Meanwhile, the aqueous extracts of M. oleifera had mean inhibition zones only at the highest percentage concentration (100%), which indicates that the plant's aqueous extracts might possess limited bioactive antifungal constituents at lower concentrations.Also, the ethanolic extracts of T. cordifolia and M. oleifera could only inhibit some strains of the Candida species.However, S. cumini significantly inhibited all the Candida species.This trend of the result may indicate that despite the effectiveness of polar solvent extracts in antimicrobial activities, only S. cumini significantly showed a mean zone of inhibition across all investigated Candida strains.
Additionally, a significant reduction in fungal growth in terms of the zone of inhibition around the disk during the disk diffusion assay was observed.Typically, only the aqueous and ethanol extracts of S. cumini showed concentration-dependent effectiveness against all Candida species tested.This observation agrees with earlier reports that documented the antibacterial activities of S. cumini and implicated its high bioactive phytochemical components for its antimicrobial activities [25].However, the chemical properties of extraction solvents play decisive roles in demonstrating plants' antifungal properties, as the presence of the phytochemicals extracted by the solvents may relate to the diverse antifungal activities against the Candida strains [15].Syzygium cumini ethanolic extracts were more effective against Candida strains than aqueous ones, suggesting that ethanol may extract more antifungal ingredients.However, T. cordifolia and M. oleifera extracts showed no effect, implying Candida resistance or insufficient bioactive compounds in these plants.
During the RBTB assay, all sterility control wells for all tested bacteria remained blue in color after overnight incubation, followed by 4 hours of incubation with resazurin.This suggests that the antifungal effects of the plant extracts on the Candida strains may not be appropriately evaluated using the RBTB assay.This further clarifies why the RBTB assay was unable to evaluate the antifungal effects of the plant extracts on the Candida species.
In addition, all combinations of fluconazole with the plant extracts (water and ethanol) showed a synergistic effect on all Candida species.The combination with fluconazole led to around three times the inhibitory effect of the plant extracts against the Candida species.This result substantiates the theory that the combination of antifungal agents against Candida species is more effective [29].Candida glabrata showed the greatest synergistic activity, followed by a combination with T. cordifolia.Also, ethanolic extracts showed a more synergistic effect with fluconazole against fungal strains than aqueous extracts.This correlates with a previous study that implicated the high hydrophobicity ability of the ethanolic extracts in increased membrane fluidizing effects that disrupt short-range interactions and facilitate membrane leakage as the cause of the increased synergistic effects [23].The different resistance patterns of the Candida species also validate previous studies that established that each strain of the Candida species.employs a different resistance mechanism [29].
The great potency of S. cumini extract, particularly its ethanolic extract, can be attributed to several factors that collectively contribute to its superior antifungal activity compared to T. cordifolia and M. oleifera extracts.This is in tandem with a study conducted by Khan et al. [30].These factors include the concentration of bioactive compounds, such as phenolic compounds, flavonoids, tannins, and essential oils, in higher concentrations than the other extracts.Another factor is the choice of solvent used in the extraction of the bioactive compounds.Ethanol is often more effective in extracting a wide range of phytochemicals, including those with antifungal properties.For this reason, the ethanol extract of S. cumini may contain a higher concentration of bioactive compounds compared to those of T. cordifolia and M. oleifera.Selective antifungal activity can also be a factor responsible for variations shown in the efficacy of S. cumini compared with other plant extracts.Different plant extracts may exhibit varying degrees of specificity against different fungal species [22].In the case of S. cumini, its extract shows remarkable efficacy against C. glabrata, as evidenced by the notably low MIC values.This selectivity may be attributed to specific bioactive compounds within the extract that interact more strongly with the cellular structures or mechanisms unique to C. glabrata.
This study had a few limitations despite providing valuable insight into the in vitro clinical relevance of the plant extracts and their potential therapeutic applications.It does not address the effects, as well as safety, on human subjects; hence, generalizability cannot be assumed in the human population.Therefore, further research should extend beyond in vitro assessment.Such investigations should include preclinical and clinical trials to validate and establish the safety of these plant extracts for use as therapeutic agents by humans.This multifaceted approach is essential to building a strong basis for the possible use of the extracts as medicinal agents.
Conclusions
Syzygium cumini, M. oleifera, and T. cordifolia emerge as promising sources for the development of effective and sustainable antimicrobial interventions.Notably, S. cumini stands out as a particularly robust candidate, demonstrating superior antifungal properties against all tested Candida species as compared to the other plant extracts.The effectiveness of both aqueous and ethanolic extracts of these plants against the selected Candida strains suggests a great inhibitory potential possessed by the plant extracts.Considering factors such as production cost, bioavailability, and minimal side effects even at higher dosages, plant extracts emerge as compelling alternatives to expensive synthetic drugs, which are notorious for their adverse effects.Also, S. cumini ranked the most active antifungal and synergistic agent, followed by M. oleifera, an indication of their potential in the management of opportunistic fungal infections.However, studies that encompass in vivo assays that would further establish the antifungal potentials of the aqueous and ethanolic extracts are recommended.More human clinical trials and safety evaluations are also needed to further validate their efficacy and establish their optimal dosage and mode of administration.
TABLE 1 : Mean zones of inhibition of plant extracts against Candida species
NI, no inhibition; Conc, concentration.
|
2024-01-26T16:44:26.783Z
|
2024-01-01T00:00:00.000
|
{
"year": 2024,
"sha1": "695f251f6436209e7ccc9a69678c28c99cad750a",
"oa_license": "CCBY",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "d279ea449e8422b0e7fbc612dbd3755b7152fc31",
"s2fieldsofstudy": [
"Medicine",
"Biology",
"Environmental Science"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
239472317
|
pes2o/s2orc
|
v3-fos-license
|
Improved Lipophilicity and Aqueous Solubility Prediction with Composite Graph Neural Networks
The accurate prediction of molecular properties, such as lipophilicity and aqueous solubility, are of great importance and pose challenges in several stages of the drug discovery pipeline. Machine learning methods, such as graph-based neural networks (GNNs), have shown exceptionally good performance in predicting these properties. In this work, we introduce a novel GNN architecture, called directed edge graph isomorphism network (D-GIN). It is composed of two distinct sub-architectures (D-MPNN, GIN) and achieves an improvement in accuracy over its sub-architectures employing various learning, and featurization strategies. We argue that combining models with different key aspects help make graph neural networks deeper and simultaneously increase their predictive power. Furthermore, we address current limitations in assessment of deep-learning models, namely, comparison of single training run performance metrics, and offer a more robust solution.
Introduction
Oral bio-availability, drug uptake, and ADME-related properties of small molecules are key properties in pharmacokinetics. For drugs to reach their intended target, they need to pass through several barriers either by passive diffusion or carrier-mediated uptake typically mediated by lipophilicity and aqueous solubility. Compounds with poor solubility are unable to achieve that and, therefore, pose a higher risk in attrition and overall cost during development [1].
Methods based on deep-learning have proven successful in predicting molecular properties [2] and are becoming more and more a routine part of the modern computer-aided drug design toolbox for molecular design and med-chem decision support. Since molecules can be represented as graphs, an obvious approach is to employ a graph-based architecture for deep-learning, which leads to the utilization of graph-based neural networks (GNNs). These kinds of networks are capable of learning representations for a specific task in an automated way and, therefore, can eliminate the complicated feature engineering process where domain specialists have to select the list of descriptors themselves [3]. They became increasingly popular in the last few years [4][5][6] especially due to their success in chemical property prediction [5,[7][8][9][10][11].
One of the first GNN models used for physicochemical property prediction was introduced by Micheli [12] in 2009. It predicted the boiling point of alkanes with a recursive architecture for structured data input and achieved an improved state-of-the-art performance. Lusci et al. [13] were the first to apply an undirected cyclic graph recurrent neural network on predicting aqueous solubility successfully. In the following years, several recurrent, spatial, and spectral graph-based neural networks were introduced [3, [14][15][16]. One of them was the message passing framework, which was extended to include directed edges [3]. This network, called directed-edge message passing network (D-MPNN), is one of the most successful GNNs to predict chemical properties [1].
Despite the success, one important limitation with message passing networks is the graph isomorphism problem, meaning that they are unaware of the structural role of each node or edge [17]. Most standard GNNs, such as the D-MPNN, are incapable of distinguishing between different types of graph structures to determine whether they are topologically identical [18]. Compounds, such as naphthalene and 1,1-bi(cyclopentane), are perceived as the same structure by these networks. This can be problematic because they have vastly different chemical properties. To address this issue, graph isomophism networks (GIN), another group of GNNs, have recently received attention [18,18,19]. They are capable of distinguishing between these compounds by reformulating the message passing framework to incorporate the Weisfeiler-Lehman (WL) hierarchy. They try to be at least as expressive as the Weisfeiler-Lehman graph isomorphism test (WL-test) [20] and have shown good results in chemical property prediction [18,19] despite often falling short with respect to speed and accuracy to other frameworks, such as the D-MPNN [21]. Inspired by the key property of the GIN and the success of the D-MPNN framework, we combined the key characteristics of both architectures. By doing so we not only address the isomorphism problem but also incorporate one of the most successful and powerful GNN frameworks to improve lipophilicity and aqueous solubility prediction.
When comparing new machine learning architectures with previously published methods, the standard approach is to compare single performance metrics, such as root mean squared error (RMSE) values on a test set to show model performance [21,22]. This can lead to reproducibility issues as stochastic algorithms like neural networks can vary greatly in their prediction, even without changing their hyperparameters, simply by using different training, validation, test set splits or non-deterministic weight initializations [23,24]. One of the reasons for this is the complex landscape that optimizers have to navigate through in modern machine learning models. In real world applications these landscapes can have multiple local minima and it is especially hard for non-deterministic optimization algorithms like stochastic gradient descent to find the global minimum, therefore often retrieving different results when repeated [25]. This problem can be intensified by using small datasets with different random splits for training and evaluation. Such an approach can lead the optimization algorithm into different local minima and makes it almost impossible for the model to generalize [2]. It is, therefore, difficult to compare different deep-learning model architectures with each other even when using the same data [23]. Another challenge is especially prominent in the GNN domain, where the optimal features for node or edge representation are unknown. Deep-learning benchmark studies often use the same data but different representations for their input data which makes it difficult to make a fair comparison between the models [2,3].
To mitigate these problems, we use the exact same data split to train, evaluate, and test each of the used models with different node and edge features, as well as learning strategies to obtain an average performance independent of the used features and training approaches. Such a procedure is time consuming as multiple models have to be evaluated several times. Nevertheless, obtaining a better overview of the behaviour of GNNs under these different constraints will facilitate the understanding of these architectures and ultimately help advance GNNs beyond the current hype to more explainable and robust models.
Our contribution is a novel graph neural network architecture called directed edge graph isomorphism network (D-GIN). It extends the directed edge message passing (D-MPNN) framework [1] by the graph isomorphism network (GIN) [18]. An overview of the D-GIN model is shown in Figure 1. Our novel architecture shows improved performance compared to its individual, less complex networks, and we demonstrate that combining models with different key aspects help make graph neural networks deeper while simultaneously increasing their predictive power. We evaluated our models by applying different learning and featurization strategies and compared their average performance under different constraints. (4) iteratively updated by an additional trainable identifier (epsilon). (5) Hidden node features are aggregated to generate the molecular embedding (h G ) which is used as input for (6), the feed-forward neural network.
Materials and Methods
This section gives a detailed overview of the used data, molecular representation, and the different machine learning methods used throughout this work. The most common notations are shown in Table 1. Table 1. Common notations used throughout this publication.
Notation
Definition τ A non-linear function (e.g., sigmoid or relu) cat(, ) Vector concatenation t Iterator of t steps Edge e uv ∈ E between node u and v N(u) Neighbors of node u N(u)/w Neighbors of node u except w n The number of nodes m The number of edges d The dimension of a node feature vector b The dimension of a edge feature vector X ∈ n×d Feature matrix of a graph
Experimental Data
A total of 10,617 molecules annotated with experimentally derived logD and logP values or logS and logP values were used for model training and predictions. The selected molecules were derived from the Delaney lipophilicity dataset containing experimentally evaluated logD and logP values at pH 7.4 [26] and an aqueous solubility set with logS and logP values [27]. Each dataset was evaluated and molecules were neutralized in both sets. For the aqueous solubility data, salts were stripped off and molecules with logS values lower than −10.0 or higher than 0.0 were removed. The original preprocessed and post-processed data can be found in the GitHub repository [28]. The splitting of each dataset into three subsets for training, evaluation, and testing was completed randomly in a ratio of 81:9:10 for the (training, evaluation, and testing). The data splitting was performed with the same seed for each of the models to be able to compare them using the exact same training, evaluation, and test data. The minimum value of each of the logD, logP, and logS properties was used as an offset to ensure only positive property values. The resulting lipophilicity dataset consisted of 4174 compounds. In total, 3380 were used for training, 376 for evaluating and model selection, and 418 for testing. The post processed solubility dataset contained 6443 molecules. Overall, 5219 compounds were allocated for training, 579 for evaluation, and model selection, and 645 for testing.
Training Approaches
The training strategies differ in the used dataset and the training target (logD, logP, or logS). Under these constraints, seven different types of strategies were used. The first multi-task learning strategy used a combined approach of logD, logP, and logS values referred to as "logD/P/S". Three additional multi-task strategies utilized a combination of two physicochemical properties and are notated as either "logD/P", "logD/S", or "logS/P". Three other single task strategies are only learned on a single physicochemical property and are referred to as either "single task logD", "logP", or "logS". When physicochemical properties from different datasets were used, the individual datasets were first split into training, evaluation, and test sets. Afterwards, each physicochemical property was evaluated and tested individually so that the evaluation and test results of the multi-task learning approaches can be compared to those with a single-task learning strategy.
When testing either single-, or multi-task models, the combined root mean squared error (RMSE) for all properties was calculated as the measure for the best model. For logP, we only used the results from either the first multi-task approach ("multi-task logS/D/P") or the single-task approach with logP values. The reasoning behind this was to use the same test and evaluation data for all models while trying to avoid an unbalanced data bias in favor of logP values. When training with two physicochemical properties where one was logP, we only used the data that had both properties. For example, when training on the lipophilicity dataset which had logP and logD values, we did not include logP compounds from the aqueous solubility dataset and vice versa.
Molecular Graphs
A graph is defined as G = (V, E), where V is a set of nodes and E denotes a set of edges. Let v ∈ V be a node with feature vector x v and e uv ∈ E be an edge pointing from u to v with feature vector x e uv . The adjacency matrix A shows the connectivity of the nodes and in our case it was binary as we did not weigh any connections. It is defined as a n × n matrix with A uv = 1 if e uv ∈ E and A uv = 0 if e uv / ∈ E. We use directed, heterogeneous graphs where e uv = e v u. Heterogeneous graphs contain different types of nodes and edges with their corresponding featurizations.
Molecular Featurization
Five different types of edge and vertex featurizations X were being used for the GNNs. The detailed description of x and x e can be found in Tables A1-A6 in the Appendix A. The feature vectors for the non-GNN models consist of 8 different settings-fingerprints (ECFP or MACCSKeys-shown in Table A7 in the Appendix A) used either in combination with standardized RDKit [29] descriptors or without the descriptors. The descriptors were a combination of all possible and standardized RDKit descriptors, which had a total length of 208. The parametrization of the ECFP was either 1024, 1536, or 2048 bits with a radius of 4. Featurization 3 ( Table A1 in the Appendix A) and 4 ( Table A2 in the Appendix A) only differ in the way the size of ring systems are being represented. Either as a float value calculated by 1 divided by the size of the ring or as a one-hot encoding with 10 possibilities. The node and edge featurization in 5 (Table A3 in the Appendix A) includes two node features (chemical element and formal charge) and one edge feature (bond order). Featurization 6 ( Table A4 in the Appendix A) includes the same node description as 5 and the edge featurization of 3. Featurization 7 (Table A5 in the Appendix A) has the same node featurization as 3 and the same edge featurization as 5. Featurization 8 (Table A6 in the Appendix A) includes a set of optimized node and edge features. This was performed by using a trained D-GIN model and then removing one node or edge feature at a time and observing the RMSE of the prediction. The five node features and the three edge features that had the biggest impact on the RMSE were then taken as the featurization.
The graphs and their featurization were implemented using python version 3.7.8 and the toolkit CDPKit [30].
Directed-Edge GIN (D-GIN) and Reference Models
D-GIN is an extension of the directed-edge message passing neural network of Yang et al. [1] without the additional feature engineering in combination with the graph isomorphic network (GIN) of Xu et al. [18]. Its high level representation can be seen in Figure 1. The principle construction of the network can be seen in the Equations (1)- (8). First, the directed edges were initialized as followed by a t ∈ 1, . . . , T iteration of after which the messages for each directed-edge was being summed as then the message m u was being concatenated as and another message passing over l ∈ 1, . . . , T 2 was performed by x u , if GIN.
afterwards the updated feature vectors h T node of each node were aggregated over the whole molecule as The readout phase was then defined asŷ = f (h G ) where f (·) was a feed-forward neural network. The D-MPNN consisted of Equations (1)-(5) but then used the hidden feature vectors for each node directly by applying Equation (5) and then immediately Equation (7) to encode the whole graph as h G .
GIN on the other hand was initialized and trained, as shown in Equation (6) in order to update the hidden feature vectors of each node. After l update step, the hidden feature vector of each node served as the input of Equation (7) to achieve the aggregated representation h G for the whole graph. D-GIN used all of these functions in a combined way described above Equations (1)- (8). The main principle behind this approach was to first use the key aspect of directed-edge message passing to propagate information via directed-edges to form messages Equations (1)-(4), which then updated the hidden node features Equations (5). These updated hidden node features were then used in the GIN message passing to further propagate information Equations (6) and (7) while also learning . These two information propagation phases are the key aspects of the two different sub-architectures.
Graph Neural Network Implementation, Training, and Hyper-Parameter Search
All GNNs have been implemented and trained using python version 3.7.8 and Tensor-Flow 2.3.0 [? ]. We used TensorFlow's keras models as our super-class and then transferred Equations (1)-(8) into the "fit" method of the keras model. A hyper-parameter search was conducted to find the best parameters which were further used to train all models. Further details on the hyper-parameters are given in the corresponding model's configuration files accessible via the graph_networks Github repository [28].
Each GNN model type was trained twice with either 24 different settings when training on the logD or logS property or 12 on the logP property-in total 48 or 24 training runs per model type were performed. Each non-GNN model type was trained with 8 different settings. For training, evaluation, and testing we split each of the datasets as described in Section Experimental Data. Each of the GNNs were trained for 1600 epochs and the model with the best performance was identified using RMSE as the evaluation metric on the validation set. To evaluate the model type performance, we used the model with the best RMSE of the two runs performed for each model setting. When evaluating the average model type performance, the average RMSE of the different model settings was used for the calculation. To evaluate models with several properties, we summed all RMSEs. For example, when using logD and logP for training, we summed the RMSE of the logD and logP prediction on the evaluation set to receive a combined RMSE. When the combined RMSE was below the last best combined RMSE, the model weights were saved. We used these models to test the model on the test set. Each model was run two times and the results with the best test set performance were taken.
Additionally, the 95% confidence interval range was calculated by applying bootstrapping 100 times while leaving out 10% of the test dataset.
To generate consensus models between GNN and non-GNN models, we combined the best GNN model for each physicochemical property with the best non-GNN model. We did this by adding the predicted log values of one model with the other and then divided it by two. These hybrid models are then called according to their GNN model type plus consensus (e.g., D-GIN cons.).
Other Machine Learning Approaches
We used the random forest (RF), support vector machine (SVM), and k-nearest neighbor (K-NN) implementations of scikit-learn (Version 0.23.2 [32]). Default hyperparameters were used. The featurization is described in Table A7. When using descriptors as input, we standardized them with the scikit-learn StandardScaler. For the fingerprints and descriptors we used version 2020.09.2 of the RDKit [29] python package. Each of the models were trained in a single-task manner for each of the property values.
Results and Discussion
For clarity, we define certain terms used throughout this publication that might have ambiguous meanings. The term "model type" refers to different kinds of machine learning algorithms. For example, a model type can be RF, SVM, KNN, D-GIN, GIN, or D-MPNN. The term "model" refers to a trained model instance with particular training and featurization strategies. The term "training strategy" is used to distinguish between different singleand multi-task training approaches trained with a combination of molecular properties. For example, logD/S/P is used to show that logD, logS, and logP values were used during training. The term "featurization strategy" is used to describe the different node and edge features utilized for the models to train on (Tables A1-A6 in the Appendix A). In addition, we distinguish between consensus (cons.) and non-consensus models. These hybrid models are a combination of the best GNN and best non-GNN models (SVM and D-GIN for logD and logS and RF and D-GIN for logP). To obtain consensus predictions, the predicted property values of the two models were combined and averaged. The averaged values were used as "new" predictions for the RMSE calculation and referred to their GNN model type plus cons (e.g., D-GIN cons).
Overall, 6 different machine learning model types were used in this study. The three GNN model types were D-MPNN, GIN, and D-GIN. The three non-GNN model types were random forest (RF) regression, support vector machines (SVM), and the k-nearest-neighbor (KNN) algorithm. Each model type was trained with the same hyperparameters, but 7 different learning strategies and 6 different node/edge featurization strategies. We trained each GNN model type for each physico-chemical property with all possible strategies twice. Subsequently, the best performing model from each of the two runs (measured on the evaluation set) was selected resulting in 24 models for the logD and logS property and 12 for the logP property, which were then used on the test set and their performance was reported.
The results of this approach are reported and discussed in two parts. First, we discuss different GNNs and non-GNN methods used in this work to identify the best performing model type according to its average performance across all used strategies (discussed in Section General Model Performance). Subsequently, we investigate the impact of the 6 different training strategies (i.e., multi-task vs. single task learning), as well as different featurizations on the performance (discussed in Sections Impact of Molecular Featurization and Impact of Training Strategies).
A dataset of 10,617 molecular structures with annotations for one of the three physicochemical properties was assembled for model training, evaluation, and testing. It included 4174 logD, 6443 logS, and 10,617 logP experimentally measured values. The same training, evaluation, test set was used for all GNN and non-GNN model types.
General Model Performance
In the following, the reported results vary by the used model type. Each combination of featurization and training strategy was used to calculate a total of 24 RMSE values for the logD and logS property, and 12 for the logP property per model type. This resulted in a RMSE distribution shown in Table 2 and Figure 2. For each of these distributions, the average, minimum, and maximum RMSE was calculated and will be reported and discussed subsequently. Table 2. Overview of the best performing machine learning model types independent of training and featurization strategy for prediction of logD, logS, and logP. The performance was calculated as the distribution average over all used model root mean squared error (RMSE) values. In total 24 models were used for the loD and logS property, and 12 for the logP property. RMSE values highlighted in dark and light gray show the best and next best models. Red asterisks mark the lowest RMSE for the non-consensus models for each property prediction.
Molecular
Model Table 2 shows the RMSE distribution average of the different machine learning model types regardless of their training and featurization strategy on the hold-out test set. For each value the standard error of the mean was calculated and added.
For logD property prediction, the D-GIN model type performed with mean, minimum, and maximum logD RMSE of 0.615 ± 0.039, 0.553, and 0.7048, and the corresponding consensus model with 0.575 ± 0.0192, 0.548, and 0.622, making it the best performing model type (results shown in Table 2, and Figure 2). The consensus GIN performed on average (distribution mean of logD RMSE values of 0.666 ± 0.029) better than the best non-GNN method (distribution mean logD RMSE of 0.740 ± 0.068). For the logS prediction, the best model type was the D-GIN consensus model with a average RMSE value of 0.738 ± 0.028 (shown in Table 2 and Figure 2). It performed on average better than the best performing non-GNN model type (SVM), which performed with an average RMSE value of 1.006 ± 0.154 (but it also had a single run with a RMSE value of 0.729 making it the model type with the best single run performance and highlighting the importance of multiple repetitions for reporting model type performances). The consensus D-MPNN also outperformed the D-GIN.
The consensus D-GIN (average RMSE value of 0.455 ± 0.028) and consensus D-MPNN (average RMSE value of 0.475 ± 0.027) showed the best average performance for logP prediction ( Table 2, and Figure 2). The RF and SVM model types also performed with low minimum RMSE values of 0.470 and 0.493, respectively. However, their average RMSE values (RF: 0.681 ± 0.224 and SVM: 0.693 ± 0.134) were higher than the D-GIN and D-MPNN model types.
Consensus models are often used in deep learning applications typically combining either different models that were trained on slightly different training data or multiple model architectures with different strengths and weaknesses. Nevertheless, further investigations are required to give a rationale of why in all our invested cases, the consensus models performed better than their individual counterparts. Furthermore, it should be noted that a direct comparison between the average performance of the GNNs and non-GNN models (RF, SVM, and KNN) can be difficult since the amount of information about a single molecule fed to each of the different model classes is quite different. For example, the non-GNN methods used a wide range of different descriptors and fingerprints shown in Table A7. The reason why the D-GIN outperforms the GIN and D-MPNN could be its higher complexity and network depth. It uses the key aspects of both sub-models and might be able to better abstract higher-order features. This could be facilitated by including skip connections between edge feature extraction mainly performed in the first (D-MPNN) and node feature extraction while learning in the second (GIN) part. This increased complexity could have helped to perform better than its individual parts.
Impact of Molecular Featurization
The average performance of each featurization strategy across all model types and training strategies is shown in Table 3. Considering the performance for all physicochemical properties, featurization strategy 5 showed the highest RMSE (mean logD/logS/logP RMSE of 0.813 ± 0.099, 1.099 ± 0.180, and 0.760 ± 0.110). This trend was also observed when separating according to the model type (shown in Table 4 and Figure 6). The reason for the relatively bad performance of featurization 5 might be that it only included two node properties (chemical element and formal charge), as well as only a single edge feature (bond order- Table A3 in the Appendix A). Table A4 in the Appendix A) also displayed considerably worse performance than other strategies when used in combination with the GIN architecture, for which the mean RMSE performance for logD and logS properties were worse than using featurization strategy 5. One explanation could be that the GIN utilizes node features quite extensively and featurization 6 only included two node feature types similar to featurization 5. The additional edge features in strategy 6 without the appropriate architecture to deal with them could push the optimizer of the GIN network into the wrong direction rather than help with the property prediction.
Although it is easy to identify bad featurization strategies, it is difficult to come up with an unambiguous recommendation for the best performing featurization strategy. The mean RMSE across all training strategies and model types in Table 3 show that featurization 3 and 4 (Tables A1 and A2 in the Appendix A) achieved very good results for logD with a RMSE value of 0.689 ± 0.079 and 0.694 ± 0.072, for logS with a RMSE 0.954 ± 0.146 and 0.948 ± 0.142 and for logP with a RMSE 0.596 ± 0.120 and 0.591 ± 0.105, respectively. Both featurization strategies utilize the maximum number of node and edge features used in this work. They only differ in the way molecular ring sizes are described. Featurization 3 used a float value calculated by 1 divided by the size of the ring system whereas featurization 4 used a one-hot encoding of ten instances (0, 3,4,5,6,7,8,9,10,11). Table 4 shows the mean RMSE values concerning featurization and model type. As performance criteria for featurization strategies we used the sum of model ranks in Table 4. Applying this approach, featurization 3 with two models as best and three models as second-best performers achieved a better ranking than featurization 4 with one model ranked best and two models as second best. Both strategies perform similarly well. Featurization 8 (shown in Table A6 in the Appendix A) used a set of optimized node and edge features. Node and edge features were optimized by masking single edge and node features at a time and evaluating their impact on the test set RMSE. The five node features and the three edge features that had the biggest impact on the RMSE were subsequently used. This approach also revealed that the size of ring systems for the node features appears to be of importance and was, therefore, included in 8. Using featurization 8, we were able to achieve two times the second-best performance. It shows an average good performance, but not as good as featurizations 3 or 4, even though its edge and node features were selected for maximum impact on the final prediction. The mean RMSE of featurization 6 and 7 (Table A5 in the Appendix A) in Table 3 show diminished results compared to featurization 3 and 4.
When evaluating the rank score, the featurization strategy that performs either best or second-best for each physicochemical property, the best featurization strategy was number 6. It was used in four of the best performing runs and once in a second-best run. However, it only performed well in combination with two GNN architectures (D-GIN and D-MPNN) and strongly underperformed with the GIN. The D-GIN and D-MPNN architecture types use primarily edge features for their information propagation and featurization strategy 6 provided these. It utilized only two-node feature types, potentially reducing the noise for the feature extraction to a minimum in this setting.
On average, featurization strategies 6 and 7 performed similarly well. However, when separating the results at a model type level, it became evident that there was a strong model architecture dependency, so it seems important to choose the features according to the architecture at hand (Figures 3-5). Furthermore, featurization 3 might perform worse than featurization 6 or 7. Nevertheless, when unsure which features to use, simply adding more features could be the safer option rather than using less. This observation is also supported by comparing featurization 3 or 4 to, e.g., 6, 7, or 8. When analyzing the results for the non-GNN models and their different featurizations, the mean RMSE variance was large in comparison to the GNN models. Moreover, in similar deep-learning benchmark studies that predicted molecular properties, predominantly fingerprints have been used. From Tables A13-A15 in the Appendix A, one can see that especially featurizations that include descriptors in addition to fingerprints perform exceptionally well. We think that when comparing GNN with non-GNN models, differences in used features should be taken into consideration when trying to identify and understand (deep-learning) method performance.
Impact of Training Strategies
The impact of different training strategies are shown in Table 3. The lowest mean logD RMSE can be obtained by a multi-task strategy that involves learning on both logD and logP values. This is similar to the best training strategy for the logS property, which is a multi-task approach including logS and logP properties. As for the logP property, the best approach is a single-task strategy including logP values, however the multi-task approach which combines all physicochemical properties achieves similarly good performance.
When analyzing the logD/S/P RMSE predictions with respect to training strategy and model type, Table 5 and Figures 7-9 show that there is no particularly favorable learning strategy for any of the model types. The datasets used in this study are specific for one particular physicochemical property. When comparing different learning strategies we thus focused on one particular physicochemical property for each model type. Starting with the results for the prediction of the logD property in Table 5, we can see that the overall best model (red asterisk), as well as the two best models for each model type (dark gray), are multi-task models. In particular, the models with a combination of logD and logP properties perform well. Tables A8-A11 in the Appendix A. Table 5. Impact of the training strategies on the different molecular properties. Each model type is evaluated separately. For each property and model type, the mean, minimum, and maximum RMSE are shown. The dark gray boxes represent the best RMSE for the particular property and model type, the light gray the second best. The red asterisk highlights the overall best RMSE for the particular property. Considering all combinations of training and featurizations strategies for each model, the learning strategy with the best average, as well as the best minimum logD RMSE was obtained using the logD/P multi-task training approach resulting in RMSE values of 0.719 ± 0.105 and 0.553, respectively (Table 3. Yet, using this multi-task learning strategy we also obtained single run performance worse than using a single-task learning strategy with only logD values, showcasing once more the importance of validating multiple learning and featurization strategies. The results are similar for the prediction of logS values: again, the multi-task learning strategy performs better than its single task counterpart. The best model for logS prediction was obtained by training on logD, logS, and logP values. Considering all combinations of training and featurization strategies for each model, the best average, minimum, and maximum logS RMSE of 0.979 ± 0.166, 0.795, and 1.325, respectively, was observed during the multi-task training with all properties. We should note here that while it seems that the average performance is improved by multi-task learning, the variance of model performance is also increased.
Conclusions
We introduced the directed-edge graph isomorphism network (D-GIN), a novel graph neural network that showed improved average performance for aqueous solubility and lipophilicity prediction compared with other baseline models. We showed that by combining different models with distinct key characteristics, we can increase the depth of the model while also improving its predictive power. Furthermore, applying different training strategies and featurizations constraints enables to obtain more information regarding general, average model performance. This strategy showed that the D-GIN model outperforms other machine-learning models on average and argued that comparing the mean performance rather than single metric values of the best performing model type gives more insight into the general behavior and ultimately facilitates a better understanding and higher robustness of deep-learning models.
In concurrence with previous publications [33][34][35], we showed that there is a tendency towards multi-task learning approaches for the GNNs utilized in this survey. On average they performed better than their single-task counterpart for the corresponding physicochemical property. We could not find clear evidence that more than two properties increase the model's performance.
Furthermore, we highlighted that the usage of additional features did not improve the GNN model performance. However, we also conclude that very little featurization led to the worst performance. In general it is necessary to be aware of the type of GNN that is used and whether its architecture focuses more on edge or node features. When trying to obtain the best performing model it can be advisable to do feature engineering, but when in doubt which features to use, it can be safer to use more than less. We showed that this awareness can help improve the GNNs predictive power at hand.
For the non-GNN models, we could conclude that by excessively adding descriptors to the molecular fingerprint the performance of these models increases substantially. We further argued that for future comparisons it would be advisable to include not only fingerprints but also descriptors to the non-GNN baseline models to be more competitive.
By combining the best GNN model with the best non-GNN model we could see a slight improvement in the overall performance in all cases. Consensus models have often shown to improve performance. However, in this case, further investigations are needed to attain to a conclusion on why this is the case.
We showed that advanced deep-learning methods such as GNNs do have great potential in the physicochemical property prediction area and, when applied properly, can serve as a promising and robust method for any computer-aided drug discovery pipeline, especially for chemical property prediction.
Conflicts of Interest:
There are no conflict of interest or disclosure associated with this manuscript.
Appendix A
The appendix includes informational materials that show the featurization of the GNN and non-GNN baseline models in Table A7, MACCSKeys --Yes The last column consist of the unique identify. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A9. Shows the logD RMSE and r 2 results for the D-MPNN model type used during this survey. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A10. Shows the logD RMSE and r 2 results for the GIN model type used during this survey. The last column consist of the unique identify. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A11. Shows the logD RMSE and r 2 results for the non-GNN model types used during this survey. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A12. Shows the logS RMSE and r 2 results for the D-GIN model type used during this survey. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A13. Shows the logS RMSE and r 2 results for the D-MPNN model type used during this survey. The last column consist of the unique identify. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A15. Shows the logS RMSE and r 2 results for the non-GNN model types used during this survey. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A16. Shows the logP RMSE and r 2 results for the D-GIN model type used during this survey. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A18. Shows the logP RMSE and r 2 results for the GIN model type used during this survey. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP. Table A19. Shows the logP RMSE and r 2 results for the non-GNN model types used during this survey. The last column consists of the unique identifier. Training strategy 0 represents a training strategy combining logD, logS, and logP, 1 represents a strategy combining logD and logP, 2 stands for a combination of logD and logS, 3 represents a strategy using logS and logP, 4 represents a strategy using logD, 5 represents a strategy using logS, and 6 represents a strategy using logP.
|
2021-09-17T07:50:27.907Z
|
2021-08-30T00:00:00.000
|
{
"year": 2021,
"sha1": "cba94617cae626f17efd48921b109e9f78d0267b",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/1420-3049/26/20/6185/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "8ddcaf37d266d7d98f86cb987f7fdab8cb03d097",
"s2fieldsofstudy": [
"Computer Science"
],
"extfieldsofstudy": [
"Medicine",
"Computer Science"
]
}
|
12070669
|
pes2o/s2orc
|
v3-fos-license
|
Organic narrowband near-infrared photodetectors based on intermolecular charge-transfer absorption
Blending organic electron donors and acceptors yields intermolecular charge-transfer states with additional optical transitions below their optical gaps. In organic photovoltaic devices, such states play a crucial role and limit the operating voltage. Due to its extremely weak nature, direct intermolecular charge-transfer absorption often remains undetected and unused for photocurrent generation. Here, we use an optical microcavity to increase the typically negligible external quantum efficiency in the spectral region of charge-transfer absorption by more than 40 times, yielding values over 20%. We demonstrate narrowband detection with spectral widths down to 36 nm and resonance wavelengths between 810 and 1,550 nm, far below the optical gap of both donor and acceptor. The broad spectral tunability via a simple variation of the cavity thickness makes this innovative, flexible and potentially visibly transparent device principle highly suitable for integrated low-cost spectroscopic near-infrared photodetection.
M ixtures of organic semiconductors with displaced frontier orbital energies are applied in organic lightemitting diodes (OLEDs), solar cells and photodetecting devices. In molecular blends that allow for charge transfer, it is energetically favourable for an excited electron to reside on the acceptor molecule (A) and a hole to reside on the electron donating molecule (D) 1 . An encounter of both charges at a donor-acceptor interface results in the formation of an intermolecular charge-transfer (CT) state which can redissociate into free carriers or recombine to the ground state 2 . Their decay can be optimized for efficient emission as recently demonstrated for 'exciplex'-type OLEDs with a respectable external electroluminescence quantum yield of 15% (ref. 3).
Furthermore, CT states play an important role in organic photovoltaic and photodetecting devices, as they mediate between photo-generated excitons in neat absorbers and free charges 4 . As low-energy recombination centres, they limit the open-circuit voltage of organic solar cells 5,6 as well as the minimum achievable dark current in photodetectors 7 . Current research efforts in both device classes aim to synthesize absorber materials with sensitivity in the near-infrared (NIR) by lowering their optical gap 8 . However, despite several strategies [9][10][11][12][13] , new organic materials rarely result in devices with reasonable responsivity above 1 mm (refs [14][15][16][17]. Moreover, most organic donor-acceptor (D:A) blends offer a rather broadband absorption, while a tunable narrowband absorption would be of particular interest, for example, for spectroscopic applications 18,19 .
Here, the use of direct intermolecular CT absorption could provide an elegant solution: as its absorption onset is determined by the energetic difference between the highest occupied molecular orbital (HOMO) of the donor and the lowest unoccupied molecular orbital (LUMO) of the acceptor, it can be substantially redshifted as compared to the absorption of neat materials 20 . However, due to its intermolecular nature, CT absorption is typically two orders of magnitude weaker than singlet absorption of the neat absorbers, leaving it often undetected 20 . To increase the CT absorption, ultrathick D:A blends of about 10 mm thickness have been suggested 21 . Due to a high series resistance, such a device can only be read out when applying extraction voltages in the order of 100 V. As a result of low on/off ratios, and presumably low speed, this approach to exploit CT absorption for photodetection is apparently abandoned. Therefore, in contrast to CT emission in 'exciplex' OLEDs, CT absorption is practically unused in organic optoelectronic devices up to now.
In this work, we introduce a resonant optical microcavity device which exploits the weak, but spectrally broadband CT absorption to achieve narrowband photodetection and tunable resonance wavelengths. We demonstrate a 41-fold enhancement of the external quantum efficiency (EQE) at the cavity resonance wavelength, with full-widths at half-maximum (FWHMs) down to 36 nm. At short circuit, we achieve EQEs exceeding 20% with photocurrent generation predominantly due to direct CT state excitation. A variation of the cavity thickness allows to tune the resonance wavelength in the CT absorption band over a vast range, from 810 to 1,550 nm, using a single D:A blend. We believe the original device concept introduced in this paper to open doors towards organic, flexible, cheap and ultracompact NIR spectrometers.
Results
The resonant cavity device architecture and a simplified energy level diagram are shown in Fig. 1a,b. We utilize microcavities formed by a fully reflecting and a partially transmissive silver (Ag) mirror [22][23][24] . The simultaneous use as electrodes allows a compact design incorporating both optical and electrical elements. The photo-active D:A blend is sandwiched between transparent transport layers which allow selective charge extraction towards the outer electrodes, analogous to state-ofthe-art organic photovoltaic diodes [25][26][27][28] . Within the Fabry-Pérot interferometer formed by the two Ag mirrors, electromagnetic waves with a wavelength l res experience constructive interference when l res ¼2nt cav =j, with j being a natural number. The effective optical cavity thicknessnt cav is determined by the average index of refractionn and the effective geometrical cavity thickness t cav , that is, the distance between both Ag mirrors extended by the field penetration into both metal layers 29,30 .
To demonstrate the concept, we use a photo-active blend comprising buckminsterfullerenes (C 60 ) as electron acceptor and zinc phthalocyanine (ZnPc) as electron donor whose chemical structures are depicted in Fig. 2b. Their energy levels are displaced as illustrated in Fig. 1b. The ZnPc:C 60 blend is known to display a substantially redshifted CT state as compared to the neat absorbers 31 . A volume ratio of 1:1 provides a maximum contact interface between ZnPc and C 60 . A sensitive EQE spectrum of a reference photovoltaic device with minimized cavity effects is depicted in Fig. 2a as crossed green line. The measurement confirms the presence of CT absorption in the NIR above 850 nm, that is, below the optical gap of neat C 60 (700 nm) and ZnPc (815 nm) whose absorption spectra are shown as grey lines marked as I and II, respectively.
Cavity-enhanced CT absorption. To make use of the broadband weak CT absorption, we greatly enhance the optical field in the NIR utilizing a resonant microcavity. The thicknesses of the transport and Ag layers are optimized via optical transfer-matrix simulations 32 for an absorber blend of 50 nm, as shown in Supplementary Fig. 1. To reach narrowband cavity resonances in the spectral region of CT absorption, we choose transport layer thicknesses of above 60 nm which exceeds those typically used in organic solar cells [25][26][27]33 . The EQE spectrum of a device fabricated with optimum transparent electrode ( logarithmic scale as purple line marked as~. The EQE reaches 18% at the resonance wavelength l res of 950 nm with an FWHM of 36 nm. As compared to the reference device without cavity enhancement, we found a 27-fold increase in EQE at 950 nm. Moreover, a neat material filter sequence outside the cavity quenches the photo-response stemming from strong absorption above the optical gap of ZnPc, as shown in Fig. 1a and Supplementary Fig. 2. Resonance wavelength tuning. To demonstrate the tunability of the detection wavelength l res , we vary the thickness of both transport layers between 60 and 105 nm, resulting in a spectral shift of the cavity resonance. As shown by the EQE spectra in Fig. 2a, we obtain resonances distributed over more than 200 nm within the CT band of ZnPc:C 60 . A maximum EQE of 23% is achieved for a resonance wavelength of 875 nm, as depicted by the dark blue line marked as . On increasing cavity thickness, the resonance wavelength shifts from 875 to 1,085 nm with increasing EQE amplification up to 41 times (for more details see Supplementary Fig. 3). While the EQE decreases for longer wavelengths, the large cavity enhancement still guarantees values above 1%, even 380 meV below the optical gap of neat ZnPc. Furthermore, to demonstrate an important asset of organic devices, two samples, with a detection wavelength of 995 nm (b) and 1,035 nm (c), are deposited onto a flexible substrate. Exemplary, one fully flexible encapsulated device is depicted in Fig. 2c. To allow for further device integration, we demonstrate photodetectors with visible transparency by shaping both electrodes partially translucent, as shown in Supplementary Fig. 8.
Angular dependence of resonance. Since the device class introduced here is based on optical interference, the cavity enhancement depends on the angle of light incidence. Figure 3 shows the absorption (3a) and EQE spectra (3b) of a device resonating at 950 nm, marked as~in Fig. 2a, as a function of the angle of incidence with respect to the substrate normal. The resonance wavelength decreases for non-normal excitation, with the peak position following a parabolic dispersion, leading to a reduction in resonance wavelength of more than 50 nm for angles above 45°. As illustrated in Fig. 3c, this behaviour is reproduced by transfer-matrix simulations. Close to normal incidence, there is however a rather broad angular range, spanning from À 20°to þ 20°, where the resonance shift does not exceed 10 nm which is substantially smaller than the FWHM of 36 nm.
Identification of parasitic absorption. In the reported devices, the peak EQE is currently limited by parasitic absorption, which decreases the number of photon transits through the D:A blend. This observation is evidenced in Fig. 4a depicting the experimentally measured device absorption (filled area) and the corresponding EQE (hatched area) of three previous ZnPc:C 60 devices whose EQEs peak at 910, 950 and 995 nm, respectively. While the peak EQE significantly drops ( À 59%) with increasing l res , the device absorption remains almost unchanged ( À 3%). Therefore, parasitic absorption greatly exceeds CT absorption. More insight into the origin of the parasitic absorption at the cavity resonance is obtained from optical simulations, depicted in Fig. 4b. We identify the transmissive top electrode, where almost every second photon absorption occurs, as main responsible. Moreover, the transparent and reflective electrode together yield the highest parasitic contribution with between 60 and 80% of the resonance absorption, followed by the doped N4,N4 0 -bis(9,9dimethyl-9H-fluoren-2-yl)-N4,N4 0 -diphenylbiphenyl-4,4 0 -diamine (BF-DPB) hole transport layer with about 15%.
Longer detection wavelengths. While providing the proof of principle, ZnPc:C 60 has a CT absorption band limited to wavelengths not exceeding 1,100 nm. To detect photons with lower energies, we exchange ZnPc by 2,2 0 ,6,6 0 -tetraphenyl-4,4 0 -bipyranylidene (TPDP), chemical structure in Fig. 5b) as donor with an elevated HOMO level 34 which reduces the onset energy of CT absorption. A sensitive EQE measurement of a TPDP:C 60 solar cell with minimal interference effects reveals a remarkably broad and featureless CT band from 725 nm down to 1,600 nm, as shown in Supplementary Fig. 4. Utilizing the approach outlined above, we tune the cavity by thickness adjustments of both transport layers (marked as [triangle down]) -and for selected devices also by thickness variations of the D:A blend (D). For more details on the layer thicknesses used, we refer the reader to Supplementary Table 2. Following this strategy, we achieve a remarkably broad resonance tunability from 810 to 1,550 nm as shown in Fig. 5a. We like to emphasize that the latter detection wavelength, addressing CT states 1 eV below the optical gap of both neat C 60 and TPDP, settles among the highest wavelengths achieved with organic photodetectors so far [14][15][16] . Up to now, we observe rather low internal quantum efficiencies for the material blend TPDP:C 60 , decreasing with blend thickness from 20 to 3%. However, we expect future alternative D:A combinations to achieve higher internal yields for CT excitation in this wavelength range.
Discussion
The device concept outlined in this work has great potential as an original class of organic, narrowband NIR detectors, with an easily tunable detection wavelength. It is worth emphasizing that the wavelength selectivity, provided by these devices, is explicitly due to the weakly absorbing nature of CT states: in contrast, more strongly absorbing neat material transitions do not allow such a strong cavity enhancement and narrowband resonances as observed here, except within their narrow absorption tail region (for more details see the Supplementary Discussion) 22,35 . Moreover, the outlined detection principle undergoes a paradigm change: while conventionally excitons diffuse 36 from either the absorbing donor or acceptor to a joint D:A interface to decay into CT states, those steps are skipped here in favour of a direct absorption of the latter. Demonstrating detectors with EQEs around 20%, this work underlines experimentally that CT states can be rather efficiently converted into photocurrent-which is contrary to their previous perception as trap states 37 . Given the achieved dark currents, which are still limited by extrinsic device shunts, we estimate an upper limit for the specific detectivity at the resonance wavelength of the ZnPc:C 60 -based detectors of 10 11 Jones at short circuit (details are provided in the Supplementary Methods and Supplementary Fig. 5). We expect a further reduction in noise current and enhancement in detectivity by optimizing the absorber layer thickness or introducing undoped blocking layers. On a variation of the excitation intensity, we observe no deviation from a linear photo-response over more than 5 orders of magnitude, as shown in Supplementary Fig. 6. We measure rise and fall times (10 to 90% or reverse) of 3 and 151 ns, respectively (see Supplementary Fig. 7 and Supplementary Table 1). The latter is partly delayed by discharge dynamics 38 of a resistor-capacitor circuit of a 0.25 mm 2 large device which might be accelerated by reducing the device area. However, this response time is already sufficiently short for numerous applications related to NIR photodetection such as contact-free movement detection, noninvasive subsurface vision or night vision 9 .
A further increase in EQE and specific detectivity is expected when improving the ratio between CT absorption and the competing parasitic absorption. A first group of approaches aims to enhance the interfacial absorption, for example, by increasing the D:A blend thickness or by exploiting intercalating D:A blends with rather high CT absorption coefficients 39,40 . A second strategy consists of reducing the competing amount of parasitic absorption. As discussed in Fig. 4, the EQE of the presented ZnPc:C 60 devices can be improved via transport layers with suppressed NIR absorption. A much more drastic enhancement in EQE height and FWHM is expected when replacing the conducting Ag mirrors, being the dominant source for parasitic absorption: for this purpose, low loss mirrors such as distributed Bragg reflectors 41 or other high-quality resonators 42,43 , paired with NIR transparent electrodes with a high in-plane conductivity would offer a promising perspective.
A multitude of NIR applications in biomedicine, pharmacy and agriculture relies on spectroscopic analysis-such as disease detection 44 , determining blood concentrations of glucose, oxyhemoglobin and water 9,45 , analysing and interacting with brain functions 46,47 ; raw material and on-line quality monitoring 48 ; or determining nutrient compositions and optimal harvest dates 49,50 . Following previously reported approaches, the realization of an organic NIR spectrometer would require several different donor or acceptor materials with varying optical gaps 18,51 . The replacement of the neat material extinction by the interfacial CT absorption which extends over several hundreds of nanometers provides an elegant, robust and cheap alternative: here, all detection wavelengths within the spectrometer range can be addressed solely by a thickness variation for a single D:A blend, as outlined in Fig. 5a. Especially for the analysis of chemical compositions as common task for NIR spectroscopy, an even further extension of the detection wavelengths into the infrared would be desirable. Hereby, the outlined strategy will provide photosensitivity also beyond 1,550 nm on a proceeding reduction of the CT absorption onset, by an appropriate choice of the frontier energy levels of both D and A. However, using organics, we expect a detection limit at about 2 mm. Here, fundamental interatomic vibration modes will cause strong parasitic absorption 9 and, consequently, reduce the number of photon transits in the resonator. In summary, we introduce an innovative class of organic narrowband NIR photodetectors based on mixtures of C 60 and donor materials with a high HOMO level. An optical cavity device architecture enhances the photocurrent for wavelengths within the intermolecular CT absorption band. Using mixtures of ZnPc:C 60 , we obtain narrowband photodetection at wavelengths below the optical gap of ZnPc and C 60 with EQEs of above 20% and spectral widths down to 36 nm. For photodetectors based on TPDP:C 60 blends with a lower CT absorption onset, we demonstrate a tunability of the resonance wavelength over a strikingly broad range from 810 to 1,550 nm by simple variations of the cavity thickness. We believe that, due to its mechanical flexibility, light weight, scalability, low fabrication cost and potential transparency at visible wavelengths, the introduced device class will become a valuable candidate for integrated spectroscopic NIR photodetection.
Methods
General fabrication procedure. Precleaned glass is used either as a neat rigid substrate or with a prestructured layer of 90 nm indium tin oxide (ITO; Thin Film Devices, USA) deposited on top. Flexible devices are processed onto 125-mm-thick films of planarized polyethylene naphthalate (pPEN; Teonex (R) PQA1M, DuPont Teijin Films, UK). Before device deposition, the flexible pPEN substrates are covered with 20 nm of aluminium oxide (AlO x ) as gas barrier by means of plasmaenhanced atomic layer deposition (Sentech SI ALD LL, Sentech Instruments, Germany), as earlier reported in ref. 52.
The subsequent layers composed of organics, fullerene, oxides and/or metals are deposited via thermal evaporation under controlled vacuum with a base pressure of 10 À 8 mbar (K.J. Lesker, UK). Evaporation rates, layer thicknesses and, where applicable, mixing ratios are controlled via quartz crystal microbalances, with rates not exceeding 1 Å s À 1 . The geometrical intersection of the bottom Ag or ITO electrode and the top Ag or aluminium electrode defines a photo-active area of 6.4 or 0.25 mm 2 .
After evaporation, all glass samples are covered with special encapsulation glasses which leave a sealed hollow volume filled with nitrogen above the device. A ultraviolet-cured epoxy glue (XNR 5592; Nagase ChemteX, Japan) is used to seal the sample at the rim of the encapsulation glass. All samples on flexible substrates are sealed with another flexible barrier film against oxygen and moisture. For this, a pPEN substrate with predeposited 20 nm atomic layer deposition AlO x is laminated (full area) onto the device. This is done using a 25-mm-thick, ultraviolet-cured, proprietary barrier glue (Tesa SE, Germany), containing a latent getter. The lamination is carried out at room temperature in inert atmosphere. AlO x films of both flexible barriers are placed directly adjacent to the device to minimize edge diffusion.
As shown in Supplementary Table 1, a ZnPc:C 60 reference device with minimal optical interference uses an ultra-thin, highly transparent Ag electrode and transport layers thinner than 40 nm to resonate at the peak extinction of ZnPc above its optical gap. Further cavity-enhanced ZnPc:C 60 photodetectors are built utilizing an alternative hole transport layer with F 6 -TCNNQ as disclosed p-dopant. As shown in Supplementary Fig. 8, the formation and spectral shift of resonance peaks can be reproduced. Moreover, a reference device with neat C 60 as photo-active layer, marked as I, follows the layer sequence Glass|ITO|MoO 3 |C 60 |BPhen|Ag. A further neat ZnPc reference device, marked as II, follows the layer sequence Glass|ITO|C 60 F 36 |BF-DPB:C 60 F 36 |ZnPc|BPhen|Al.
TPDP:C 60 series. TPDP is synthesized from dimerizing pyrylium salts via a Wittig reaction as described in ref. 55. The layer sequence of all TPDP:C 60 devices is documented in Supplementary Table 2.
Measurement techniques. For details on the measurements techniques, we refer the reader to the Supplementary Methods. Data availability. The data that support the findings of this study are available from the corresponding author on reasonable request.
|
2018-04-03T00:25:57.836Z
|
2017-06-05T00:00:00.000
|
{
"year": 2017,
"sha1": "bedb152aba67cc6464e9ca66e1e644125c5995aa",
"oa_license": "CCBY",
"oa_url": "https://www.nature.com/articles/ncomms15421.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "bedb152aba67cc6464e9ca66e1e644125c5995aa",
"s2fieldsofstudy": [
"Physics",
"Chemistry"
],
"extfieldsofstudy": [
"Medicine",
"Materials Science"
]
}
|
209289295
|
pes2o/s2orc
|
v3-fos-license
|
A review of clinical photoacoustic imaging: Current and future trends
Photoacoustic imaging (or optoacoustic imaging) is an upcoming biomedical imaging modality availing the benefits of optical resolution and acoustic depth of penetration. With its capacity to offer structural, functional, molecular and kinetic information making use of either endogenous contrast agents like hemoglobin, lipid, melanin and water or a variety of exogenous contrast agents or both, PAI has demonstrated promising potential in a wide range of preclinical and clinical applications. This review provides an overview of the rapidly expanding clinical applications of photoacoustic imaging including breast imaging, dermatologic imaging, vascular imaging, carotid artery imaging, musculoskeletal imaging, gastrointestinal imaging and adipose tissue imaging and the future directives utilizing different configurations of photoacoustic imaging. Particular emphasis is placed on investigations performed on human or human specimens.
Introduction
Photoacoustic (PA) imaging (PAI), or optoacoustic imaging, is a hybrid imaging modality that merges optical illumination and ultrasound (US) detection [1,2]. It has been rapidly gaining popularity and explored for biomedical imaging applications in the last two decades. As pure optical imaging methods cannot maintain high-resolution imaging in deep biological tissues due to optical scattering, the capability to achieve high resolution optical contrast images in biological tissues up to centimeters depths makes PAI a promising technique for various clinical imaging applications. In PAI, a nano-second pulsed laser (pulse duration < 10 ns) is commonly used to illuminate the biological sample. The molecules absorb the optical energy and convert it into heat, generating a temperature rise. The thermoelastic expansion from the temperature rise generates acoustic waves which is detected using ultrasonic transducers. As sound scatters 1000 times lesser than light [3,4], the acoustic signal propagates much longer in biological tissue without significant attenuation.
PAI capitalizes mainly on the intrinsic optical absorption of chromophores in human tissue such as hemoglobin, melanin, lipid and water. As each of these chromophores exhibits its own characteristic absorption spectra, PAI at multiple wavelengths allows their relative quantification and helps to investigate physiological changes in disorders to understand the mechanism behind them and how they can be managed effectively. Due to its label-free nature, it encourages patients' compliance and is capable of long-term longitudinal monitoring. Because PAI can differentiate between deoxy-hemoglobin (Hb) and oxyhemoglobin (HbO 2 ), measuring the sO 2 in the blood vessels is possible, which is a significant characteristic of ischemia, hypoxia or hypoxemia physiological states for other applications [5]. PAI can exploit the optical absorber in skin, melanin, for imaging of melanoma, pigmented lesions and hair follicles. The presence of lipids in certain areas such as arterial plaques and sebaceous glands makes it convenient for PAI interrogation. However, if the endogenous contrast is not sufficient due to the similarity of absorption spectra of one or more chromophores, various exogenous contrast agents can also be employed to augment PAI for better contrast and deeper imaging [6][7][8]. It is interesting to note that clinically relevant FDA approved contrast agents such as indocyanine green, ICG [9] and methylene blue [10] are photoacoustically strong chromophores.
Generally, PAI systems can be grouped into three configurations subjected to their combination of optical illumination methods and acoustic detection methods: tomography, mesoscopy and microscopy systems (Fig. 1) [11]. A typical photoacoustic tomography (PAT) system uses wide-field illumination and detects the acoustic waves by a single element [12,13] or array transducer [14,15]. The use of a single element transducer is time-intensive and affords only static imaging, making it inadequate for in vivo clinical imaging applications. By using multiple transducers for detection, the emitted acoustic waves can be captured at multiple angles and hence ideal for real time monitoring of the imaging region with increased acquisition speed. As the transducers can be arranged in 2D or 3D spatially, a 2D cross-sectional or 3D volumetric image can be reconstructed from all projections. Depending on the applications and imaging targets, the transducer array can be mounted on a stationary platform [16,17] or be used in hand-held imaging mode [18,19]. The latter imaging configuration is especially practical in clinical practice for better accessibility to the various parts of the human body. The 2D array transducer can be either linear or curved. Linear array based PA systems can be developed using commercial ultrasound machines [20][21][22]. They are however hampered in detecting acoustic signals at limited angles from the region of interest as compared to curved transducer array systems [17]. If the transducers can be arranged spatially in 3D, acoustic waves can be detected from all in plane directions and hence complete volumetric images can be acquired in a single frame. A laboratory prototype of 3D array transducers with wide field-of-view (FOV) [23,24] and commercial handheld transducer with small FOV [25] can be used for various clinical imaging needs. Several companies, including iThera Medical GmbH, FUJIFILM VisualSonics, TomoWave Laboratories Inc., ENDRA Life Sciences Inc., Canon and Seno Medical Instruments, etc. provide commercial transducer-array-based PAT systems for medical imaging. Even though PAT systems are able to visualize vascular structures and intrinsic contrast at centimeters depth, they are not well-suited for high-resolution superficial vascular imaging applications.
Photoacoustic microscopy (PAM) on the other hand is a high-resolution PAI technique which can further be categorized into two implementation methods: acoustic resolution-PAM (AR-PAM) and optical resolution-PAM (OR-PAM) [26]. Deep tissue imaging at high resolution (∼ 20−50 μm) beyond optical diffusion limit (∼ 1 mm in biological tissue) can be accomplished in AR-PAM by utilising weak optical and tight acoustic focusing. The lateral resolution of AR-PAM is dependent on the centre frequency of the ultrasound transducer and the acoustic lens used, with the most common configuration being 50 MHz ultrasound transducer and 0.44 N.A acoustic lens to give a lateral resolution of ∼45 μm and penetration depth of up to 3 mm [27,28]. In order to achieve high lateral resolution down to single capillary (∼ 5 μm) using AR-PAM, high-frequency transducers are needed (higher than 400 MHz central frequency), which restricts the penetration depth to less than 100 μm. Instead, the lateral resolution can be enhanced by a tight focusing of an optical beam as in an OR-PAM setup [29,30], which enables organelle and cellular level imaging from its high lateral resolution spanning hundreds of nanometer to several micrometers [31]. In an OR-PAM setup, a spatially filtered laser beam is focused using a microscope objective to pass through an optoacoustic beam combiner comprising of both right angled and rhomboid prisms with a coating of silicon oil in between which functions as an optically transparent and acoustically reflective film. An acoustic lens below the rhomboid prism provides acoustic focusing. An ultrasonic transducer placed on top of the rhomboid receives the acoustic signals. For OR-and AR-PAM image acquisition, two-dimensional continuous raster scanning imaging heads are used. An A-line is the time-resolved PA signals multiplied by the speed of sound, 1540 m/s in soft tissue. Multiple A-lines acquired during the continuous Y stage movement produces a two-dimensional B-scan. Multiple B-scans are acquired and processed to obtain the maximum amplitude projection (MAP) photoacoustic images. Recently, the use of microelectromechanical systems (MEMS) has improved OR-PAM imaging speed significantly [32,33]. The implementation of MEMS in the setup will help in the miniaturization of the technique and will be highly beneficial for the development of endoscopic devices and in clinical translation.
iThera Medical GmbH, modified the concept of AR-PAM by using a customized ultra-broadband transducer and a two-arm fiber bundle light delivery, with this specific implementation termed as photoacoustic mesoscopy or raster-scanning optoacoustic mesoscopy (RSOM). Herein, wide field optical illumination and a single tightly focused high center frequency (typically ≥ 50 MHz) US transducer with a wide bandwidth range (10-180 MHz) are utilized to achieve high spatial resolution, (typically 15-40 μm) [34]. By reconstructing the acquired data and segmenting them into two different frequency bands, e.g. 10-30 MHz (low) and , and then spatially merging them together, the location of capillaries and bigger blood vessels can be differentiated. Achieving a depth of 2 mm, this configuration has been demonstrated to be highly useful in understanding conditions related to human skin like psoriasis [34] and eczema (unpublished data).
As such, all three configurations of PAI systems mentioned above have demonstrated unparalleled advantages in certain clinical applications, including deeper imaging of biological tissue, high spatial resolution for in vivo imaging, contrast-free and stain-free imaging capability of endogenous molecules which are discussed in detail below.
Breast imaging
Photoacoustic imaging of the breast has garnered immense interest among researchers worldwide owing to the penetration depth it can comfortably achieve to cover the entire, if not, most of the breasts. Breast cancer is the most commonly occurring cancer in women with over 2 million new cases in 2018 [35]. Imaging plays a significant role in breast screening, diagnosis, staging and monitoring therapeutic interventions. While X-ray mammogram is the most commonly used method to screen and diagnose breast cancer in spite of it being painful, having the ill-effects of exposure to ionizing radiations and reduced accuracy in denser breast [36], ultrasonography follows next, especially to differentiate between benign and malignant cysts [37] with magnetic resonance imaging (MRI) and other nuclear imaging modalities [38] prescribed to lesser extent. Angiogenesis and hypoxia being few of the major hallmarks of cancer, sensitivity of optical imaging to the different forms of hemoglobin make it highly reliable over other imaging modalities and encourage using it in combination with other radiology methods [39] to improve the sensitivity. However, because of the penetration limit of light to ∼1 mm in tissues, the clinical translation of optical methods for breast cancer have not been successful. PAI effectively breaks through this penetration depth by combining functional imaging and contrast of optical tomography and high spatial resolution of US to reach depths of up to 4 cm [23] and thus gathering immense interests from many research groups, particularly for breast cancer as it is one of the major superficial organs. Several configurations of the photoacoustic systems have been and are being developed by various groups to obtain the maximum possible information from breasts of various phenotypes to give high spatial and temporal resolution images of angiographic structures with minimal motion artifacts. Different in vivo studies performed on normal and breast cancer patients and ex vivo studies performed on breast specimen will be discussed briefly in this section.
In vivo imaging
The first use of PA for imaging breasts in patients was demonstrated as early as 2001 by Oraevsky et al. [40]. Following this, several studies demonstrating the technical utility of various manifestations of PA breast imaging systems to visualize malignancies in human breasts were published [41][42][43][44][45][46]. One of the first clinical studies performed by Manohar et al. was using a stand-alone clinical breast imaging system called the Twente Photoacoustic Mammoscope which uses a single NIR laser, 1064 nm for excitation and an ultrasound detector array with 588 elements providing illumination from the cranial direction. The system has metamorphosed through a few versions to acquire PA signals from a field of view (FOV) of 9 × 8 cm 2 within 10 min with a resolution of approximately 3.5 mm [47]. With the PA images obtained using this system, 32 of the 33 lesions could be detected successfully [24]. Moreover, the PA appearances of these high intensity lesions which can be broadly categorized into -1) mass-like, 2) ring-like and 3) non-mass or scattered were similar to those observed on MR dynamic contrast enhanced (DCE) imaging and vascular stained histopathology of breast malignancies suggesting the contribution of vasculature to the high intensity lesions spotted in malignancies than lipids or collagen which could also contribute to PA contrast [47]. Most importantly, the contrast of the lesion was independent of the breast density [24], which when higher makes the diagnosis of tumor nearly impossible on X-Ray mammogram [41]. The problem of imaging denser breasts has been demonstrated to be overcome by also using a volumetric three dimensional photoacoustic handheld probe consisting of a spherical matrix array of 256 ultrasound sensors each with a central frequency of 4 MHz [48]. In healthy volunteers of ages 25 and 40 whereby the breast tissue is known to contain dense fibroglandular tissue that will make tumor diagnosis difficult on X-ray mammogram, Deán-Ben et al. have shown volumetric views of the breast vasculature and surrounding parenchyma at a single wavelength (800 nm) excitation up to a depth of 1 cm. In addition, by using multiple wavelengths, such as 730, 760 and 850 nm from a tunable optical parametric oscillator (OPO) laser for tissue excitation, they have offered high resolution volumetric maps of blood oxygenation parameters like HbO 2 , Hb and total hemoglobin (HbT) [49]. By offering PA images at a high resolution of 200 μm, this study is the first of its kind to help look for angiogenic biomarkers and regions with high microvessel density which are the characteristics of breast cancer growth and invasion [50,51].
Interestingly, PAI's capacity to derive oxygenation status of blood vessels has been demonstrated to aid largely in assessing anti-cancer treatment driven changes in the blood vessels [52]. Following chemotherapy with taxane, the relative oxygen saturation in peritumoral blood vessels was observed to be low indicating hypoxia (Fig. 2). At the same time, finer intratumoral blood vessels became noticeable suggesting increased blood perfusion possibly due to the chemotherapy associated lowering of interstitial blood pressure leading to normalization of blood vessels. Apparently, all these finer changes happened without any change in the tumor size as observed by ultrasound, thus reinstating the potential advantages that PAI can bring to the clinics. In the same study, remarkable differences were noticed in the shape and the number of blood vessels between the diseased and contralateral breast, opening the avenue for analysis of finer blood vessels to diagnose early breast cancer. Moreover, the observation of centralizing vasculature signal tending towards the tumor in 61 % of invasive breast carcinoma (IBC) cases as opposed to only 35 % of ductal carcinoma in situ (DCIS) cases may eventually help in distinguishing IBC from DCIS lesions. The authors were able to provide more comprehensive vascular structural features with depth profile unlike previous studies mainly by exploiting the technological advances. The authors addressed the welldocumented drawbacks of limited view associated with planar configuration of transducer arrays [11] by using hemispherical array and applying the breast misshapement algorithm from the MR breast shape image to that in a PA image, from which functional information about peritumoral and intratumoral vasculatures were obtained. In addition there have been a few multicenter studies with the first CE marked photoacoustic breast imaging system, Imagio by Seno Medical Instruments demonstrating the ability of ultrasound guided photoacoustic imaging to rightly diagnose benign and malignant breast masses [53] based on the relative amount of HbO 2 and Hb and further downgrade benign masses classified as Breast Imaging Reporting and Data System (BI-RADS) 4a and 4b to BI-RADS 3 and 2 thus potentially decreasing benign breast biopsies [54].
The capacity of PAI to distinguish between not just the different forms of hemoglobin but between many intrinsic chromophores like melanin, lipids and water was realized with the introduction of fast tunable OPO lasers [55]. As human breasts are composed mainly of vasculatures, fats and water, it is of benefit not just to look at the angiography of the breast but also to look at the distribution of fats within the breasts and in the tumors. In a pilot study, Diot et al. used a handheld photoacoustic probe with extended spectral range spanning 28 wavelengths from 700 nm to 970 nm to compute the contributions of Hb and HbO 2 , total blood volume (TBV), lipid and water and to derive multispectral optoacoustic tomography (MSOT) patterns in normal and cancerous breasts of various stages [56]. Angiogenesis, the characteristic hallmark of cancer was visualized as increased peripheral vascularization on Hb, HbO 2 and TBV images and decreased intratumoral vascularity, mainly because the intratumoral vessels were too small to be detected by the transducer. Tumor areas appeared to have heterogenous TBV patterns with rim measuring more than 30-fold compared to the neighboring non-cancerous regions and presented as disruptions in fat and water layers in the place of tumor as the tumors were replacing the original fat and water layers of the breast. Additionally, skin infiltration of breast tumor presented itself as a disruption of melanin and skin vascularization. Understanding these breast cancer features of MSOT could potentially guide surgery and overall therapy planning.
In spite of the compelling demonstrations of clinical utility of PAI in breast cancer as discussed above, there are several avenues for development addressing which the clinical applicability can be actualized. For instance, PAI has been demonstrated to be sensitive to hormonerelated physiological changes of breast parenchyma [57] like MRI [58], elastography [59] and near-infrared spectroscopy measurements [60] which will likely have an impact on the contrast of the images. By understanding and characterizing the normal ranges of breast vascularity parameters in pre-and post-menopausal women and at different phases of menstrual cycle, quantitative cut-offs for differentiating between benign and malignant lesions could be established. Also, as breast imaging is largely affected by motion artefacts arising from the movements of heart and lungs, continued efforts are taken to correct for breathing motion through post processing algorithms [52]. As such, shorter imaging times of the scanner will be an added advantage. When most of the mammoscopes described above take a few minutes for imaging which itself is good enough to induce motion artefacts, a single breath hold photoacoustic computed tomography system (SBH-PACT) developed by Lin et al. [23] will be a substantial alternative to address this. SBH-PACT could reveal the angiographic morphology from the nipple to the chest wall in a healthy 27 year old woman in a single breath hold's time of 15 s. Addressing the limited view drawback [11] by using ring detectors that can be switched between 2D and 3D modes, eight out of nine breast cancer lesions were accurately detected using high spatial (255 μm) and temporal resolution images. However, in order to keep the acquisition time short, single wavelength imaging (1064 nm) was used, making it difficult to obtain oxygenation information of tumors.
Ex vivo imaging
For early stage breast cancers, breast conserving surgery (BCS) or lumpectomy in conjunction with radiation therapy is emerging as a popular and efficient alternative to total mastectomy [61]. However, 20-70% of BCS patients undergo more than one surgical procedure before achieving tumor free margins [62,63]. Current standard of care for verifying tumor free margins includes histopathology and frozen sections [64,65] which suffer from long procedure times, on-table ultrasound [66] which suffers from operator dependent variability and xray mammogram [67], which has shorter processing time but reduced sensitivity and specificity due to poor contrast in denser breast tissues as mentioned earlier and the lack of functional specificity. When optical technologies like diffuse reflectance spectroscopy [68], optical coherence tomography [69] and Raman spectroscopy [70] are contributing to improve the sensitivity and specificity of tumor margin detection, they are still limited by long processing times, limited penetration depth and FOV. PAI with proven capability in rapid-deep tissue biochemical imaging [55] especially for in vivo breast cancer imaging (as discussed earlier) is emerging as a technique of immense interest for ex vivo breast specimen analysis too, particularly for assessing the margins. Interestingly, all clinical studies in this space have used ultrasound guided photoacoustic tomography systems.
In one of the pioneering studies, Li et al. used multiple wavelengths spanning from 1100 to 1250 nm in the second optical window at a step size of 10 nm to excite the excised breast specimen fixed in 10 % buffered formalin and obtained the distribution of hemoglobin and fat [71]. By classifying fatty tissue with scattered connective tissue as normal tissue and the rest as cancerous, assessment of 12 specimens fixed in 10 % buffered formalin using PAI resulted in 100 % sensitivity and 75 % specificity in assessing the tumor margins. The reduced specificity was mainly attributed to the dense connective tissue being picked up as cancer. This could be improved by using strict classifications to identify the two which could be achieved by looking further into their mechanical properties through frequency differences between the two. As an extension of this concept, Li et al. engineered an intraoperative multimodal ultrasound and PAT system that made use of overtone absorption of fat at 1197 nm as contrast [72]. The usage of single wavelength greatly reduced the scan time to 2 min covering an area of 10 × 10 cm 2 breast specimen. In a clinical study involving breast cancer specimen from 66 patients, this system was observed to assess tumor margins with a sensitivity of 85.5 % and specificity of 90 % through machine learning algorithms.
In a pilot exploratory study performed by Goh et al., they made use of the differences in the absorption of hemoglobin and fat in the wavelengths ranging from 700−1100 nm to image their distribution in freshly excised breast specimens [17]. Tumors being hallmarked by angiogenesis, intense Hb signals were observed on all their edges. Samples with negative margins were noted to have continuous intense lipid layers on all edges of the tumor and those with positive margins were noted to have discontinuous lipid layers as observed in the in vivo PAI imaging of breast cancer by Diot et al. [56]. Based on these classifications, an assessment of 15 fresh specimens containing 90 margins in total (6 margins per sample) revealed a sensitivity of 100 % and a specificity of 97.6 % (in press). The reduced specificity was attributed to two of the negative margins being picked up as positive margins. Interestingly, the two incorrectly identified margins had close margins of < 0.5 mm as identified by histology. As no tumor on ink for invasive ductal carcinoma (IDC) and 2 mm for DCIS are the acceptable surgical margins according to the Society of Surgical Oncology [73,74], it can be considered that PAI was observed to have 100 % specificity as well.
Alternatively, an OR-PAM system operating with a laser in the UV range (266 nm), termed UV-PAM, has been demonstrated to offer histology-like imaging of fixed but unprocessed human breast specimen [75]. Diagnostic criteria such as nuclear size, internuclear distance and packing density make UV-PAM very promising in the detection of cancer cell clusters at the cancer specimen margins. UV-PAM in the published version would take 7 h for imaging 1 cm x 1 cm at a spatial resolution comparable to high-NA optical microscopes. However, by improving the optics in UV-PAM, it is possible to scan a BCS specimen with an average size of 5 cm diameter in 4 min.
Dermatologic imaging
The human skin, being easily accessible, makes it an ideal location for non-invasive imaging. Skin disorders are usually diagnosed by visual inspection and occasionally confirmed by skin biopsies and following histopathology. While this method remains the gold standard for skin disorder diagnosis, they are invasive and unaesthetically pleasing, especially on easily scarred Asian skin. Non-invasive imaging can potentially negate the need for skin biopsies with faster diagnosis times, and can benefit the treatment management of the disorder. However, the skin architecture is complex and can be regarded as a multilayer optical medium and this should be reflected and captured by these imaging modalities. Photoacoustic imaging in the mesoscopy scale allows for the different layers of the skin to be discerned [73]. RSOM imaging of a volunteer's palm revealed the different skin layers, the uppermost layer making up epidermis of about 200 μm thickness followed by the epidermis-dermis junction [76]. It is also capable of visualizing vessels from capillary loops near the epidermis-dermis junctions, down to the arterioles and venules in the vascular plexus as described above. Additionally, the spectral capabilities conferred by MSOM imaging can differentiate between melanin, HbO 2 and Hb [77]. The absorption maps of these tissue components can be reconstructed in volumetric spatial maps with the superficial epidermal layer distinguished by strong absorption of melanin and the vascular plexus underneath consisting of deoxy-and oxy-hemoglobin maps. The sO 2 level of the vascular network was calculated at different depths and found to be 85 ± 4 % at the skin surface, and 54 ± 7 % of a deeper vein in the dermis. One limitation of RSOM/MSOM imaging is that it is sensitive to the subject's motion (e.g. breathing) due to the focality of the ultrasound detector relative to the subject's skin. To overcome this, a motion correction algorithm was reported that first depended on the segmentation of the skin's melanin [78] which subsequently evolved into an automatic algorithm by taking into account the vertical shifts of the skin at various body regions [79]. While the RSOM/MSOM and PAM systems allow for high-resolution visualization of vessels abnormalities; a common indicator for skin diseases; tomographic PAI systems will be more useful for larger and deeper skin structures such as the delineation of skin cancer.
Pilosebaceous units
MSOT has been exploited for the imaging of pilosebaceous units whereby the hair follicle structure was elucidated including the hair shaft, the dermal papilla and the sebaceous glands in the vicinity as seen in Fig. 3 [80]. By imaging at different wavelengths, the sO 2 levels of the capillary bed encompassing the dermal papilla was determined to be close to 99 % for a healthy hair follicle. Imaging of pilosebaceous units is useful for the diagnosis and understanding of pilosebaceous disorders such as alopecia and acne. This is also highly significant in the context of developing hair care products and hence might generate tremendous interest among multinational giants in the personal care products industry.
Skin cancer
The number of skin cancer cases worldwide has risen noticeably in recent years. Skin cancer can be categorized into three main types; basal cell carcinoma, squamous cell carcinoma, and melanoma, in which the first two types are collectively called non-melanoma skin cancers and are seldom life threatening. Melanoma skin cancers are rare but they are potentially aggressive and usually start in pigmented lesions such as a mole or birthmark. Surgical intervention of these lesions is the gold standard for treatment which has evolved towards tissue sparing for better aesthetic results, namely, Mohs micrographic surgery (MMS) whereby the cancer is removed in layers until no cancer remains. Since MMS is time-consuming and requires special expertise and training, it can benefit from pre-operative 3D imaging of the skin cancer's size and depth for exactly defined excision with improved clearance, lower relapse rates and shorter surgery time. Skin-directed imaging tools can also enhance the management of skin cancers when aggressive tumors are detected subclinically.
Label-free in situ PAI has been exploited to measure the size of cutaneous cancers due to the functional contrast of these lesions. Several studies using multispectral PAI configurations equipped with handheld transducer probes were demonstrated to measure the depth of in situ pigmented skin lesions which were validated with histological measurements after the lesions were excised or undergone MMS [81][82][83][84][85][86]. The handheld implementation makes it flexible to access the skin cancer lesions which usually appear on sun-exposed areas such as the face and neck. As Chua et al. reported, when the MSOT 3D map of non-melanoma skin cancers were acquired, lesions margins could be visibly differentiated from the soft tissue in the vicinity due to the presence of melanin chromophore (Fig. 4). The length and the deepest infiltration dimensions of melanin signals were measured and they show a high correlation with the length and depth of histological resections of the tumors via a linear regressive fit (r = 0.9) [82]. Generally, lesion dimensions measured by PAI are overestimated compared to histological measurements obtained after excisional surgery, probably due to the dehydration of the tissue samples in histology preparation. Notably, most non-melanoma skin cancers appear pigmented in people of color while they appear erythematous on patients with lower Fitzpatrick phototypes. For example, erythematous lesions were shown as a mass of HbO 2 signals with no melanin signals present (Fig. 4). Therefore, the spectral imaging ability of these systems in differentiating signatures of hemoglobin from melanin allow both pigmented and non-pigmented cutaneous lesions to be mapped with accurate lesion dimensions [82].
As mentioned earlier, melanoma skin cancers can spread throughout the body and the first site of metastasis is the nearby lymph nodes. Identification of metastasis in the lymph nodes is imperative in accessing the extent of the disease. Both in vivo and ex vivo studies examining the capability of multispectral PAI to detect melanoma metastasis in human lymph nodes have been reported [87][88][89][90]. Stoffels et al. reported that MSOT imaging improved the accuracy of tumor metastasis rate in sentinel lymph node biopsies compared with the established standard protocols (22.9 % versus 14.2 %) [88]. To achieve in vivo imaging of the lymph nodes, the exogenous contrast agent ICG was Fig. 3. PA orthogonal images and 3D projection of a single pilosebaceous unit imaged on the forehead. Unmixed PAI signals show the spatial maps of HbO 2 (red), Hb (blue), lipid (green) and melanin (yellow), revealing an ellipsoid-like lipid structure near the hair shaft. Notably, the sO 2 of the capillary bed feeding the follicle was highly oxygenated. Scale bar; 3 mm. Reprinted with permission from Ref. [80].
administered peritumorally into patients with melanoma. The dye was visualized in the sentinel lymph node up to a depth of ∼10 mm beneath the skin with a 100 % agreement rate with the conventional radioactive tracing and pathological analysis methods.
These reported studies exhibit how PAI can be exploited to preoperatively map the skin cancers to aid the surgeons in planning their excision surgery to save time and increase the accuracy of lesion margin removal. However, the challenges for PAI in skin cancer detection remain as the system's detection limit of the imaging depth is only about 10 mm [82,83] and thus the technical design of the handheld probes have to be modified to address this pitfall. Furthermore, 3D mapping of the skin lesions are sensitive to motion so acquisition with artefacts have to be discarded. Future studies to validate the use of PAI in delineating skin cancer lesions would be using the pre-operatively determined tumor dimensions in the surgical planning process and judging the extent of cancer clearance.
Inflammatory skin diseases
Inflammation is a systematic disease and can be manifested in macrovascular and microvascular disorders. While the former can be evaluated through sonography, pathological changes in peripheral small vasculatures can be visualized using PAI by virtue of the absorption of Hb and HbO 2 . Therefore, inflammatory systemic diseases manifested in dermatologic conditions such as psoriasis and eczema can be possibly imaged and monitored by PAI. Aguirre et al. reported the use of single wavelength RSOM in a pilot study to visualize skin architecture and vascular features in psoriatic skin, permitting the identification and quantification of inflammation landmarks [91]. Compared to healthy skin, psoriatic skin displayed increased epidermal thickness, elongated and dilated capillary loops and heightened dermal vascularization (larger diameter, denser in square area and more tortuous) which was corroborated with histological examination as shown in Fig. 5. From these objective measurements of the inflammation metrics, a calculated index of PA features were conceived which correlated well with the clinical Psoriasis Area Severity Index (PASI), the latter of which do not reflect the structural characteristics of the skin.
The vascular abnormalities seen in eczema patients are similar as reported by Zabihian et al. whereby a high-resolution PAM system integrated with optical coherence tomography (OCT) system was used [92]. Hypervascularization was observed in a subject with chronic hyperkeratotic hand eczema whereby the capillary loops were closer together and capillaries were present between the curves of the loops. In a subject with dyshidrotic hand eczema where scar tissue was existent, the vessels in the vascular plexus were narrower at the border between the scar tissue and healthy skin [92]. Evidently, inflamed skin lesions exhibit thickened epidermis and vascular patterns with dilated vessels arranged in a disordered network. The acanthosis of the skin lesions may cause high scattering within the tissues and may limit the imaging depth of OCT, diminishing the visibility of dermal structural features. Another inflammation indicator could potentially be the oxygenation status of the vessels measured via PAI as reported by Ishida et al. [93]. After a week post-treatment with ixekizumab of a patient with psoriasis on extremity digits, the oxygenation saturation level increased from 72 % to 89 % in the palmar digital veins. This increase in oxygenation was attributed to the lower oxygen consumption in the tissues as inflammation decreases, thus increasing the oxygenation saturation level in the vessels.
Vascular imaging in extremities
PAI's label-free sensitivity to both HbO 2 and Hb makes it an attractive modality to study and visualize vasculatures in all superficial organs. The different configurations of PAI systems have been utilized to grant in vivo structural and functional information (e.g. sO 2 ) in human skin. Mesoscopic and microscopic configurations are suited for imaging cutaneous microvascular network down to individual vessels, as superficial vasculature can act as a site for surrogate markers of peripheral vascular diseases described below. Other modalities exploited for quantifying peripheral vasculature include magnetic resonance angiography (MRA) and MRI. However, these techniques require the use of external contrast agents and workup which may limit their clinical relevance. Laser speckle sensing has also been adopted to measure blood flow but being single-wavelength, it cannot quantify the tissue oxygenation level nor image the vasculature architecture. Meanwhile, the tomographic configuration of PAI in hand-held mode can be useful to image and interrogate specific larger and deeper blood vessels. Feasibility studies utilizing both volumetric 3D [48,94,95] and 2D array [96] hand-held probe designs to image human vasculature has been demonstrated. Taruttis et al. demonstrated the use of multispectral optoacoustic tomography (MSOT, iThera Medical GmbH) equipped with a handheld detector with a curved geometry (135°coverage) which was capable of penetrating 10 mm deep into the tissue to image major blood vessels and microvasculature [97]. Large blood vessels in the millimeter-scale such as tibialis posterior and dorsalis pedis arteries in the leg were visible from the contrast attributed to HbO 2 signals. MSOT was also shown to resolve microvasculature of diameter 90−110 μm on the medial and distal big toe. The arterial pulse was also observed in real-time MSOT visualization as each laser pulse generated an image frame.
Cutaneous microvasculature
The clinical studies described in this section were acquired on healthy subjects to judge the feasibility of PAI as an imaging tool to visualize microvasculature for potential disease applications. AR-PAM has been demonstrated to show the microvascular response on the palm to intermittent arterial occlusions by measuring the relative changes in oxygen saturation level and capillary perfusion [98]. A decrease in PA signals was observed during an occlusion after releasing which there Fig. 5. RSOM imaging of healthy skin and adjacent psoriatic lesion with histology validation. RSOM cross-sectional images, clinical images and histological images of (a, c, e) psoriatic skin and (b, d, f) adjacent healthy skin. Psoriatic skin exhibited elongated capillary loops near to the skin surface, thicker epidermis, absence of melanin and increased vascularization of the dermis which is validated by the histology images. Scale bars; 200 μm. Reprinted with permission from Ref. [91].
was an increase, indicating a reduction and improvement in perfusion during and after ischemia respectively. The same change was reflected in the relative Hb concentration as well following occlusion and release. PAM imaging has also revealed that the microvasculature structure differs in between acral and non-acral areas [99]. Compared to the microvasculature in the palm, strong PA signals were observed in the epidermal-dermal junctions of the forearm. This was attributed to the denser network of perfused capillaries in the forearm, creating stronger and broader PA signals.
The mesoscopy configuration of PAI systems can penetrate beyond the optical diffusion limit and image over a bigger volumetric space in the skin. The ultra-wideband transducer in the RSOM system makes it capable of resolving the microvascular structure of the skin that includes small vessels from capillary loops near the epidermis-dermis junctions and bigger arterioles and venules in the vascular plexusin a label-free mode [100]. The high frequency band during reconstruction revealed the capillary loops extending from the lower plexus, represented as 'dot'-like structures. Capillary loops are situated below the junction whereby only the apex of the loops are visualized due to the directivity of the photoacoustic signals in the axial direction. From the low frequency band, the bigger vascular plexus forming the lower horizontal plexus can be visualized. The RSOM system when fitted with per-pulse tunable 100 Hz laser with a range of 460-650 nm is capable of spectral measurements to resolve the different main tissue absorbers i.e. melanin, Hb and HbO 2 [77]. This particular setup is termed Multispectral Optoacoustic Mesoscopy (MSOM). This also granted additional spectroscopic data to obtain the sO 2 values in single blood vessels, measuring about 85 ± 4 % in superficial vessels and about 54 ± 7 % in a draining vein in the dermis of the forearm. These values can be corroborated by other optical measurements such as reflectance spectroscopy and near-infrared spectroscopy of the bulk blood sO 2 values of the forearm. The RSOM system has also been reported to resolve the microvasculature on the nailfold of healthy subjects, even in cases whereby the epidermis thickness is beyond the penetration depth of a standard clinical optical microscope [101]. Studying the nailfold microvasculature is useful to assess the early diagnosis of systemic sclerosis or scleroderma. The capillary density and maximal capillary diameter of the nailfold microvascular architecture were both quantified to derive objective parameters to aid clinicians in diagnosing systemic sclerosis, by observing avascular areas and atypically large capillaries. The quantified parameters as measured by RSOM was comparable to that measured by the optical microscope. Given RSOM'S capabilities to resolve the different layers in the skin morphology, the skin microvasculature response to local temperature rise could also be investigated [102]. Upon local heating of both volar and dorsal forearm, vascular density was observed with the RSOM signals in the deeper depths in the lower frequencies (bigger structures), signifying local vasodilation. Quantitation of the vasodilation revealed more than 1.7 times blood volume change and an increase in vessel diameter by 63 %.
PAI may also be valuable in the imaging of perforator vessels in anterolateral thigh flaps, usually used in the reconstruction surgery of the head and neck. In investigating this hypothesis, Tsuge et al. performed in vivo imaging of anterolateral thigh perforators and their furcate architecture in the subcutaneous tissue of healthy subjects, using a PAI system with a hemispherical detector array that acquires data in a spiral pattern within a plane [103]. The microvessels in the subcutaneous layer, particularly those in slanting or horizontal alignment, were visualized up to a depth of 13 mm while the PAI system demonstrated a limitation in visualizing subfascial vessels. Mapping of the vasculature in 3D revealed the branching architecture of perforators, suggesting the potential of PAI to be used as an intraoperative imaging modality during thin anterolateral thigh flap surgery.
These above-mentioned studies mainly exploited the endogenous tissue contrast to study the vascular architecture and blood oxygenation. In contrast, Lutzweiler et al. demonstrated PAI's capability to measure the dynamic kinetics and clearance of an externally-administered contrast agent in a volunteer's finger [104]. ICG was injected into superficial vein of the volunteer's arm. A customized curved detection array with an angular coverage of 266°was utilized in the study to fit a finger and to acquire 10 cross-sectional images of the finger per second. Vasculature on the palmar side of the finger could be visualized through endogenous contrast and by monitoring the ICG PA signals in these vessels over time, ICG's kinetic data could be fitted into a dual exponential function to derive the circulation and decay time of the contrast agent.
Vascular dysfunction in diseases
PAI's ability to resolve larger blood vessels and microvasculature makes it an attractive modality to investigate diseases whereby vascular dysfunction or remodeling is a marker for disease activity and progression. By studying the markers, an improved understanding of the diseases can be achieved to provide better management of the disease for the patients and healthcare providers, including improved disease stratification and more efficient assessments and follow-up. One such disease as mentioned above is systemic sclerosis. Systemic sclerosis is an autoimmune disease of the connective tissue, typified by skin and internal organs fibrosis and progressive vasculopathy. Masthoff et al. evaluated the extent of microvascular dysfunction in systemic sclerosis patients compared to healthy individuals with the use of MSOT equipped with a handheld detector with a center frequency of 4 MHz, offering 2D cross-sectional images with a FOV of 30 × 30 mm 2 and 100 μm reconstructed pixel size [105]. In comparison to healthy individuals, the subcutaneous finger tissue of systemic sclerosis patients exhibited considerably lower HbO 2 and HbT MSOT values (∼ 50 %). When systemic sclerosis patients of stable prognosis were compared to their counterparts with progressive prognosis, the latter exhibited lower Hb, HbO 2 and HbT MSOT values. This proof-of-concept study provided evidence of inter-individual variation of the disease and how non-invasive longitudinal monitoring of these patients is essential for individualized treatment plans. Raynaud's phenomenon is the most common manifestation of systemic sclerosis and it is characterized by spasms of arteries, thus impeding blood flow and it is exacerbated by stress or the cold. This medical condition has been similarly investigated via PAI by Eisenbrey et al. to enumerate tissue oxygenation in the nail bed and nailfold as an avenue to diagnose and assess Raynaud's phenomenon [106]. PAI was performed on nail beds of patients with Raynaud's associated systemic sclerosis along with healthy subjects using a Vevo 2100 LAZR system equipped with an LZ-250 probe in oxy-hemoglobin quantification mode at baseline and upon cold stimuli. While the number of vessels imaged in the FOV were the same between the disease and control groups, Raynaud's patients exhibited a more distinct decrease in sO 2 levels of the vessels upon cold stimuli with time compared to the healthy subjects. The authors maintained that measuring the oxygenation status of the vessels upon cold stimulus is more useful in the diagnosis of Reynaud's phenomenon.
PAI has also been exploited to quantify imaging features for the diagnosis and management of vascular malformations [107]. Vascular malformation refers to congenital vascular abnormalities of a combination tangle of veins, arteries and lymph vessels or individually. MSOT imaging of venous malformations and arteriovenous malformations revealed that they exhibit higher HbO 2 :Hb ratios by approximately 40 % compared with healthy tissue with arteriovenous malformations exhibited significantly higher HbO 2 :Hb ratios compared with venous malformations. When 3 patients presented with arteriovenous malformations underwent embolization treatment (i.e. obstructing a blood vessel), the HbO 2 :Hb ratios decreased significantly but was still higher than that of healthy subjects. Unsurprisingly, partial embolization treatment only resulted in a conservative reduction in HbO 2 :Hb ratios. Venous malformations are however treated with drugs to shrink the malformations, i.e. percutaneous sclerotherapy. Similarly, the HbO 2 :Hb ratio of the venous malformations in 2 patients decreased upon successful sclerotherapy. This study signifies that PAI is a good modality in facilitating the intervention management of patients with vascular malformations.
Wound imaging
Chronic wounds are wounds that exhibit little or no healing over extended periods of time. Although the precise mechanism behind the prevention of wound healing is unknown, it is thought that a local loss of oxygenation is one of the causes. Therefore, sO 2 levels of the wound is potentially one of the parameters in studying chronic wounds. As multi-wavelength PAI is able to measure sO 2 level, it can be exploited for wound imaging to provide label-free 3D information about the oxygenation at ultrasound resolutions. Petri et al. first demonstrated the use of PAI in assessing chronic wounds, in a study whereby 5 patients were imaged with PAI at baseline and post-application of topical hemoglobin spray onto their chronic leg ulcers [108]. Topical hemoglobin spray is a novel therapeutic intervention to increase the oxygenation in the ulcerated areas by performing as a transporter for ambient oxygen to the wound in situ. PAI imaging showed that the spray increased the oxygenation of the ulcers on the surface and in the deeper ulcerated tissue. In a subsequent study with 49 patients, about half of the patients were identified as non-responders to the hemoglobin spray therapy with a less than 10 % increase in oxygen saturation level, indicating that their leg ulcers were incapable of self-healing and therefore necessitate a removal of the sclerotic tissue by surgery [109]. This shows the potential of PAI as an adjunctive diagnostic tool with a treatment intervention in appraising chronic wounds and tailoring wound care management to the individual patient.
Carotid vessel imaging
Carotid artery disease is the most common form of heart disease, with stroke being the risk factor. Carotid duplex ultrasonography (DUS) is frequently employed in clinical practice to screen for any blockage or narrowing of the carotid arteries commonly caused by atherosclerotic plaques. DUS grants information on the plaque's location, calcification presence and mobility characteristics without the use of any contrast agents or ionizing radiation. However, it is unable to dispense information on the plaque's composition and structure. Because PAI bestows a comprehensive platform for structural and functional imaging, it is advantageous for cardiovascular imaging. Herein, we will concentrate on non-invasive PAI of the carotid artery [110]. The use of intravascular PA (IVPA) imaging, a catheter-based approach delivering a direct visualization of human coronary arteries [111][112][113] is beyond the scope of our review and will not be discussed herein. Its clinical use is still inundated with challenges and is for now restricted to visualizing ex vivo human carotid artery tissues. Efforts to emulate non-invasive PAI of diseased human carotid artery have been reported whereby human carotid the ex vivo artery tissue was embedded in a neck mimicking phantom and imaged [114].
In one of the first demonstrations of non-invasive PAI of the carotid artery, Dima et al. designed both a linear and curved ultrasound array probe for a PAI system tailor-made for carotid imaging [115]. The former array performed at 5-7 MHz frequency with 128 elements while the latter array was designed for central frequency of 5 MHz. The curved array demonstrated better imaging performance, giving a more thorough tomographic visualizations, with the vessels imaged showing superior likeness to the tissues. When the curved array was placed on the lower right neck of a healthy subject, the right common carotid was visualized, up to 18 mm under the surface of the skin. Nearby, the right internal jugular vein was identified along with the external jugular vein. The PAI images were corroborated by DUS which was acquired at the same location on the neck.
Merčep et al. undertook a hybrid MSOT, pulse-echo US and color Doppler imaging approach via a multi-segment detector array that simultaneously offers blood flow measurements and sO 2 levels in the carotid artery region of a healthy subject [116,117]. Vascular structures appear in the MSOT images due to the hemoglobin contrast while they appear hyperechoic in the US images. Images acquired revealed a distinction between the sO 2 levels of different vascular structures occupying at least 1−2 cm beneath the skin surface in the carotid space. As expected, the MSOT HbO 2 signals were higher in the common carotid artery, while more Hb signals were present in the superficial microvasculature as well as the jugular vein. The blood flow of the common carotid artery was measured between the regular values of 30 and 40 cm/s as captured by color Doppler imaging. This multi-modal approach is advantageous as the complementary information imparted can aid in clinical decisions for an informed judgement of the carotid artery disease severity.
Musculoskeletal imaging
PAI's capacity to provide high resolution optical contrast in soft tissues has been explored to map the morphology of cartilage, synovium, vascularity and bone tissue in human joints, particularly in inflammatory arthritis [118]. Inflammatory arthritis is a group of diseases whereby inflammation is present in the joints and often other tissues. Due to the inflammation, extreme hyperemia of the joint is observed from the hypervascularization and dilation of veins and capillaries present there. As the metabolic state of the inflamed tissue is apparent in its physiological states, i.e. blood volume, and blood oxygenation, PAI can be exploited to detect and study the inflamed tissue. A portable PA/US imaging system equipped with a hand-held linear probe emitting a single wavelength was first proposed by van den Berg et al. for the imaging of clinically evident synovitis in rheumatoid arthritis patients and control healthy subjects [119]. A superficial blood vessel was observed in both the inflamed and non-inflamed joints with more PA signals and more convergent PA features observed underneath the former. When the amplitude PA signals were quantified, the signals in the inflamed joints were 4 to 10-fold higher as compared to those from control groups. This feasibility study has demonstrated that PAI is sensitive to clinically evident synovitis in inflamed joints and should be further explored.
Jo et al. reported a PA and US dual-modality imaging system equipped with a linear probe and tunable dye laser to identify inflamed finger joints in patients and by quantifying the hyperemia and hypoxia associated with inflammation in their finger joints in relation to that of healthy subjects [120]. As the PA and US images were acquired simultaneously, the functional information of hyperemia obtained from PA could be co-registered with the US morphological tissue structures. The PA images revealed that the hyperemia present in the finger joints of the patients were visualized as a sheet or blob of blood while the hemoglobin signals observed were 66 times lower in the synovium of the normal subjects' joints. Furthermore, the measured sO2 levels were lower in the arthritic joints in relation to the normal joints (0.582 ± 0.034 vs. 0.651 ± 0.020). This reported findings suggested that PAI, as a complement to US, is capable of investigating additional in vivo physiological biomarkers of inflammatory arthritis.
Gastrointestinal imaging
Under acute conditions of the clinical phenotypes of inflammatory bowel diseases i.e. Crohn's disease (CD) and ulcerative colitis, the care management entails constant adjustment of therapeutic interventions to avoid severe irreversible complications. Like many other inflammatory disorders, CD also presents with poor correlation between intestinal inflammation with standard clinical scoring system and laboratory assessment, which demands for a more accurate objective assessment of the risk for relapses and associated complications. Existing colonoscopy based approach is time consuming and also associated with multiple levels of risk. In this context, hand-held MSOT system can be exploited to extract functional information in the colon as a rapid and noninvasive approach for the objective assessment of CD's severity [121,122].
In a pioneering study conducted in 108 patients with varying degrees of CD severity, a clinical MSOT Acuity system fitted with a handheld detector (3−4 MHz, 256 transducer elements) acquired images from abdomen to image deeper organs and to monitor the changes in colon upon inflammation [121]. This study demonstrated a major distinction in single wavelength acoustic data at 760 nm and HbT signals between patients who show no inflammation in the intestines via scope and those with low grade inflammation. Furthermore, HbO 2 and sO2 values increased progressively in correlation with disease severity as shown in Fig. 6 [122]. Quantified parameters from MSOT images were also compared between patients with active disease and those in remission, which was established from clinical scoring, endoscopic scoring and histologic data [121,122]. The capability of MSOT to spatially map the vascular Hb and HbO 2 absorbers as markers for perfusion and inflammation in the intestinal walls with CD, offered functional information to objectively assess and to evaluate the level of inflammation in a reliable and noninvasive way.
Adipose tissue imaging
The critical role of adipose tissue in understanding metabolic disorders has recently been reported. There are two forms of adipose tissue: white adipose tissue (WAT) and brown adipose tissue (BAT). WAT acts as 'reservoir' for energy storage containing triglycerides, while BAT is specialized for thermogenesis to burn excess calories and produces heat mainly through the uncoupling reaction facilitated by uncoupling protein-1 (Ucp1). Over the last few years, there is a renewed spark in understanding these mechanisms and also the recently discovered "browning" process that involves cold induction or Ucp1 activation resulting in "brown-like" or "beige" adipocytes dispersed inside the major population of WAT but exhibiting genetic and biological characteristics of BAT, [123,124].
Reber et al. utilized MSOT system fitted with hand-held 2D probe for noninvasive, label-free imaging of BAT in humans and also to resolve BAT activation based on hemoglobin gradients, which is in accordance with reported studies employing positron emission tomography (PET) [125]. Images from the BAT in the neck and the supraclavicular fossa of three subjects were acquired. Since the oblique resolution of MSOT system is higher than that of MRI and PET, a detailed view of the triangular muscle tissue could be observed just underneath the skin, corresponding to the other modalities, a seen as in Fig. 7. Upon BAT activation by cold exposure, the HbO 2 signals in the BAT region increased considerably by 3-fold while no meaningful signal increase was observed in the muscle region (Fig. 7). The PA signal in the muscle exhibited a minor change of 1 % ± 2 % as compared to that of BAT region (372 % ± 105 %). The difference in lipid and TBV in the WAT and BAT were also observed in ten volunteers [125]. Resolving the spectral signatures of both regions revealed a precise difference in the MSOT spectra between BAT and WAT, with the former exhibiting 4fold higher MSOT signal in the 700−900 nm region. Strong MSOT signal at 930 nm observed in both BAT and WAT region is due to the fat/lipids content which is confirmed by the absence of such signal in muscle. This study was instrumental in understanding the fundamentals of fat tissue metabolism and could be very well translated for understanding the role of adipose tissue in controlling obesity and diabetes. However, standardization of the MSOT approach with well-established MRI study protocol is necessary so that better image co-registration could be realized for more accurate comparison between water and lipid (fat fraction in MRI) signals.
In another study, Buehler et al. used a similar MSOT system with 2D hand-held probe to image subcutaneous lipomas (fatty tumors) [126]. They imaged six volunteers with lipomas located at different areas and Identification of the intestinal wall was done using B-mode ultrasound image. It was observed that in contrast with the patient in remission, clinical and endoscopic CD activity was associated with increased signals for HbO 2 and sO 2 in the intestinal wall. Reprinted with permission from Ref. [122].
correlated the results with clinical diagnostic US. MSOT images acquired at single wavelengths revealed better contrast in visualizing subcutaneous lipomas than that of US, particularly at wavelengths where fat exhibits minimum light absorption and hemoglobin absorbs the most. Spectral un-mixing showed that lipomas exhibit characteristic differences in signature between muscle tissues and normal tissues primarily due to the difference in absorption profiles of hemoglobin and lipid. This result is coherent with the fact that lipoma tissues are fatty and vascularized tissues while the primary absorbers in the muscle is blood and fat tissue per se contain less vessels than lipomas. The use of MSOT in conjunction with US for imaging lipomas can potentially enhance the diagnosis of subcutaneous soft tissue lesions due to the complementary images offered. Future studies could include developing MSOT imaging specific biomarkers based on spectral differences among other types of superficial soft tissue masses such as mesenchymal tumors, skin appendage, metastatic tumors (e.g. carcinoma, melanoma), tumor-like lesions (e.g. lymphoma, myxoma) and inflammatory lesions (cellulitis, abscesses).
Challenges and future directions
From this review, one can appreciate the extending list of organs and conditions that PAI helps to visualize. With breast cancer topping the list with CE marked Imagio™ and MSOT Acuity in markets for diagnosis, several novel clinical applications of PAI can be expected in the coming years with the advent of new custom proprietary PAI systems designed for clinical use. One such transpiring clinical application is in gynecology whereby multispectral PAI has been employed in pre-clinical studies to examine placental, fetal and maternal sO 2 levels in normal and pathologic pregnancies in relation to pregnancy-related diseases [127][128][129][130]. Recently, translation to the clinical stage was realized with a transvaginal fast-scanning photoacoustic endoscopy configuration which enabled high-resolution imaging of the microvasculature in the cervix down to capillary-size resolutions [131]. This system was tested in two pregnant subjects to observe the angiogenesis associated with cervical vascular remodeling during pregnancy. Parameters such as the microvessel density and total microvascular coverage area could be quantified from the images, with the former being the more reliable metric to quantify the extent of cervical remodeling. Having a transvaginal detector can potentially be used for in vivo ovarian imaging as well to detect and characterize ovarian cancer which have been demonstrated on ex vivo ovarian tissues previously [132,133]. However, several gaps need to be filled before such studies can be translated into clinical trials, including compliance to the laser safety limit, special design of transducer array to fit larger scanning areas, shortening of image acquisition time, development of large fieldof-view imaging systems and achieving better contrast in the region of interest (ROI) compared to other regions of less interest. A full suite of contrast agents, those clinically approved for other optical imaging modalities [10] and totally new formulations [134] continually being tested for photoacoustic activity can offer advantage in enhancing the in vivo contrast of ROI.
Another area of potential clinical interrogation is to study neural activities and cognitive functions in the brain. While extensive preclinical brain studies using PAI have been reported, clinical translation of PAI brain imaging is challenged by the light and ultrasound attenuation through the thick human skull. This setback was addressed by Nie et al.who reported the first demonstration of transcranial PAI through an intact cadaver human skull by incorporating a photon recycler in the PAI system to reflect back-scattered light back to the skull [135]. Advances in wavefront shaping techniques which are known to significantly enhance the in situ optical flux and thus the signal to noise ratio are recently gaining importance in improving the penetration depth of photoacoustic imaging, particularly through turbid medium like the skull [136][137][138][139]. This can potentially accelerate the clinical application of transcranial PA brain imaging, allowing the functional and pathological changes from diseases and different stimuli in the human brain to be observed.
While the clinical use of PAI has shown promising results in the pilot studies reported, multiple directions still exist for this imaging modality in its passage into clinical practice. Firstly, these studies are with a limited number of subjects, thus requiring larger cohort multicenter studies to establish the trends observed. Multicenter studies notably require high inter-operator and intra-operator reproducibility for the same PAI system utilized. While recent clinical studies have already started to evaluate these parameters [57,140], there is yet another aspect to seriously contemplate on. As many of these clinically tested PAI systems are prototypes from academic institutions and not of commercial quality yet, images from these systems considerably differ due to several factors such as laser power variability and image acquisition setup, thus requiring standardization across the spectrum of the systems. Efforts by Bohndiek et al. which are backed by volunteers from PAI community to successfully form a consortium termed International Photoacoustic Standardization Consortium (IPASC) to establish bestpractice guidelines for image acquisition, analysis and reporting and a standardized validation for technical systems will largely help to verify the robustness of different PAI systems [141].
Secondly, adapting PAI systems to be portable and economical [142] is the key to gain widespread adoption in the clinics and such attempts are in the early development stage. Instead of utilizing expensive OPO or tunable dye lasers as the illumination source, low cost light emitting diodes (LED) are being investigated as a substitute laser illumination source in PAI systems [143]. AcousticX (Cyberdyne Inc., Tokyo, Japan) is one such commercial LED-based PAI system. The system utilizes LED illumination array from both sides of a regular ultrasound transducer, comprising of 4 rows of 36 single LED elements to illuminate an area of 50 × 7 mm 2 . To acquire multispectral data, each pair of LED array illuminates alternatively at 690 nm and 850 nm (Fig. 8). LED arrays working at other wavelengths of illumination at 470, 620, 690, 750, 810, 850, 930 and 980 nm have also been developed [143]. The system can achieve a lateral resolution between 0.55 and 0.59 mm and an axial resolution of 0.268 mm [144]. The advantages of the LED-based PAI setup include low cost, small footprint and negating the need for laser calibration and laser safety goggles. However, the larger pulse width and low power of LED limiting the efficiency of generated acoustic signals and penetration depth respectively have to be tackled to further improve their output energy before the LED-based PAI systems could achieve similar imaging quality and depth as conventional laser-based PAI systems. Moreover, the impossibility of tuning LED make them unsuitable for photoacoustic spectroscopic applications.
Another method to make PAI portable is by using Fabry-Perot film US sensors in the place of piezo transducers as demonstrated by Beard et al. [145,146]. A fully-optical PAI system consisting of a fibre-coupled laser as the illumination source and a Fabry-Perot sensor for acoustic detection has been demonstrated to image acoustically small structures volumetrically and at higher sensitivities, providing a spatial resolution in the range of 75−125 μm depending on the imaging depth and the FOV, which is typically 1−2 cm in both x and y axis. By being able to detect reliably the thermally induced peripheral vasoconstriction induced by thermal stimuli, it can be a prospective evaluation tool for patients with perivascular diseases By bringing several other significant advantages, such as high sensitivity, miniaturized sensor size, full transparency, wider response frequency, higher axial resolution and possibility of contact free measurements, fiber based ultrasound transducer opens up more opportunities in all kinds of PA endoscopy applications. Furthermore, clinical integration of PAI systems depends on a balance of functionality and seamless end-user operability. The systems have to be designed with the clinician or care provider as the final end-user in mind and such user involvement is vital for successful implementation of PAI in the clinics. As such, its tandem use with other conventional clinical imaging modalities such as MRI and OCT in a clinical workflow can facilitate this. Pre-clinical studies demonstrating multi-modal imaging of PAI with MRI [147][148][149] and OCT [150,151], fluorescence, two-photon and second-harmonic generation imaging systems [152][153][154] have been demonstrated in recent years, to exploit the complementary information granted from anatomical landmark imaging with high molecular and functional contrast delivered by PAI. As far as technology evolution goes, combining PAI with another optical imaging modality into one integrated system may not face as many setbacks as combining PAI with MRI for instance. Thus, it is important to identify the unmet clinical needs to be addressed by a multimodal integrated system as compared to sequentially performed PAI and MRI. Combining PAI and MRI would also require registration algorithms to align PAI molecular information with high-field MRI images based on mutual fiducial markers [149,155].
Thirdly, regulatory red tapes by regulatory authorities such as FDA in USA and CE Mark in European Union, present one of the challenges in translating PAI to the clinics. These regulations were put in place to assure quality, performance, and compliance of these technologies with American and European clinical standards. Nevertheless, a few commercial PAI systems are on their way to get regulatory approvals for clinical use, beginning with the first CE-marked PAI system in 2014, Imagio by Seno Medical Instruments, Inc. in Europe for diagnosing breast cancer. Other commercial PAI systems have followed suit such as MSOT Acuity (iThera Medical GmbH) with its CE Mark recently awarded for clinical use. Another present setback is clinicians having differing opinions on the value and effectiveness of PAI in the clinics, which can be mitigated by the active facilitation of the technology via an external national agency. The agency can act as a mediator across policy, systems and organizational levels which are important determinants in implementing a new technology in the clinics.
Finally, advancements in PAI systems has opened new doors in its applications pertaining to disease diagnosing, treatment, and monitoring, while generating copious amounts of data. Beyond just visual interpretation involved in clinical diagnostic paradigm, evaluations of such data can be enhanced by advanced computational analyses. While current systems come with relatively time consuming image re-construction and spectral un-mixing algorithms, further refining these to potentially reduce the processing time and improve the usability and image interpretability like in conventional sonography will impact the confidence level among clinicians to deduce reliable real-time functional information. Moreover, artificial intelligence (AI) can be utilized to make use of such generated data to translate them to clinically interpretable information. Acquired PA images could be deconstructed by AI into phenotypic descriptors to better quantify their visualizations. Such descriptors could characterize the shape, size, and pattern of the vascular remodeling in diseases, for instance. Deep learning, a subfield of AI, is an upcoming automatic approach that 'learns' phenotypic descriptors from a representative data to perform human-directed task-specific applications and can outperform human capabilities. Clinically, the use of AI in tandem with PAI can potentially result in better disease management and patient outcomes by augmenting the qualitative assessments made by clinicians and introducing earlier interventions.
Declaration of Competing Interest
Singapore Bioimaging Consortium has signed Research Collaboration Agreements free of financial interests with iThera Medical, GmbH and MicroPhotoAcoustics Inc. individually. VN is a shareholder in iThera-Medical GmbH, Munich, Germany.
|
2019-11-14T17:11:11.277Z
|
2019-11-07T00:00:00.000
|
{
"year": 2019,
"sha1": "f57e80650202f1b98ac4194d10bcf4aa2492dc44",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1016/j.pacs.2019.100144",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "062526f98740515eb71fdddaf0e8616fbb50cac0",
"s2fieldsofstudy": [
"Medicine",
"Engineering"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
55148919
|
pes2o/s2orc
|
v3-fos-license
|
Photocatalytic Degradation of Anthracene in Closed System Reactor
Polycyclic aromatic hydrocarbons (PAHs) represent a large class of persistent organic pollutants in an environment of special concern because they have carcinogenic and mutagenic activity. In this paper, we focus on and discuss the effect of different parameters, for instance, initial concentration of Anthracene, temperature, and light intensity, on the degradation rate. These parameters were adjusted at pH 6.8 in the presence of the semiconductormaterials (TiO 2 ) as photocatalysts overUV light.Themain product of Anthracene photodegradation is 9,10-Anthraquinone which isidentified and compared with the standard compound by GC-MS. Our results indicate that the optimum conditions for the best rate of degradation are 25 ppm concentration of Anthracene, regulating the reaction vessel at 308.15 K and 2.5mW/cm of light intensity at 175mg/100mL of titanium dioxide (P25).
Introduction
Polycyclic aromatic hydrocarbons (PAHs) constitute an important and hazardous group of priority contaminates [1].Several researchers determined the concentrations of these compounds by different chromatographic techniques, for instance, HPLC [2][3][4], GC-MS/MS in [5], and GC-MS in [6,7].The main processes that successfully eliminated PAHs from the environment included the microbiological transformation and degradation, bioaccumulation, biological uptake volatilization, photooxidation, and chemical oxidation [8].Photolysis and ozonation are the most important methods for transformation for most PAHs adsorbed on natural substances in an environment [9].Photolysis of PAHs led to the formation of photodimers and photooxidation products [10].In recent years, the release of toxic and organic contaminants into aquatic environment as a result of human activities has drawn much attention and is considering a baffling problem facing researchers today [11].Zinc oxide and titanium dioxide are universally considered as the most important photocatalysts due to their lower cost and their considerably low band gap energy (∼3.2 eV) [12].Nanoparticles of titanium dioxide were considered to be more efficient than bulk powder in photocatalytic field [13].Several previous works used titanium dioxide as catalyst for degradation of different organic pollutants [14][15][16].
Semiconductors have been used for pollutants degradation in water to be less harmful inorganic material.Both catalysts titanium dioxide and zinc oxide have photocatalytic properties which made these catalysts the best for photodegradation of water pollutants [17].Attention has been focused in the past decade on using nanocrystalline TiO 2 as a photocatalyst for the organic pollutants degradation.TiO 2 semiconductor has a wide band gap about 3.2 eV, which corresponds to the UV-range radiation.The formation of an electron hole pair occurs within the conduction and valance bands of TiO 2 semiconductor after absorption in UV range.Water molecules can be oxidized to hydroxyl radical by positive hole.Scheme 1 shows the mechanism diagram for photodegradation of Anthracene.The hydroxyl is a radical, frequent, and powerful oxidant.The oxidation of organic pollutants seems to be mediated by a series of reactions started by hydroxyl radical on the TiO 2 surface.Recombination for the produced hole from valance band and separated electron from conductive band it can appear either in the volume or on the surface of the semiconductor particle accompanied with heat releasing.To this end, both UV light source and TiO 2 are necessary for photooxidation reaction to occur [18].The photodegradation of PAHs compounds in water using TiO 2 catalyst has proved high efficiency in [19].Furthermore, many researchers dealt with the PAHs in water by photocatalytic degradation for TiO 2 .Woo et al. in [20] investigated photocatalytically oxidation using TiO 2 of 5 selected PAHs, namely, naphthalene, acenaphthylene, phenanthrene, Anthracene, and benzo[a]anthracene.Gu et al. in [21] studied degradation of phenanthrene on soil surfaces photocatalytically with the addition of nanoparticulate anatase TiO 2 under UV-irradiation.Vela et al. in [22] discussed photocatalytic processes using semiconductor materials (ZnO and TiO 2 ) to remove the residual concentrations of several PAHs from groundwater.Theurich et al. in [23] reported themechanism of the photocatalytic transformation of naphthalene and Anthracene qualitatively in aqueous suspensions of titanium dioxide.Indeed, catalyst TiO 2 can play as efficient photocatalyst in the oxidation of PAHs and convert it to safer compounds especially with Anthracene, Fluorene, and Naphthalene by artificial or sunlight illumination [19,24].To this end, our aim in the present paper is to study the effect of photocatalytic reactions on the degradation of Anthracene using titanium dioxide under different experimental conditions.
Experimental Procedure
where 0 , are the PAH concentration at times (zero and ), respectively, and is the rate constant.First-order degradation rate constants were determined by regression analysis.
Results and Discussion
Several parameters were studied to indicate the effect of these degradation rates as follows.
Effect of the Initial Anthracene Concentrations.
The effect of initial Anthracene concentration on the reaction rate is the first parameter studied in this work.Figure 1 shows that Anthracene concentrations decrease with time increases.The rate of degradation increases as the initial concentration increases as well.For photochemical reactions the higher concentration causes a higher light absorption and consequently accelerates the degradation rate [26].Figure 2 shows the relation between rate constant and initial concentrations of Anthracene. the best one at which the degradation rate is fastest.Figure 3 shows the effect of temperature on the concentration of Anthracene with time of reaction.Figure 4 illustrates the Arrhenius plot for the relation between ln and 1/, activation energy calculated by Arrhenius equation.
Effect of
The influence of temperature on the degradation rate is typical.The greatest increase of the rate of degradation about two times can be achieved after about 180 min at 308.15 K; initial rates within the first 30 min appear to increase with increasing the temperature.This phenomenon is related to the effect of temperature on the stability of Anthracene molecule.Luo et al. [27] reported that higher temperature slightly enhances the rate constant of Pyrene.
Effect of Light Intensity.
UV light intensity has an important role in the process of photocatalytic degradation.Figure 5 shows effect of light intensity on the degradation rate of Anthracene molecule.This figure indicates that the reactions followed pseudo-first-order rate constant with increasing UV light intensity from 1-2.5 mW/cm 2 .The results indicate the perfect degradation at light value: 2.5 mW/cm 2 .
Therefore, when light intensity increases the number of photons increases which means that the formation of electrons and holes increases, and hence, electron-hole recombination is negligible.However, at the lower light intensity, electron and hole pair separation competes with recombination which in turn decreases the formation of free radicals [28], causing less effect on the rate of degradation of the Anthracene as shown in Figure 6.
Photodegradation Products of Anthracene.
The major product for photodegradation of Anthracene is 9,10-Anthraquinone.Anthraquinone is characterized by GC-MS (2010-SHIMADZU).Standard solution 25 ppm of 9,10-Anthraquinone (Sigma Aldrich) was prepared and compared with that produced by oxidation.Figures 7 and 8 illustrate that there are no differences between them.The proposed degradation pathway of Anthracene is as in Figure 11.
During the photocatalytic degradation experiments with Anthracene only 9,10-Anthraquinone was detected as an intermediate, in agreement with Theurich et al. [23].Figures 9 and 10 show the GC chromatogram and mass spectra for 9,10-Anthraquinone after exposure to the light intensity in the same environment of perfect conditions for Anthracene degradation.
Conclusions
The photocatalytic degradation of Anthracene using artificial UV light has been achieved.The observations of these investigations demonstrate the importance of selecting the optimum parameters for degradation to obtain a high degradation rate, which is considered essential for any application of photocatalytic oxidation processes.The experimental work in controlled pH media at closed system reactor has found that the main product of oxidation of Anthracene is 9,10-Anthraquinone, which is safer for environment than Anthracene.The rate of photodegradation in present UV light has been found to be maximum in neutral medium with optimum concentration of 25 ppm of Anthracene.The optimum temperature for degradation is 308.15K.The optimum light intensity is 2.5 mW/cm 2 at pH 6.8.The degradation of Anthracene increases with the increase of light intensity.Nevertheless, the increase of light intensity leads to the increase of the number of electron-hole pairs and increases the degradation of Anthracene.
Temperature.The oxidation of Anthracene molecule was studied at different temperatures, to indicate .76 kJ mol−1
Figure 5 :− 1 )Figure 6 :
Figure 5: Effect of light intensity on the degradation rate of Anthracene.
Figure 7 :
Figure 7: Chromatogram of GC-MS for standard 9,10-Anthraquinone and produced by oxidation of Anthracene before exposure to UV light.
Table 1 :
The average recovery of Anthracene and relative standard deviations RSD ( = 5).
2.1.Chemicals and Reagents.Anthracene was purchased from Sigma Aldrich, Germany, and used without further purification.Acetonitrile (anhydrous, ≥99.98%),Dichloromethane (anhydrous, ≥99.98%),Acetone, ethyl acetate (anhydrous, ≥99.98%), and methanol HPLC-gradient grade were purchased also from Sigma Aldrich, Germany.Titanium dioxide particles were purchased from Degussa (P25), anhydrous Na 2 SO 4 (extra pure Allied Signal, Riedel-de-Germany).2.2.Preparation of Stock Solution of Anthracene.A set of dilutions of Anthracene solution at the concentration of 100 mg/L were made in the following solvents: methanol, dichloromethane, acetonitrile, ethyl acetate, and acetone.Anthracene solutions in the above solvents were prepared and stored in room temperature (20 ± 2 ∘ C) in dark place to keep it from the light degradation.Calibration curve for Anthracene solution has been achieved by preparation several concentrations (0.1, 0.5, 1, 2, 4, 8, 16 and 30) mg/L.All glassware used for experiments was washed in chromic acid mixture for 12 h with methanol, deionized water, and acetone and then dried at 110 ∘ C for 3 h.madzu,Japan).The study revealed that the Anthracene level had no effect on the percent decrease of the compound during evaporation process for the solvent.The average recovery of analytes for every liquid media and corresponding relative standard deviations RSD ( = 5) were represented in Table1.Chromatographic conditions are listed in Table2.
Table 2 :
Chromatographic conditions were used for determination Anthracene by GC.
∘ C 2.4.Photolysis Experiments.The experiments were carried out in glass dual wall reactor closed system type, to keep the temperature constant using chiller (Julabo model EH/Germany) as temperature controller.Agitation of the reaction mixture was provided by a magnetic stirrer (Heidolph-Mr3001).The photoreactor operated in a batch mode.The study was carried out for selected compound Anthracene (Sigma Aldrich) without additional purification.The pH of the reaction solution adjusted about 6.8, pH by adding an exact volume of Sodium hydroxide or sulfuric acid.2.5.Kinetics of the Photocatalytic Process.Kinetics ofAnthracene degradation was calculated by the first-order equation:
|
2019-04-05T03:29:09.434Z
|
2014-07-08T00:00:00.000
|
{
"year": 2014,
"sha1": "d5fc5edbc8986f319841a69c983367b4749aace9",
"oa_license": "CCBY",
"oa_url": "https://downloads.hindawi.com/journals/ijp/2014/503825.pdf",
"oa_status": "GOLD",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "d7eb5bed3b9c34598be7b3e7fe0be121ba591a71",
"s2fieldsofstudy": [
"Environmental Science",
"Chemistry"
],
"extfieldsofstudy": [
"Chemistry"
]
}
|
251158635
|
pes2o/s2orc
|
v3-fos-license
|
Antivirals With Activity Against Mpox: A Clinically Oriented Review
Abstract Mpox virus is an emergent human pathogen. While it is less lethal than smallpox, it can still cause significant morbidity and mortality. In this review, we explore 3 antiviral agents with activity against mpox and other orthopoxviruses: cidofovir, brincidofovir, and tecovirimat. Cidofovir, and its prodrug brincidofovir, are inhibitors of DNA replication with a broad spectrum of activity against multiple families of double-stranded DNA viruses. Tecovirimat has more specific activity against orthopoxviruses and inhibits the formation of the extracellular enveloped virus necessary for cell-to-cell transmission. For each agent, we review basic pharmacology, data from animal models, and reported experience in human patients.
Human mpox, caused by the mpox virus, a member of the genus Orthopoxvirus within the Poxviridae family of doublestranded DNA (dsDNA) viruses ( Figure 1) [1][2][3][4], was first described in a 9-month-old infant in the Democratic Republic of Congo in 1970 [5]. Since then, it has resulted in multiple outbreaks in Central and West Africa, and occasionally in Europe and North America [6], most notably 47 human cases in the US Midwest in 2003 [7]. This outbreak was attributed to prairie dogs that became infected though contact with rodents imported from Ghana [8]. Human infections in endemic areas have been described in association with close contact with infected animals through hunting and skinning, or household rodent infestation [9]. Human-to-human transmission has also been described in household contacts of index cases, particularly among those who are unvaccinated against smallpox [10]. Proposed routes of transmission include salivary or respiratory secretions; contact with skin lesions, body fluids, or contaminated fomites; and possibly fecal shedding [10][11][12]. It is estimated that smallpox vaccination provides 85% protection against mpox, explaining the increase in susceptible hosts since smallpox eradication and discontinuation of routine smallpox vaccination [13]. The clinical course and possible complications of human mpox are illustrated in Figure 2 [9,[14][15][16]. Genomic sequencing of mpox isolates from the United States, West Africa, and Central Africa demonstrated the existence of 2 clades: the Congo Basin (CB) clade and the West African (WA) clade, including the 2003 US samples [17]. The CB clade is associated with increased human-to-human transmission, more pronounced rash, viremia, severe illness, and a higher case fatality rate (10.6% vs 3.6%) compared with the WA clade [6,17]. Diagnosis is made by combining the clinical and epidemiological picture with a viral assay, most commonly a viral DNA detection assay by real-time polymerase chain reaction [18]. The optimal specimen is a lesion exudate or crust material. Infections can be diagnosed retrospectively with serological testing [19]. For years, the management of mpox infections has relied on supportive care and management of complications; however, the recent development of new antivirals, such as tecovirimat and brincidofovir, has opened new therapeutic opportunities [20].
As of 17 June 2022, 2525 confirmed cases of mpox have been reported from 37 countries not known to be endemic for mpox. The highest number of cases have been described in the United Kingdom, Spain, and Germany [21]. Preliminary data suggest the ongoing outbreak is related to the WA clade. A particular clinical manifestation reported is the initial appearance of the rash in the genital or perianal area, suggesting close physical contact as the route of transmission [22]. In light of this unprecedented outbreak, this review aims to provide a clinically oriented discussion of 3 antiviral agents with known activity against mpox: cidofovir (CDV), brincidofovir (BCV), and tecovirimat.
Basic Pharmacology
Although CDV (Vistide, Gilead) has broad activity against many DNA viruses including orthopoxviruses, it is only Food and Drug Administration (FDA) approved for the treatment of cytomegalovirus retinitis [23,24]. Cidofovir is a prodrug, which must first enter host cells, then is phosphorylated by cellular enzymes into the active form, CDV diphosphate (CDV-pp) [24]. Once phosphorylated, CDV-pp has a prolonged intracellular half-life [25,26]. During DNA replication, CDV-pp is incorporated into the growing DNA strand and slows synthesis of DNA ( Figure 3). Cidofovir diphosphate may also inhibit DNA polymerase 3′-5′ exonuclease activity [24].
Resistance to CDV has been well described. Using serial passage with increasing CDV concentrations, resistant poxviruses can be selected in vitro [27]. These mutations appear to be similar in mpox and vaccinia virus and are due to point mutations in the conserved poxvirus DNA polymerase 3′-5′ exonuclease and the DNA polymerase catalytic domains [27,28]. Resistance to CDV typically occurs in a stepwise fashion, with moderate resistance occurring with single mutations and higher levels of resistance occurring with multiple mutations [27]. Studies have demonstrated that CDV-resistant virus is significantly less virulent than wild-type strains, as challenges with wild-type virus were commonly lethal, while CDV-resistant virus caused a mild disease course. These data indicate that CDV resistance is slow to develop and is associated with a fitness cost for orthopoxviruses [27,29].
Pharmacokinetic Data
Cidofovir is poorly absorbed orally and only available by intravenous infusion. Plasma CDV is rapidly renally filtered and secreted, whereas intracellular phosphorylated metabolites have a prolonged half-life, which allows for weekly or biweekly dosing (Table 1) [30,31]. Poxviruses known to infect humans within the Poxviridae family; 4 genera include the species that are most commonly known to infect humans. While not characterized as human pathogens, additional orthopoxviruses, such as mousepox and rabbitpox, serve as the infectious agent in animal models that most closely replicate human infections with other orthopoxviruses such as smallpox (variola). Figure created with BioRender.com.
Animal Data
Various animal models have evaluated the efficacy of CDV for the treatment of multiple orthopoxvirus infections, including cowpox, vaccinia, mpox, and ectromelia (mousepox) viruses [32]. The majority of these studies evaluated the use of CDV at the time of orthopoxvirus exposure or soon (24-48 hours) thereafter, and it is unclear how time to treatment in these models correlates with the timeline of human infection. Nevertheless, in mice infected with vaccinia and cowpox viruses, intraperitoneal CDV prevented mortality when given up to 96 hours after infection, a time point almost halfway through the disease course in this animal model. Cidofovir reduced viral titers in the lungs, liver, kidney, and spleen [33] in a T-celldeficient murine model of progressive vaccinia. Topical CDV prevented disease progression when given within 2 days of infection and decreased lesion severity up to 5 days postinfection, while systemic CDV decreased lesion severity when administered up to 15 days postinfection [34]. Further, in mice infected with cowpox virus, CDV has been shown to not only decrease viral loads but also to decrease cytokine levels in plasma and tissue, including interleukin (IL)-2, IL-3, IL-6, and IL-10 [35]. It is unclear if CDV has immunomodulatory effects or if these results are due to reduced viral titers.
In cynomolgus monkeys vaccinated with vaccinia virus, systemic CDV reduced the size of lesions at the vaccine site and promoted more rapid healing of the initial lesion [36]. In nonhuman primates exposed to mpox, CDV has been shown to prevent lesion development when given up to 48 hours after infection, while monkeys treated with placebo had numerous lesions and viremia [37]. Taken together, systemic CDV appears to be most effective when given early after mpox exposure, but may be useful at decreasing disease manifestations even when given relatively late in the mpox disease course.
Toxicity
Cidofovir is associated with dose-limiting nephrotoxicity, which is characterized by proteinuria followed by glucosuria, decreased bicarbonate, uric acid, and phosphate. If CDV is continued, this leads to serum creatinine elevation, which can be severe [31][32][33]. Nephrotoxicity due to CDV is dose-related [32] and is due to accumulation of CDV in kidney proximal tubule cells through organic anion transporter 1 (OAT1) [34]. Nephrotoxicity can be partially ameliorated by probenecid, which is an inhibitor of OAT1 transport and reduces CDV accumulation in proximal tubular cells [34]. In phase I/II studies The virus replicates at the initial infection site, resulting in a local inflammatory response. The virus then spreads to the regional lymph nodes and via the bloodstream (primary viremia) to lymphoid organs, which explains the signs and symptoms seen during the prodrome phase, including lymphadenopathies. The virus spreads again to the bloodstream (secondary viremia), leading to the end-organ involvement with the skin rash and other complications. Fever starts during the prodrome phase and resolves within 3 days of rash onset. Lymphadenopathy is a specific manifestation of mpox, differentiating it from smallpox and varicella. The skin lesions evolve from macules, to papules, to vesicles and pustules, and finally to crusts and scabs, each phase taking about 2 days on average. The skin lesions then resolve, often with pitted scarring. Additional complications can occur from secondary bacterial infection or viral spread to other organs and could lead to death. The frequency of these complications is reported based on a description of cases from the 1981-1986 outbreak in the Democratic Republic of Congo and might not reflect the severity of other outbreaks caused by a different clade of the virus. Specific characteristics of the 2022 outbreak are highlighted. Figure created with BioRender.com.
in patients with AIDS, pre-hydration and probenecid reduced rates of nephrotoxicity, especially at CDV doses greater than 3 mg/kg (Table 1) [36]. Due to this nephrotoxicity, CDV is contraindicated in patients with serum creatinine greater than 1.5 mg/dL, creatinine clearance of 55 mL/minute or less, or 2+ or greater proteinuria, and it is recommended to avoid concomitant nephrotoxic medications [33].
Clinical Data in Humans
In humans, CDV has been used to treat cases of infection with poxviruses. The activity of the intravenous (IV) formulation was documented in patients with molluscum contagiosum receiving CDV for a concomitant AIDS-associated cytomegalovirus (CMV) retinitis, with subsequent resolution of molluscum lesions [38]. Additional case reports mention the use of IV CDV as part of a multipronged management approach for ocular cowpox [39,40]. It has also been used in 1 patient with eczema vaccinatum in combination with tecovirimat [41]. Topical CDV has been successfully used to treat children and adults with molluscum contagiosum or orf. The strengths of the compounded creams varied from 1% to 3%, and the used vehicles differed, although vehicles containing propylene glycol were preferred, given that propylene glycol can enhance the bioavailability of CDV [42][43][44]. The lesions typically demonstrate acute inflammation after application of CDV, followed by dramatic resolution [45]. In some patients, the lesions recurred after discontinuation of topical CDV; however, they were successfully managed with either an additional course of topical CDV [43] or curettage [44]. In 1 patient with recalcitrant molluscum contagiosum, 1% CDV was injected into skin lesions with a 0.05-mL volume per lesion, with complete remission of the treated lesions without scarring, and with the antiviral activity being limited to the treated skin lesions [46].
Basic Pharmacology
Brincidofovir is a lipid-conjugated CDV analogue that is marketed under the brand name Tembexa (Chimerix). Brincidofovir was FDA-approved in 2021 for the treatment of smallpox [47]. Like CDV, BCV has broad activity against dsDNA viruses but has lower half-maximal effective Notably, mpox virus undergoes its entire life cycle inside the cytoplasm since it carries all the enzymes it needs for DNA replication and protein synthesis, thus obviating the need for an intranuclear stage. Viral particles are assembled into intracellular mature viruses, then released as extracellular enveloped viruses during cell lysis. Cidofovir and its prodrug brincidofovir inhibit DNA synthesis by incorporation of cidofovir diphosphate into the growing DNA strand. Tecovirimat inhibits membrane protein p37, which is essential for the formation of the extracellular enveloped virus upon cell lysis. Figure adapted from "Generic Viral Life Cycle" by BioRender.com (2022); publication and licensing rights obtained from BioRender. Retrieved from https://app.biorender.com/biorender-templates. [46][47][48][49][50]. The added alkoxyalkyl moiety in BCV is structurally similar to lysophosphatidylcholine (LPC), which allows BCV to be taken up by the small intestines [25]. Contrary to CDV, which slowly crosses cellular membranes, BCV readily enters host cells due to its lipophilicity [25]. Brincidofovir is then hydrolyzed by cellular phospholipases into CDV [25] and phosphorylated into CDV-pp. Cidofovir diphosphate reaches higher intracellular concentrations after BCV administration due to its ability to cross cellular membranes more efficiently. Like CDV, BCV has a prolonged intracellular half-life and inhibits poxviruses DNA replication ( Figure 3) [25,26]. As BCV is converted into CDV, crossresistance between BCV and CDV is expected.
Pharmacokinetic Data
Initial studies in humans have shown that oral BCV is absorbed in the fasting state and has lower peak CDV concentrations in plasma [51]. This gives BCV the convenience of oral dosing (Table 1). In addition, BCV demonstrated a significantly higher penetration into lung, spleen, and liver tissues, albeit with lower concentrations in the kidneys [52]. Unlike CDV, which is transported into the proximal convoluted tubules by OAT1, where it accumulates and causes renal damage, BCV is not a substrate for OAT1 [52,53]. Thus, BCV does not accumulate in the kidneys and has a lower risk for nephrotoxicity [52,53].
Animal Data
Brincidofovir has been tried in multiple poxvirus animal models [54][55][56][57]. In mice infected with ectromelia virus, CDV and BCV reduced mortality significantly compared with placebo [54]. Furthermore, BCV prevented mortality when given within 5 days of intranasal ectromelia virus challenge, which is thought to be analogous to the time of first lesion appearance in mpox [54]. In a rabbitpox model in which therapy was initiated on the first day of lesion appearance, rabbits treated at day 3 postinfection had improved survival (88%) compared with those treated at day 4 (67%) [55]. There was no statistical improvement from placebo if given later than day 4, regardless of when lesions occurred [55]. Similarly, an intradermal rabbitpox model showed BCV improved survival when started immediately at the time of fever (around day 2 postinfection) or within 24 to 48 hours with 100% versus 93% survival, respectively [56].
The prairie dog mpox model is very similar to the mpox infection course in humans and is characterized by a 10-to 13-day incubation period, followed by about 2 days of fever, ultimately leading to the appearance of generalized lesions [57]. In prairie dogs, BCV was shown to improve survival when given shortly after mpox exposure [57]. Taken together, these models indicate that early treatment with BCV is key for treatment efficacy, and ideally this would be taken as soon as infection is known, or as soon as prodrome or lesions develop.
Toxicity
Pooled data from phase I/II/III studies indicate that common adverse effects with BCV include gastrointestinal and hepatocellular toxicity (Table 1) [58]. These adverse effects appear to be dose and frequency related [58]. Compared with CDV, BCV has lower rates of nephrotoxicity and the advantage of oral administration [58].
Clinical Data in Humans
Brincidofovir has been administered to select patients with infections caused by poxviruses. A summary of the published case reports is presented in Table 2. Additionally, BCV has been evaluated for the prevention and treatment of other dsDNA viruses. A phase II trial studying BCV for primary CMV prophylaxis in allogeneic hematopoietic cell transplant (HCT) recipients showed a significant reduction in CMV events in the 100-mg twice-weekly arm compared with placebo. In this trial, diarrhea was dose-limiting at 200 mg twice weekly [59]. Nevertheless, a subsequent phase III trial evaluating the same indication failed to demonstrate a difference in clinically significant CMV infection between BCV 100 mg twice weekly and placebo and showed a higher rate of serious adverse events in the BCV arm. The increased rate of adverse events was mostly driven by acute graft-versus-host disease and diarrhea. Additionally, there was slightly higher all-cause mortality at week 24 in the BCV group [60]. Another phase II trial evaluated Consider monitoring proteinuria as potential early marker of nephrotoxicity; probenecid has drug interactions due to inhibition of OAT1 Abbreviations: ALT, alanine aminotransferase; BID, bis in die (twice daily); CDC, Centers for Disease Control and Prevention; CDV, cidofovir; CMV, cytomegalovirus; CrCl, creatinine clearance; CYP, cytochrome P; dsDNA, double-stranded DNA; EC50, half-maximal effective concentration; FDA, Food and Drug Administration; G, gastric; IV, intravenous; N/A, not applicable; NG, nasogastric; NS, normal saline; OAT1, organic anion transporter 1; PO, per os (by mouth); Q8h, every 8 hours; Q12h, every 12 hours; SCr, serum creatinine; t1/2, half-life; UGT, Uridine 5'diphospho-glucuronosyltransferase; ULN, upper limit of normal.
BCV for preemptive therapy of adenovirus viremia in allogeneic HCT recipients and showed a numerically lower rate of treatment failure and all-cause mortality in the BCV 100-mg twice-weekly arm. This did not reach statistical significance, likely due to a lack of power. Nevertheless, the BCV group had a higher rate of acute graft-versus-host disease [61]. Additional retrospective studies of BCV have shown its activity when used for resistant CMV and herpes simplex treatment [62] and for herpes simplex and varicella zoster prophylaxis [63]. There is currently an ongoing phase II clinical trial evaluating intravenous BCV in patients with adenovirus infection (NCT04706923).
Basic Pharmacology
Tecovirimat (ST-246) was FDA approved in 2018 for the treatment of smallpox and is marketed under the brand name TPOXX. Tecovirimat has activity against orthopoxviruses but has no notable activity against other dsDNA viruses. Tecovirimat targets the V061 gene in cowpox, a gene that is homologous to the vaccinia virus F13L gene. This encodes for membrane protein p37, which is a well-conserved protein in orthopoxviruses and is responsible for the formation of extracellular enveloped virus (EV) [64,65]. EV is thought to be the major contributor to cell-to-cell transmission and transmission through the bloodstream to distant tissues [65,66]. Tecovirimat does not inhibit DNA or protein synthesis and does not inhibit the formation of mature virus, which remains in the host cell until cell lysis (Figure 3) [64]. Resistance to tecovirimat can occur with a single amino acid mutation at position 277 [65]. It is unknown if mutation of the p37 protein confers a fitness disadvantage to orthopoxviruses, although vaccinia viruses with engineered mutations in the F13L gene had decreased plaque size and a decrease in extracellular EV formation [65]. Tecovirimat has activity against CDV-resistant vaccinia virus strains, and there is no documented cross-resistance between tecovirimat and CDV or BCV [65].
Pharmacokinetic Data
Tecovirimat is available in IV and oral formulations. When administered in the fed state, tecovirimat can achieve a better absorption, with up to 1.6 times greater Cmax than at fasting. Tecovirimat appears to have saturable absorption at doses greater than 400 mg, with higher doses resulting in nonproportional increases in Cmax and area under the curve (AUC) [68].
Animal Data
Tecovirimat has been shown to be effective in multiple animal models of orthopoxviruses, including against mpox virus in macaque monkeys [69,70] and prairie dogs [71]. Tecovirimat decreases lesion severity even when administration is delayed [69,72]. Administration of tecovirimat within 4-72 hours after poxvirus exposure demonstrated efficacy at preventing death and a reduction in the severity of lesions in various animal models [70,[73][74][75]. Tecovirimat has been shown to decrease viral spread of vaccinia virus to distant tissues [64,66]. Altogether, tecovirimat is a promising agent in animal models for the treatment of mpox infection.
Tecovirimat appears to have synergistic activity when coadministered with BCV. In cell culture experiments with cowpox and vaccinia virus, the addition of tecovirimat reduced EC 50 values of BCV [76]. In mice infected with cowpox, BCV and tecovirimat appeared to be synergistic, especially when therapy was significantly delayed, as the combination reduced mortality compared with either drug alone [76]. The duration of treatment with tecovirimat has been studied in various animal models. Fourteen-day courses have been shown to be more protective against death [73]. Courses of less than 5-7 days in duration may lead to rebound of infection, as discontinuation of tecovirimat prior to day 10, when T-cell immunity develops, may lead to worse outcomes [74]. In immunocompromised patients, prolonged courses or combination therapy may need to be considered.
Toxicity
Phase I and II studies of tecovirimat have demonstrated that tecovirimat is safe and well tolerated (Table 1) [69]. Due to poor water solubility, IV tecovirimat is solubilized with B-cyclodextrin. Although the drug labeling recommends caution in patients with renal impairment, previous studies evaluating IV voriconazole and remdesivir, which are formulated with B-cyclodextrin, have not shown significant toxicities of this solubilizer in patients with renal impairment [77,78]. Furthermore, rapid infusion with the IV product should be avoided, as elevated Cmax following rapid infusion in animal models resulted in reversible central nervous system toxicities, including ataxia, tremors, and lethargy [79].
Clinical Data in Humans
Tecovirimat has been administered to select human patients with infections caused by orthopoxviruses. A summary of the case reports is presented in Table 3. Two patients received it for mpox. Limited details are available about the first patient, except for complete recovery [80]. The second patient received a 2-week oral course initiated 5 days after rash onset, achieved full recovery with no treatment-related complications, and was discharged from the hospital after a 10-day stay [20]. Of interest, 1 immunocompromised patient developed resistance to tecovirimat during a prolonged treatment course for progressive vaccinia; however, he received BCV concomitantly and he completely recovered [67]. There are 4 registered ongoing clinical trials evaluating tecovirimat as oral or intravenous formulation for orthopoxviral exposure (NCT02080767, NCT05380752) and its safety, tolerability, and pharmacokinetics when administered for 28 days (NCT04971109, NCT04957485).
FUTURE DIRECTIONS
In conclusion, the 3 antivirals reviewed here demonstrate activity against mpox. Given their favorable tolerability profile, tecovirimat and BCV are promising therapeutic options. Larger studies should seek to identify the patients at highest risk of complications due to mpox infection (eg, immunocompromised, pregnant women, children, older adults) who might benefit the most from antiviral therapy, and to determine the optimal starting time and duration of antiviral therapy.
Note
Potential conflicts of interest.
|
2022-07-30T06:16:50.762Z
|
2022-07-29T00:00:00.000
|
{
"year": 2022,
"sha1": "4076384808e4c5faac389ddd0c0421491c483dfb",
"oa_license": "CCBYNCND",
"oa_url": "https://academic.oup.com/cid/advance-article-pdf/doi/10.1093/cid/ciac622/45473719/ciac622.pdf",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "39e838f5e8771cc8a9e61071c5af7c32e07b9431",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
220077690
|
pes2o/s2orc
|
v3-fos-license
|
Incubators: Building community networks and developing open educational resources to integrate bioinformatics into life science education
Abstract While it is essential for life science students to be trained in modern techniques and approaches, rapidly developing, interdisciplinary fields such as bioinformatics present distinct challenges to undergraduate educators. In particular, many educators lack training in new fields, and high‐quality teaching and learning materials may be sparse. To address this challenge with respect to bioinformatics, the Network for the Integration of Bioinformatics into Life Science Education (NIBLSE), in partnership with Quantitative Undergraduate Biology Education and Synthesis (QUBES), developed incubators, a novel collaborative process for the development of open educational resources (OER). Incubators are short‐term, online communities that refine unpublished teaching lessons into more polished and widely usable learning resources. The resulting products are published and made freely available in the NIBLSE Resource Collection, providing recognition of scholarly work by incubator participants. In addition to producing accessible, high‐quality resources, incubators also provide opportunities for faculty development. Because participants are intentionally chosen to represent a range of expertise in bioinformatics and pedagogy, incubators also build professional connections among educators with diverse backgrounds and perspectives and promote the discussion of practical issues involved in deploying a resource in the classroom. Here we describe the incubator process and provide examples of beneficial outcomes. Our experience indicates that incubators are a low cost, short‐term, flexible method for the development of OERs and professional community that could be adapted to a variety of disciplinary and pedagogical contexts.
addition to producing accessible, high-quality resources, incubators also provide opportunities for faculty development. Because participants are intentionally chosen to represent a range of expertise in bioinformatics and pedagogy, incubators also build professional connections among educators with diverse backgrounds and perspectives and promote the discussion of practical issues involved in deploying a resource in the classroom. Here we describe the incubator process and provide examples of beneficial outcomes. Our experience indicates that incubators are a low cost, short-term, flexible method for the development of OERs and professional community that could be adapted to a variety of disciplinary and pedagogical contexts.
general education for science majors, genomics proteomics bioinformatics, integration of courses, learning and curriculum design, original models for teaching and learning, scholarship of teaching and learning, open educational resource (OER), professional development, community networks
| INTRODUCTION
Creating undergraduate biology educational resources in rapidly emerging, cross-disciplinary topics is an especially challenging task. Bioinformatics, an interdisciplinary field at the nexus of biology, statistics, and computer science, is a particularly relevant example that highlights both the opportunities and obstacles educators face in finding and using curricular materials in newly developing fields. On the one hand, this new, interdisciplinary field, with its wealth of freely available tools and immense datasets, is a potential gold mine for authentic student research that fully supports fundamental life science concepts. On the other hand, bioinformatics requires life science students to develop competency in a variety of scientific skills including quantitative analysis, problem solving, and handling "big data," as well as the ability to work and communicate with scientists trained in diverse disciplines. Furthermore, while calls for integration of bioinformatics into life science education have been heard for more than a decade, 1-3 the core competencies that are most important for students to acquire in this new and dynamic discipline have only recently been defined. 4,5 Compounding these challenges, relatively few life science educators are well-trained in this fastmoving area where new approaches appear frequently. 6,7 Finally, creating inquiry-based curricula in an emerging interdisciplinary field requires that faculty not only master the content, but also apply effective pedagogical approaches to design curricular tools and resources. 1 Clearly, creating exercises that actively engage students, involve authentic data and tools, and are at an appropriate level for undergraduates is a complex undertaking.
To address this educational need, we describe the conceptualization, design, and implementation of a new online collaboration environment that we call "incubators." These distributed learning communities address dual aims: to build a robust collection of engaging bioinformatics educational materials, and to support faculty in the process of integrating bioinformatics resources into their life science classes. While our work focuses on integrating bioinformatics into life sciences education, this model could be easily adapted to virtually any field of interest.
| BACKGROUND
The incubator platform was developed as a collaboration between Network for the Integration of Bioinformatics into Life Sciences Education (NIBLSE) and Quantitative Undergraduate Biology Education and Synthesis (QUBES). 8 NIBLSE, a National Science Foundation (NSF) Undergraduate Biology Education Research Coordination Network (RCN-UBE) formed in 2014, seeks to expand the network of educators seeking to integrate bioinformatics as an essential component of undergraduate life sciences education (see niblse.org). As a first step, NIBLSE recently published a set of core bioinformatics competencies that are recommended for all students in the life sciences (Table 1 5 ). NIBLSE has also established a set of vetted bioinformatics learning resources (niblse. org) and is currently developing a set of learning and skill assessment tools. QUBES is an NSF-supported community of math and biology educators who share resources and methods for preparing students to use quantitative approaches to tackle authentic, complex, biological problems (see qubeshub.org). The QUBES project, with its suite of cyberinfrastructure tools and support services, provides organizations with the opportunity to host and share their activities in an online collaborative environment. One of the central features of the QUBES infrastructure is its open education resource (OER) publishing system that makes it easy to share teaching and learning resources. The QUBES publishing system has mechanisms for authenticating its published items (digital object identifiers, metadata, and tracking attributions for adopted materials) and their impact in the larger educational community (metrics on use, number of adaptations, etc.).
The incubator strategy described here leverages the expertise of the NIBLSE community with the infrastructure and collaborative environment developed by QUBES to facilitate multiple program goals. As we elaborate below, incubators have helped to (a) expand the number of available resources aligned with the bioinformatics core competencies, (b) incorporate more educators into the network, and (c) establish supportive relationships for faculty interested in expanding the integration of bioinformatics within their programs.
| Challenges to the development and incorporation of educational resources
While OERs are well known for reducing the costs of educational materials and permitting customization, 9,10 their use and development in a new field such as bioinformatics presents several challenges. Foremost, up-todate, reliable resources are not always readily available. 7 Furthermore, even when high-quality learning resources are available, faculty may not feel comfortable using them unless they have some practical guidance from other educators. 11,12 Indeed, biology instructors identified a lack of training as one of the primary barriers preventing further incorporation of bioinformatics into undergraduate curricula. 6 Developing OERs in an emerging field can be a difficult undertaking (Table 2; for a review, see 13 ). While experienced faculty may have developed curricular resources for their own use, they are often reluctant to share them widely. The resources might work well with proper framing in their own classes; however, because they are not sufficiently polished, are too narrowly focused, or lack sufficient documentation, other instructors would have difficulty adopting them for their own use. Furthermore, there is little incentive for faculty to improve and maintain these resources or make them available to others, since such scholarly contributions traditionally have not been recognized. 14 Consequently, developing a mechanism for faculty to refine and share such resources with the potential for peer-review publication would be invaluable for the larger educational community.
There have been a variety of efforts over the past two decades to develop new curricular approaches involving bioinformatics. Many of these programs have been highly successful, but their impact has been limited by various factors. Some programs require a major overhaul of an entire course or curriculum, for example, SEA-PHAGES. 15 Others involve attending a workshop (e.g., Data Carpentry (datacarpentry.org), BioQUEST (bioquest.org)), which often requires a commitment of time and funds that are impractical for many undergraduate instructors. In addition, after the initial enthusiasm from attending a workshop diminishes, faculty often are unable to implement what they have learned in their classrooms. 14,16 In response, projects such as the Genomics Education Partnership (GEP; gep.wustl.edu) supplement initial training with ongoing interactions as material is incorporated into the curriculum. However, these approaches are typically focused on a particular task (e.g., annotating the Drosophila genome) and are not widely applicable to other areas of bioinformatics. To address these limitations, we developed the incubator process (Table 2).
| WHAT IS AN INCUBATOR?
Incubators integrate technical and social structures developed by QUBES (qubeshub.org) to create new opportunities for faculty scholarship, while supporting effective lesson design and promoting adoption of materials into classrooms. Each incubator is a relatively shortlived (4-8 weeks), online faculty learning community whose membership and workflows are engineered to produce teaching resources that incorporate effective pedagogies. Because the incubators run for a short period of time and only require a few hours per week, they take a smaller investment of faculty time than attending a workshop, and they have the additional advantage of providing ongoing assistance to participants. Indeed, the final products of an incubator are often implemented in the classroom by the incubator participants.
The end result of a NIBLSE Incubator is a current, vetted resource that is designed to capture student interest as well as provide skills and knowledge explicitly mapped to one or more of the NIBLSE bioinformatics core competencies. 5 In addition, incubators typically produce supporting materials to facilitate their adoption in diverse classroom contexts by instructors lacking extensive bioinformatics knowledge. To promote broad dissemination, the resource is made freely available as a scholarly publication on the NIBLSE site (hosted by QUBES) for other instructors to adapt and incorporate into their courses (see below).
| HOW DO INCUBATORS WORK?
Below we introduce the five steps involved in the NIBLSE Incubator model: selection of the learning resource, establishing incubator goals, identifying the incubator team, engaging in remote collaborative work, and dissemination of products ( Figure 1). Here we generally describe the key features of each step and how these are tied to the successful incubation of a resource. In addition, two specific case studies of incubator implementation are included in the Appendix S1.
| Step 1: Selection of the learning resource
The incubator process begins with draft educational resources recruited from faculty with expertise in bioinformatics teaching or research. By starting with T A B L E 2 OER challenges addressed by incubators
Challenge
Incubator solution There is a shortage of vetted educational resources available for undergraduate bioinformatics education.
NIBLSE incubators increase the number and visibility of robust bioinformatics resources available to the education community.
In newer or more interdisciplinary fields, it may be difficult to determine what the learning goals for a particular exercise are or how they fit into a larger framework.
NIBLSE incubators link resources to specific bioinformatics core competencies.
Students do not see the importance of learning bioinformatics or do not see how it relates to biological questions they are interested in.
Incubators allow authors to add engaging context and multiple applications to a core learning resource, allowing instructors to better adapt the resource to their students' interests.
Faculty are inexperienced or insecure about adopting new materials into the classroom.
The novice perspective is specifically incorporated during the development of the resource to ease adoption by educators with varying experience. Background and context for those new to bioinformatics are provided, allowing the resource to be more easily adopted.
Instructors may have resources that are narrowly tailored to the specific audience at their institution.
Incubators help authors produce a more robust resource, that can be adapted to use in a variety of settings.
Faculty may feel isolated and lack support for implementing bioinformatics into their courses.
Incubators allow faculty to consult with and make connections with other faculty from diverse institutions across the country who are also implementing bioinformatics into their courses. Incubators match experienced and inexperienced users.
Faculty who have created and used bioinformatics learning resources in the classroom do not have the time to polish and publish them.
Incubators run for 4-8 weeks, with a typical time commitment of 1-2 hr each week. Collaboration facilitates polishing and publication of the resource.
Faculty with unpolished educational resources lack the incentive to make them available to others.
Incubators result in a published resource in the NIBLSE Resource Collection, with a persistent DOI, as well as statistics on public views and downloads.
preexisting education materials, the incubators necessarily engage faculty who already have some expertise teaching bioinformatics. Consequently, the incubator work can focus on pedagogical strategies to make the bioinformatics content more accessible to both students and less experienced faculty. Submitted materials are screened by a Resource Review Committee to identify suitable candidates that align with the core competencies identified by the NIBLSE community. 5 To standardize this process, we customized a previously published Learning Object Review Instrument (LORI 17 ;) to evaluate potential bioinformatics learning resources (see Appendix S1).
| Step 2: Establishing incubator goals
While the broad objectives of the NIBLSE project informed the selection of starting materials, the goals of an individual incubator are closely tied to the features of the learning resource. The LORI review process identifies the bioinformatics core competencies addressed by the learning material and provide suggestions to improve its usability and accessibility. An Incubator Managing Editor selected from the Resources Review Committee then works with the resource author to negotiate the incubator goals. These goals generally focus on addressing one or two core competencies, providing additional background or context needed by novice bioinformatics instructors, and including a richer biological context to motivate faculty and students. Some incubators also seek to develop assessment instruments to evaluate the effectiveness of the resource. While the author and Managing Editor establish the initial goals for the incubator, other members of the incubator team help shape the direction of the incubator as it progresses, particularly as it reflects each participant's teaching context and student audience.
| Step 3: Assembling the incubator team
We typically construct heterogeneous teams of 4-6 individuals to participate in an incubator. In addition to the author of the learning resource and the Managing Editor representing the NIBLSE project, an incubator team also includes a bioinformatics expert and one or two faculty members with limited experience in bioinformatics. The perspectives of both novice and expert bioinformatics instructors have proved valuable when developing the resource: Based on their experience, bioinformatics experts can recommend potential approaches to achieving the learning objectives and ensure the accuracy of all developed materials, while novices point out aspects of the materials that might be confusing to potential adopters and students. Another important benefit is that novices receive informal training in a focused environment, with minimal time commitments. The incubator team is rounded out with a representative of the QUBES project who helps with the collaborative infrastructure and establishing the norms for remote collaboration.
Incubator members have been drawn from a variety of institutions throughout the United States and beyond ( Figure 2) and are typically selected from a list of NIBLSE volunteers who have previously expressed interest in participating. In some cases, the authors or the Managing Editor invite other individuals to participate (occasionally including graduate or undergraduate students) with specific expertise or interest.
| Step 4: Engaging in remote collaborative work
The activities of the incubators are coordinated using the QUBES cyberinfrastructure, along with third-party tools such as Google Docs and Zoom video chat sessions. A private online group space is built on QUBESHub to organize materials and communication, with the QUBES project representative initially providing an orientation to effective uses of these tools for remote collaboration. Each incubator group maintains a regular schedule of synchronous video meetings (usually an hour-long session every week or two) for the duration of the incubator (generally 4-8 weeks). These meetings are used to provide background and training, share progress and feedback on resource development, and otherwise coordinate the work of the group. Development tasks are generally distributed among participants for drafting and then shared back for discussion and feedback. Striking the proper balance between synchronous and asynchronous tasks has proved important for balancing engagement and accountability with the flexibility needed to accommodate participants' regular teaching, research, and service commitments.
| Step 5: Dissemination of final products
The incubator products are published using the QUBES open education platform. To publicly recognize the important scholarly contributions of incubator participants, QUBES publications receive digital object identifiers (DOIs) to make them easier to reference and cite in lists of professional activities. Additional benefits of the QUBES publication platform include collecting standard metrics (number of views and downloads; Table 3), as well as automatically tracking attributions when new versions or adaptations of resources are released.
While a completed incubator serves as a standalone published activity, incubators can also serve as stepping stones to further professional activities. For example, two incubated resources 19,20 have gone on to be accepted for publication in CourseSource, 24,25 a peer-reviewed journal of teaching and learning materials. Resources can also serve as the basis of professional developmental opportunities, furthering their dissemination and implementation. For example, one incubated resource 19 served as the foundation of a QUBES Faculty Mentoring Network (FMN), where 12 faculty members from a variety of institutions worked together online to adapt and integrate the resource into their classrooms. To support and guide this process, the initial author of the resource in concert with NIBLSE leadership and QUBES support collaborated with the FMN participants throughout the semester as they implemented the activity in their courses. These activities, produced in the first three years of operation, clearly illustrate the potential for incubators to further the development and dissemination of bioinformatics learning resources.
| DISCUSSION
We have described a novel community-based strategy, called incubators, for the development of bioinformatics OER for the life sciences community. Incubators are a cost-effective mechanism (with regard to both time and money) to accomplish multiple faculty-and curriculumdevelopment goals: they produce polished, vetted F I G U R E 2 Map of incubator participants (as of June 2019). Author institutions are marked with a triangle, other incubator participants with a circle. Overlapping circles represent institutions participating in multiple incubators (but not necessarily the same individual). Each incubator also had a QUBES liaison (Sam Donovan or Hayley Orndorf from the University of Pittsburgh) who, for clarity, is not indicated here [Color figure can be viewed at wileyonlinelibrary.com] learning resources; provide training and support to faculty who are inexperienced in teaching bioinformatics; and expand the number of faculty that are incorporating bioinformatics into their life science courses.
During the initial implementation period, we have learned several important lessons for successful completion of an incubator (outlined in Table 4). First, the value of an overall curricular framework cannot be overstated. The NIBLSE core competencies have proven invaluable to guide the selection of needed resources for incubators and communication with potential authors and incubator participants. A commitment to engaging and active pedagogy has also been important. In addition, setting specific goals for each incubator at the outset is essential. We prefer that the Incubator Managing Editor work with the submitting authors to establish the incubator's goals prior to forming the incubator team. This allows the incubator to start quickly and reduces ambiguity about the focus of the incubator as volunteers are recruited to participate.
Second, a defined process for seeking volunteers and forming groups is essential. When recruiting incubator participants, we clearly articulate the opportunities for authorship and emphasize that incubator participation is a way to engage in professional activities related to teaching and learning, in addition to potentially enhancing their own teaching. When forming incubator teams, we favor relatively small groups to facilitate logistical details such as finding meeting times, establishing trust, and being accountable to the group. Given that overburdened instructors are slow to volunteer for work that hasn't been traditionally recognized as a valued professional activity, it is critical to maintain a list of interested faculty volunteers so that each incubator can be launched promptly. In our case, incubator participants were primarily drawn from NIBLSE members who had expressed an interest in participating in an incubator. Other participants were nominated by a NIBLSE member and individually invited to participate based on their expertise and interests, and in some incubators, highly motivated undergraduate and/or graduate students have contributed a valuable student perspective on the appeal and accessibility of a learning resource. Having an organized Managing Editor to keep the group focused and making ongoing progress is obviously important. While participating in an incubator requires a commitment of effort by each participant (in addition to their other professional responsibilities), the overall time commitment is significantly lower than an onsite workshop: Each incubator has a flexible schedule, does not require travel, and is spread out over several weeks, which gives participants more time to process each round of revision and development (unlike the compressed time frame of a workshop). At the same time, the incubator has definitive deadlines and an end point, which encourages participants to focus on achieving discrete tasks between meetings. In this way, incubator participation does not become another ongoing, openended professional responsibility to manage in an already hectic schedule.
| Incubators provide a unique combination of supporting features for the development of curricular resources
While several analogous efforts have achieved success in developing learning resources and fostering a broader educational community, the incubator process described here is unique in its combination of supporting features. Most comparable is the VIPEr (Virtual Inorganic Pedagogical Electronic resource; ionicviper.org) project that serves the inorganic chemistry educational community by "developing materials to bring current research into the classroom, building community through cyber-technology, and testing materials and technology in the classroom and assessing student learning" (ionicviper.org 26 ;). As with NIBLSE, VIPEr recognizes that educators need to engage in ongoing discussion about the effective use of resources in different contexts, difficulties others have encountered, and possible solutions. However, while VIPEr has a mechanism for posting teaching resources and hosting discussion forums, it does not employ an online, collaborative process to develop resources for publication.
Similarly, the Bioinformatics Education Dissemination: Reaching Out, Connecting, and Knitting-together (BEDROCK) project (http://bioquest.org/bedrock/) created "problem spaces" that included background information, data (typically molecular sequences), and classroom activities to engage students in exploring the data. While BEDROCK hosted workshops to bring educators together and train them in how to use the problem spaces in their classrooms, a robust infrastructure for collaborating and interacting online was not available.
| Future plans
Future plans for the NIBLSE Resource Collection are to expand the number of resources to provide greater depth in offerings. This will require offering a greater frequency of incubators and soliciting additional raw materials from current and future NIBLSE members. In concert with QUBES, we also have plans to increase the availability of commenting and discussion tools associated with each resource. Our vision is that these tools will further facilitate faculty engagement and thereby strengthen the dynamic community of bioinformatics educators created by NIBLSE. Finally, NIBLSE is actively developing assessment materials linked to the core competencies that can be used to assess the impact of the resources in the NIBLSE collection. Future incubators could then revise the existing learning resources in response to the data collected from these assessment tools.
T A B L E 4 Tips for carrying out a successful incubator • Build resources around a Framework (competencies and pedagogy) • Recruit participants with multiple viewpoints (expert, novice, instructor, student) • Keep the group small (4-7 people) • Keep the time defined (e.g., 1 hr weekly meetings for 6 weeks) • Establish goals for the resource early
|
2020-06-27T13:06:08.743Z
|
2020-06-25T00:00:00.000
|
{
"year": 2020,
"sha1": "6a0afd454308202dd36c4821c4d40ca946b86b4a",
"oa_license": "CCBYNCND",
"oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/bmb.21387",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "65ffcfb0c889cce63bd621a3d685587ac80edb91",
"s2fieldsofstudy": [
"Education",
"Biology",
"Computer Science",
"Environmental Science"
],
"extfieldsofstudy": [
"Mathematics",
"Medicine"
]
}
|
243952751
|
pes2o/s2orc
|
v3-fos-license
|
Kalām in the crossfire: A historical survey of the legitimacy of the study of theology within the Sunni school of Islam
school and investigating the most authoritative sources of doctrinal tradition within this major denomination of Islam. Our study leads us to conclude that the disapproving views of the preeminent Sunni figures should be construed, not as a denunciation of the discipline of theology per se , but as a refutation of certain theological principles and persuasions that are viewed as incompatible with orthodox Islamic faith. In addition to and preceding that conclusion, this article provides a survey of the literature concerning the views of Muslim scholars on the legitimacy of Kal ā m . After categorising these views into the two opposite camps of Kalam’s legitimacy and illegitimacy, the article then proceeds to evaluate and critically analyse them, and to resolve some of their initial incompatibilities. Contribution: The article sheds new light on the historical development of the discipline of theology within the Sunni schools of Islam. This is a little-studied and often overlooked subject that can help in attaining a better understanding of how this critical field within religious studies has emerged into its present form.
Introduction
The branch of traditional religious studies in Islam that deals with religious belief and doctrine is designated as kalām, but it is also referred to by such other descriptive titles as: al-fiqh al-akbar ('the greatest knowledge'), ʻilm al-tawḥīd ('the science of divine unity') (see Taftāzānī 1401 AH, 1:6), ʻilm uṣūl al-dīn ('the science of the principles of faith') (see Ibn Athīr Jazarī 1400 AH, 3:161),ʻilm sharīʻat ('the science of the path of faith'), and ʻaqlīyāt ('the rational principles') (see Ibn Taymīyah 1425 AH, 19:307). Shahrastānī claimed that the use of the appellation kalām to designate this discipline was first employed by Muʻtazilī theologians (Shahrastānī 1404 AH, 1:30). The function of this discipline consists of proving and articulating the principal doctrines of faith -including postulating the facts that constitute the believer's knowledge of God, shedding light on how obedience to God can be manifested, expounding on the nature of divine prophecy and the character of God's prophets -and defending the 'boundaries' of faith against the criticisms posed by disbelieving sceptics (see Muẓaffar 1422 AH, 1:19). As such, kalām covers all doctrinal beliefs of Islam, and is not restricted to the study of God and divine attributes. (Kalām and its domain are The aim of this article is to briefly introduce and to examine the views of the founders of the four Sunni schools of Islamic jurisprudence (namely, Abū Hanīfa Nuʻmān ibn Thābit, Mālik ibn Anas, Muhammad ibn Idrīs al-Shāfʻī, and Ahmad ibn Hanbal) and that of their renowned students and followers, regarding the legitimacy of engaging in the study of Islamic doctrinal beliefs or Kalām. Different, and often conflicting, views have been postulated on this matter. Some Sunni thinkers have condemned the pursuit of theology as an act of heresy and denounced its practitioners as apostates. Other Sunni thinkers have extolled this discipline as the noblest of sciences whose learning and teaching are, at least under certain circumstances, incumbent. This fundamental dispute regarding the legitimacy of the discipline of theology has resulted in a rather contentious and opaque scholarly environment. In addition to the inherent importance of the discipline of theology as such, the significance of this dispute is compounded by the impact that the attitude one adopts toward this discipline can have on the development of the intellectual and rational aspects of Islam. A negative attitude toward theology, for example, can hamper the application of rational elements to Islamic doctrine. Our focus in this article is on analysing the views of the leading thinkers and jurists of the Sunni school and investigating the most authoritative sources of doctrinal tradition within this major denomination of Islam. Our study leads us to conclude that the disapproving views of the preeminent Sunni figures should be construed, not as a denunciation of the discipline of theology per se, but as a refutation of certain theological principles and persuasions that are viewed as incompatible with orthodox Islamic faith. In addition to and preceding that conclusion, this article provides a survey of the literature concerning the views of Muslim scholars on the legitimacy of Kalām. After categorising these views into the two opposite camps of Kalam's legitimacy and illegitimacy, the article then proceeds to evaluate and critically analyse them, and to resolve some of their initial incompatibilities.
to be more elaborately defined in a separate section of the article.) The aim of this article is to analyse the views of Muslim thinkers regarding the legitimacy or illegitimacy, of the study of Kalām, and why its practitioners have been denounced as heretics and apostates by some major Muslim figures. The question of the legitimacy of studying and practising theology has historically led to great controversy. On one side of this controversy were those who declared the pursuit of this discipline to be contrary to genuine faith, even going so far as to equate it with kufr (disbelief) and zindiqah (atheism). On the opposite side of the dispute (Anṣārī Hirawī 1418 AH, 5:71; Ibn Taymiyya 1425 AH, 16:473; Dāremī 1433 AH, 33) were those who pronounced engagement in this discipline and teaching it to others as a meritorious (Qīrwānī n.d.:169;Ghazālī n.d., and even occasionally mandatory endeavour (Bayāḍī 1425:33-37;Ibn Abī al-ʻIzz 1391 AH, 1:65). The views of the representatives of each side will be discussed below.
Considering the fact that what is intuitively expected of all devout Muslims is that they should endorse and support this discipline, as its essential function is defending the fundamental articles of Islamic faith, the question naturally arises as to why some major Muslim figures have taken such a negative stance toward the discipline of kalām? What compelled them to assume such a position? Is it in fact the discipline of theology as such that is the object of their condemnation? Or is it a particular subgroup of Islamic theologians that are intended? Or could it be that this stance was in response to certain topics and discussions that proved distasteful to the religious sensibilities of the detractors?
To answer these questions, we examine the views of the proponents and detractors of kalām, focusing primarily on the views of the imams of the four main Sunni schools of jurisprudence and their prominent students and followers. The conclusion we arrive at is that the occasional condemnation of theologians and the prohibition on the study of kalām were not directed per se at the discipline as we know it. Rather, the detractors were primarily opposed to certain theological questions and principles, and therefore denounced only a certain group and a certain persuasion of theologians in virtue of their position on these problematic questions and principles.
Historical background of the present study
Throughout the history of Islamic scholarship, numerous books have been produced on the question of whether it is permissible to pursue kalām. The following is a short list of some of the most preeminent of these works:
Defining Kalām
Kalām has numerous definitions. They vary between a minimalist concept of kalām and a maximalist one. One representative example that falls into the former category is the definition offered by Taftāzānī: 'Comprehending religious beliefs by virtue of indubitable arguments' (Taftāzānī 1410 AH, 1:6). Another such definition is that proffered by Ījī. The stress in his definition is primarily on the function of the discipline of kalām in proving religious beliefs (see Ījī 1417 AH, 1:31). Ibn Khaldūn's definition exemplifies the maximalist conceptualisation of kalām. He defined kalām as: [T]he science that defends religious beliefs by virtue of rational arguments, thus countering the heretics and those who choose to diverge from the beliefs of the predecessors and the custodians of [religious] tradition. (see Ibn Khaldūn 1984:1, 458) Thus, in Ibn Khaldūn's view, kalām, in addition to providing genuine knowledge concerning religious doctrine and proving religious beliefs, is responsible for addressing and refuting the criticisms levelled at religious doctrine. Thus, drawing on these various definitions, it would be safe to say that kalām is the discipline responsible for bringing rational reasoning to bear on religious doctrine, supporting and reinforcing religious beliefs by virtue of rational and demonstrative proofs, and offering rational responses to the criticisms and objections raised by the critics.
The verbal equivalent for 'theology' in Arabic is 'ilāhīāt' or 'the divine sciences' (al-ʻulūm al-ilāhīa). Apparently, this gained prevalence in Islamic tradition after Ibn Sīnā entitled the section on the first philosophy of his al-Shifā by the term (see Ibn Sīnā 1:4). This is a later phenomenon in Islamic literature, marked with its own historical characteristics, for example, a new phase of interaction between Islamic philosophy (Falsafa) and Kalām. However, and regardless of the lexical connections, the word 'theology' conventionally and more often, is taken to be an equivalent for the science of Kalam, which is the case in this article. As such, Theology (Kalām) means the study of Islamic doctrinal beliefs. Early examples of these doctrines included divine attributes and their relation to God, human free will and divine providence, divine justice and predestination, and the nature of divine revelation. As can be seen, Kalam covers a wide range of discussions over beliefs and doctrinal principles, as opposed to jurisprudential, legal and more generally, the practice and ritual-related aspects of Islam.
The views of Sunni scholars regarding Kalām
As already noted, Muslim scholars are divided on the legitimacy of studying and engaging in the discipline of kalām. The supporters of this discipline are not content with merely asserting the legitimacy of this discipline but rather go further, deeming its pursuit to be a priority in comparison with the other branches of the Islamic studies. The detractors, however, proclaim the pursuit of kalām to be impermissible and thus deem Muslim theologians as worthy of condemnation. What follows is a detailed investigation of these two opposing positions.
Advocates of the legitimacy of Kalām
Many Muslim thinkers, predominantly the followers of the Ashʻarī school of kalām and those of the Māturīdī school (the latter which, in contrast to the former school, identifies itself with the Ḥanafī school of Islamic jurisprudence) commend the discipline of kalām as the noblest form of human knowledge. For, they argue, its subject-matter is knowledge of God, and there is no subject that can be deemed nobler than God. As such, the pursuit of kalām should be encouraged, and in fact in the instances in which critics of religious faith raise objections regarding religious tenets, thus threatening to undermine the religious belief of the masses, learning this discipline in order to refute the raised objections is obligatory. To support their position, these scholars invoke the verses of the Qur'an and the example and thought of Prophet Muhammad's Companions (ṣaḥābah) and the Successors (tābiʻūn). Below is a treatment of the prominent Sunni scholars who endorse the discipline of kalām.
Abū Ḥanīfah (d. 150 AH)
Abū Ḥanīfah Nuʻmān ibn Thābit is among the advocates of the legitimacy of kalām. He is often hailed as the most preeminent master of the study of kalām (see ʻAbd al-ʻAzīz Bukhārī 1418 AH, 1:17). He encouraged his students to pursue kalām and famously characterised the study of the principal tenets of faith as al-fiqh al-akbar ('the greatest knowledge'), as contrasted with the study of the practical laws of Islam, the sharīʻah, which he described as al-fiqh al-aṣgahr ('the lesser knowledge'). This innovative characterisation is one obvious indicator of his favourable view of kalām. Another no less obvious indicator is his authorship of a book on kalām entitled: al-fiqh al-akbar (see Bazdawī n.d.:3; Zarkashī 1421 AH, 1:17). The title of the book is self-explanatory. The reason given by the author for this designation is that its most prominent and widely debated topic is divine unity and attributes, knowledge of which is the 'greatest knowledge', and hence the title (see ʻAbd al-ʻAzīz Bukhārī 1418 AH, 1:17; Ibn Abī al-ʻIzz 1391 AH, 1:65; Taftāzānī 1401 AH, 1:6). This book had such an impact that it came to be known as the 'Book of Knowledge', and the arguments presented in it were described as offering the most effective proofs for refuting the opponents of religious faith (Isfarāʼīnī 1403 AH:184). Abū Ḥanīfah's contribution to kalām gained such widespread acceptance that some scholars maintained that those who truly wish to learn 'the science of polemics' ought to study Abū Ḥanīfah (Abū Isḥāq Shīrāzī n.d., 1:78; Abū Shāmmah Maqdisī 1424 AH, 1:80; Mazī 1400 AH, 28:436).
Abu al-Ḥasan Bazdawī (d. 382 AH)
Abu al-Ḥasan ʻAlī ibn Muḥammad Bazdawī, the great Ḥanafī jurisprudent of Transoxiana (Dhahabī 1405 AH, 18:602), is another eminent Sunni scholar who wrote a book on kalām. This is how he described this discipline: [Religious] knowledge is of two types: the knowledge of divine unity and attributes and the knowledge of the canonic law [sharīʻah] and the rules of Islamic practice. What is crucial with regard to the former knowledge is to draw on the Scripture and the tradition and to avoid deviation (hawā) and heresy (bidʻah). Our predecessors -including Abū Ḥanīfah, Abū Yūsuf, Muḥammad, and their disciples -all exemplified this truth. Abū Ḥanīfah wrote the book al-Fiqh al-akbar on this topic, demonstrating therein the divine attributes, God's ordainment of good and evil…. (Bazdawī n.d.:3)
Abū Ḥāmid Ghazālī (d. 505 AH)
Abū Ḥāmid Ghazālī is arguably the most prominent Islamic theologian. That subsequent scholars used such reverential descriptions as shaykh ('the Master'), imām ('the leader'), baḥr al-ʻilm ('the sea of knowledge'), uʻjūbat al-zamān ('the marvel of all times'), zayn al-dīn ('the adornment of religion'), and dhakāʼ mufriṭ ('extremely clever'), to name only a few, to refer to him is a testimony to his unique status within Sunni kalām (see Dhahabī 1405 AH, 19:322). His juridical affiliation lay with the Ḥanafī denomination, and in point of his theological orientation, he subscribed to the Ashʻarī school.
To prove the legitimacy of the study of kalām he cited a number of Qur'anic verses and then concluded, 'The Qur'an, from its beginning to its end, is all argumentation' (Ghazālī n.d.,. Ghazālī emphasises the opinion that the value of the discipline of kalām is due to the goals that motivate the practitioner of kalām and the dexterity of the practitioner with respect to those goals. These can include knowing the arguments for the creation of the world and the unicity of God, and having knowledge of the divine attributes (see Ghazālī n.d., 1:95).
Badr al-Dīn Ibn Jamāʻah (d. 733 AH)
Badr al-Dīn Ibn Jamāʻah Kanānī Shāfiʻī was a grand mufti and a senior judge, a Qur'anic exegete, and an eminent Ashʻarī theologian (Dhahabī 1408 AH, 2:103). He wrote many works on kalām. Quoting Abū Manṣūr ʻAbd al-Qāhir Baghdādī, he enumerated the most prominent Sunni theologians, and among the Companions, he pointed to ʻAlī ibn Abī Ṭālib as the first practitioner of kalām on account of his debates with the Khawārij concerning the notions of divine promises and threats (al-waʻd wa al-waʻīd), and with the Qadarīyyah regarding the issues of divine preordainment and predestination (al-qaḍā wa al-qadar). He named ʻAbdullāh ibn ʻUmar -who debated with the leader of the Qadarīyyah, Maʻbad Muhanī, and disavowed them -as the next greatest theologian among the Companions (Ibn Jamāʻah, Muḥammad, 1410 AH, 21-24).
Shāṭibī (d. 790 AH)
Abū Isḥāq Shāṭibī, a prominent Mālikī jurisprudent, is of the view that by invoking the notion of 'the categorical interests' (maṣāliḥ mursalah), we are able to demonstrate that the discipline of kalām is in fact grounded in Islamic law and tradition and in the Qur'an. He reasons that kalām is nothing but the study of 'the principles of faith ' (uṣūl al-dīn), and as such its function is to articulate the arguments mentioned in the Qur'an and the corpus of tradition relating to the topic of divine unity and to other such topics that belong to the purview of the Islamic belief system (see Shāṭibī 1429 AH, 1:29). This leads him to the conclusion that if in fact kalāmand by extension all the disciplines that are in the service of the Islamic faith but that were absent in the earliest period of Islam -is an unorthodox innovation that ought to be shunned, then by the same logic we must also disavow the consolidation of the Qur'an into a single codex and the universally accepted practice of copying the Qur'an (see Shāṭibī 1429 AH, 1:29).
Ibn Abī al-ʻIzz Dimashqī (d. 792 AH)
ʻAlī ibn ʻAlī ibn Muḥammad Dimashqī, more commonly known as Ibn Abī al-ʻIzz, was a prominent Ḥanafī jurisprudent whose theological affiliation lay with the Māturīdī school of kalām. In his commentary on al-ʻAqīdah al-ṭaḥāwīyyah, he described kalām as the study of the principles of faith and thus as constituting 'the greater knowledge', in comparison with the study of the secondary tenets (furūʻāt) of faith, with which jurisprudence is concerned. Not only does he deem the pursuit of kalām permissible, but also consider it to be a need surpassing all other needs in importance and a necessity transcending all other necessities in significance. Furthermore, because of its unique role in providing knowledge of God and His attributes and actions, it is effective in engendering spiritual revival and inducing mental and emotional tranquillity in its students (see Ibn Abī al-ʻIzz 1391 AH, 1:65).
Aḥmad ibn Ḥasan Bayāḍī (1098 AH)
Kamāl al-Dīn Aḥmad ibn Ḥasan Bayāḍī is a renowned Māturīdī theologian who adhered to the Ḥanafī school of Islamic law. He elaborated extensively on the legitimacy of pursuing kalām and puts forth a number of reasons demonstrating that kalām is the noblest of sciences. Bayāḍī maintained that possessing a greater knowledge of God and His attributes enhances the quality of one's worship. This connection between worship and knowledge is so strong. Bayāḍī emphasised that the proper performance of religious rituals is essentially dependent on an adequate comprehension of the principles of faith and of the science of kalām (see Bayāḍī 1425 AH:30-33).
Opponents of the legitimacy of Kalām
In opposition to the scholars who have in various ways highlighted the significance of the discipline of kalām and have argued for its legitimacy, there are others who denounce this discipline and discourage its pursuit. The legitimacy of the study of kalām and the permissibility of its pursuit have come under attack from prominent authorities from all four schools of Sunni jurisprudence, for various reasons. In some of these cases, of course, the adopted stance is not as clear-cut as is sometimes suggested but is rather based on dubious quotations that are often at odds with other passages that seem to exhibit a positive stance toward kalām. Below, we will examine several examples of those authorities who are alleged to be opposed to the study of kalām.
Abū Ḥanīfah
Some Sunni scholars -most of whom are aligned with the Ahl al-Ḥadīth movement (which espouses a very literalist interpretation of sacred texts) -cite certain passages from Abū Ḥanīfah that seem to suggest that he was opposed to the study of kalām. These scholars claim that Abū Ḥanīfah declared the study of kalām to be forbidden (see Muqirrī, 1417 AH, 1:88), and that he condemned anyone as atheist (zindīq) who sought to approach religion by virtue of polemical disputation (jidāl) (see Khaṭīb Baghdādī 1403 AH, 2:159). Ibn Taymīyyah alleges, Abū Ḥanīfah would say, 'God curse "Amrw ibn ʻUbayd, for it was he who led the people to the study of kalām and to beliefs that were irrelevant to them"' (Ibn Taymīyyah 1408 AH, 6:561). Similarly, it is claimed that when Abū Ḥanīfah was asked to answer a question regarding the philosophic notions of substance, accident, and body, he dismissed the assertions of philosophers as heretical (bidʻah) and cautioned against learning kalām. As opposed to that, he encoured all people to adhere to 'the ways of the predecessors' (ṭarīqat al-salaf) (see Qāsimī 1399 AH, 1:298, and Abū Qāsim Iṣfahānī 1419 AH, 1:115-116).
Imām Shāfiʻī (d. 204 AH)
Muḥammad ibn Idrīs Shāfiʻī (the eponymous founder of the Shāfiʻī school of Sunni jurisprudence) is cited as another prominent opponent of the discipline of kalām. He is the source of many of the most scathing criticisms of kalām. One of the many notable remarks in this connection that is attributed to him is: 'It is necessary to shun kalām and flee from it as one would flee from a lion' (Abū Qāsim Iṣfahānī 1419 AH, 1:224). The reason why Imām Shāfiʻī was so vehemently opposed to kalām is expressed in the following passage quoted from him: Shun kalām, for if one is asked concerning a juridical question, and one gives the wrong answer, such as, say, if someone is asked concerning the blood money for the murder of another soul and he answers that it is an egg, he would only be laughed at. But if someone is asked concerning a theological question and he offers an incorrect answer, he would be denounced as an apostate. (Samʻānī 1417 AH:9) Aḥmad ibn Ḥanbal (
Analysis and evaluation
The conflict of contradictory opinions regarding the legitimacy of the discipline of kalām presents a murky and confused historical picture of the debate over this very important study. This state is especially detrimental as what is at stake is more than just a theoretical position on an abstract field of study; it influences how one decides to deal with this discipline. This practical concern is further accentuated if we take account of the fact that to delegitimise the study of kalām would in effect stay the growth and development of the intellectual component of Islamic doctrine. This makes it all the more important that we examine and evaluate the reasons why the detractors of kalām assumed such a negative stance toward this discipline. To this end, we will group the various criticisms raised against this discipline into three main strands. We will see what issues gave rise to these criticisms and how they may be addressed.
First criticism: Absence of Kalām in the age of the companions
One of the primary criticisms levelled against kalām is that Prophet Muhammad never engaged in or encouraged theological discussions and inquiries. The Prophet's sole preoccupation was to teach the practical precepts and rules of Islam. The Prophet did not say anything that could be interpreted as grounds for the legitimacy of kalām. Moreover, one would also be hard pressed to find any statement from the Companions that could be brought to bear on the question of the efficacy of kalām. Thus, pursuing this discipline would be contrary to the example set by the Prophet and the Companions.
The fallacy of this criticism can be readily demonstrated. The Prophet, in line with the Qur'anic injunction 'and argue with them by virtue of that which is best' (Qur'an 16:125), was constantly debating and arguing with the unbelievers. His attitude in engaging with the unbelievers was informed by the Qur'anic dictum: 'Offer your argument if you are truthful' (Qur'an 2:111; also, Qur'an 21:24; 27:64; 28:75). And the reason for this Qur'anic approach is obvious, because with careful consideration of the contents of the Qur'an and the spirit that pervades its teachings, it will be seen that the Qur'an's call is founded on reason. With regard to the use of rational arguments, the Qur'an itself employs this method. Examples of the application of this method can be found in the Qur'an's use of a rational argument in its refutation of the polytheists' claim as to the divinity of their gods, namely, that one of the requisites of divine lordship is having the power to bestow benefits and being able to guard against affliction and harm [to one's subjects] (Qur'an 17:56 and 34:22). Another example can be found in the Qur'an's argument for the reality of the resurrection and the truth of the call of the Apostle of God to the fact that the creation of the heavens and the earth was not mere idle play (Qur'an 21:16-17). . But if what the critics have in mind is that the Companions never pursued and taught kalām as a codified field of knowledge, the same can be said with regard to such other disciplines as tafsīr (Qur'anic exegesis), ḥadīth (the study of the reports related from the Prophet and the Companions), and fiqh (jurisprudence). None of these disciplines, which all the detractors of kalām find to be perfectly credible and consistent with Islamic faith, had formally taken shape during the time of the Companions (see Ghazālī n.d.,1:96).
If, on the other hand, the thrust of the criticism concerns the use of such novel terms as substance and accident, which were borrowed from Hellenistic philosophy, the criticism would lack any substantive value, for these terms are not critical to the study of kalām; they only facilitate a more convenient way of communicating between the scholars and the students of kalām. Moreover, the same phenomenon is at play in the other disciplines of the Islamic studies, as fiqh and ḥadīth developed terms that were definitely not in use during the time of the Prophet or the Companions (see Ghazālī n.d.,1:95).
But if the criticism intends to point to the fact that theological debates and discussions were not prevalent during the time of the Companions, the answer is clear. In the early years of Islam, there was not much theological and sectarian dissension, and therefore there was no need to engage in theological disputation. Only toward the middle and the end of the age of the Companions did sectarian and theological divisions start to emerge. With the formation of such sects as the Khawārij and the Qadarīyyah, the Companions in fact started articulating sound and robust arguments in critique of the divergent beliefs of these newly formed sects. They would engage in serious debates with the prominent figures of these sects and would attempt to offer reasonable responses to their claims. For example, Abdullah ibn Abbās participated in a debate which he referred to as majles-i sirā' in which he would engage in rational argumentation concerning the issue of destiny or God's providence and divinely ordained 'fate' (qaḍā wa qaḍar) (Al-Janadī 1995, 1:99).
Second criticism: Adverse consequences
The second strand of criticism levelled against the study of kalām is based on the ostensible adverse consequences of this discipline. According to some of the detractors of this discipline, pursuing kalām can only lead to bigotry, greater polarisation and division, and enmity among Muslims, and these are all consequences that are strongly condemned in Islam (see Ghazālī n.d., 1:95).
The fallacy in this line of criticism is that it attributes the adverse consequences entailed by the incorrect methods and presuppositions utilised by some theologians to the discipline of kalām as such. Kalām itself is based on principles derived from the Qur'an and the corpus of tradition, and any negative consequence that may result from the pursuit of kalām is caused by the illegitimate methods and principles employed by some of its practitioners (see Ghazālī n.d.,1:95). When contemplating the legitimacy of kalām as such, what we need to bear in mind is that when the proper methods and principles are utilised, the results arrived at (namely, defending Islamic beliefs and offering cogent and convincing arguments to those who raise questions and criticisms about Islamic doctrine) will be legitimate and reasonable, and this is true of all the branches of Islamic studies. With the other disciplines, too, be it fiqh, ḥadīth, or rijāl, if the practitioner implements incorrect principles and methods, the results and consequences he will arrive at would naturally be invalid and illegitimate. But, it would be incorrect to cite these invalid and illegitimate consequence as grounds for undermining the legitimacy of the disciplines in question. Abū Ḥāmid Ghazālī makes the following observation in this connection: If the study of kalām leads to bigotry, division, rancour, and enmity, these are obviously unacceptable matters, and we must all avoid them. But in the same vein, arrogance, conceit, ostentation, and seeking after social and political power, which are potential consequences that may be attendant on the pursuit of such universally accepted disciplines as ḥadīth, fiqh, and tafsīr, are never cited as grounds for the illegitimacy of the disciplines in question. (Ghazālī n.d., Third criticism: Opposition of major religious leaders The third, and arguably the most significant, criticism directed at the study of kalām is the opposition voiced by prominent Sunni religious leaders against this discipline. To address this criticism, it is essential that we correctly and carefully analyse and examine the expressions of opposition purportedly voiced by the religious leaders in question. It is only through such an analysis that we can discover their true intent.
We will start with an examination of the negative attitude ostensibly exhibited by Abū Ḥanīfah. As extensively demonstrated in the section 'The Advocates of the Legitimacy of Kalām', Abū Ḥanīfah is known more for his support of the study of kalām than for his opposition. The positive remarks attributed to Abū Ḥanīfah that are supportive of the legitimacy of the study of kalām far outweigh those that are deemed to be critical. But as for why Abū Ḥanīfah could have held a positive opinion of kalām despite the negative remarks he occasionally expressed regarding it, the exposition that numerous Sunni scholars have offered of the views of Abū Ḥanīfah in this relation are enlightening. Adducing the reports indicative of Abū Ḥanīfah's positive stance toward kalām, Dhahabī offers the following explanation regarding the negative remarks attributed to him: Abū Ḥanīfah was not opposed to kalām as such but to certain theological topics and to certain practitioners of theology, such as ʻAmrw ibn ʻUbayd, who was one of the prominent figures of the Qadarīyyah and Muʻtazilī schools. He did not reject kalām per se, that is to say, as the discipline responsible for delineating and defending the boundaries of faith. (Dhahabī 1405 AH, 6:104) The same analysis is true of the apparently negative views expressed by Abū Yūsuf Shaybānī regarding kalām, which at first glance seem to imply its illegitimacy. His criticism was mainly aimed at those practitioners of kalām that exploited this study for improper objectives, employed invalid methods, and attacked their theological adversaries as heretics. A careful examination of the modifiers in Abū Yūsuf's remarks makes it clear that he was not opposed to the discipline of kalām as such. Rather, he was averse to such figures as Marrīsī and to such persuasions as the Jahmīyyah and the Muʻtazilah who bickered unduly regarding the divine attributes and accused one another of apostasy. This is how Dhahabī explains Shaybānī's apparent opposition to kalām: The examples Abū Yūsuf has in mind in denouncing kalām are the doubts and objections that arise in consequence of the views of the practitioners of kalām. They would bicker with one another regarding the verses [āyāt] and reports [aḥādīth] bearing on the topic of the divine attributes and would denounce one another as heretics in virtue of their differing views. It was this [problematic] approach that resulted in the emergence of the Muʻtazilī and Jahmī persuasions and the theory of the corporeality of God [tajsīm] and many other woes. (Dhahabī 1405 AH, 8:539) Likewise, Shāfiʻī's denunciation of all practitioners of kalām was motivated by his encounters with Abū Yaḥyā Ḥafṣ Fard. He believed the Qur'an to have been created in time, and it was on this question that Shāfiʻī debated with him (see Dhahabī 1405 AH, 10:30). What Dhahabī's report indicates is that Shāfiʻī's disapproval of kalām and his denunciation of the practitioners of kalām as heretics was aimed only at those who claimed that as the word of God, the Qur'an was a temporal phenomenon or one that was created in time. This qualified opposition was described by some later writers, erroneously, as a categorical opposition, whereas other writers correctly highlighted the qualified scope of Shāfiʻī's opposition. It is from the reports of the latter group of writers that we may accurately understand Shāfiʻī's true stance toward kalām. Ibn ʻAsākir, quoting Bayhaqī, points to this very fact in his elucidation of Shāfiʻī's position, and then he adds, Another critic of the discipline of kalām who is described as exhibiting a very negative attitude toward this discipline is Aḥmad ibn Ḥanbal. A consideration of his times and the historical conditions and events that defined his social environment can be of great help in correctly understanding the disapproving remarks quoted from him regarding the study of kalām. Ibn Ḥanbal's time saw the most intense phase in the dispute between the Ahl al-Ḥadīth (who advocated for a very literalist approach toward an understanding of sacred texts) and the Muʻtazilī theologians, especially with regard to the question of the createdness of the Qur'an. The dispute was so intense that it was characterised as a fitnah -an instance of religious dissension. The debate was such a prominent social phenomenon that in 218 AH the Abbasid caliph Maʼmūn ordered the governor of Baghdad to 'test' the judges, jurisprudents, and reporters of Prophetic tradition with the question of the creation of the Qur'an. Those who believed in the creation of the Qur'an were free to go, but those who believed otherwise were to be reported to the caliph (see Ibn Athīr Jazarī 1417 AH, 5:572). Aḥmad ibn Ḥanbal was one of the figures who resisted the governor's pressure, ultimately leading to his arrest and his forcible transfer to the court of Maʼmūn while bound and in shackles. He remained in prison for his refusal to believe in the creation of the Qur'an until the ascension of Mutiwakkil to the caliphate (see Ibn Athīr Jazarī 1417 AH, 5:576).
Under the caliphate of Mutiwakkil, the tables were turned in favour of the Ahl al-Ḥadīth and to the detriment of the Muʻtazilī theologians. Aḥmad ibn Ḥanbal and all those imprisoned for rejecting the createdness of the Qur'an were freed and belief in the creation of the Qur'an was banned. Mutiwakkil issued a decree, effective in all Islamic lands, that prohibited the study of kalām and the belief in the creation of the Qur'an (see Ibn Kathīr 1424 AH, 14:350). This clearly indicates that in that day and age, kalām was used exclusively to refer to the debate over the creation of the Qur'an.
The foregoing interpretation is further corroborated by another statement made by Ibn Ḥanbal in this relation. He is said to have remarked, 'Those who study kalām will never attain salvation, for they are not immune from being persuaded into accepting the Jahmī position' (Dhahabī 1405AH, 11:29, 1407Ibn Mufliḥ Maqdisī 1417 AH, 1:223). This quote once again shows that the object of Ibn Ḥanbal's opposition was not kalām as the discipline responsible for arguing for and defending the articles of faith but as a harmful preoccupation with such contentious topics as the creation of the Qur'an and the nature of the divine attributes, which he deemed to be heretical. This understanding is reinforced by the historical fact that Ibn Ḥanbal would himself engage in theological debates with theologians belonging to the Muʻtazilah, Qadarīyyah, and Jahmīyyah sects, attempting thereby to put pressure on his theological adversaries and convince the caliph to endorse his theological position on the issues in question.
Discussion
Having analysed the views of the proponents and detractors of kalām, we demonstrated that from the Islamic standpoint, the application of reason to the principal or creedal doctrines of faith is legitimate. This then led us to conclude that denouncing the practitioners of kalām as heretics on the basis of their use of rational argumentation is unfounded. The only genuine ground that the detractors of kalām invoke in support of their stance is the critical statements of the prominent scholars of Islamic jurisprudence and theology regarding the illegitimacy of pursuing the study of kalām. We showed, however, that when viewed in light of the socio-political conditions of their time, these statements are insufficient in proving that the persons in question were in fact opposed to the discipline of kalām as such.
The second and third centuries AH were a time of intense conflict between the Muʻtazilah and the other doctrinal persuasions in Islam. The Ahl al-Ḥadīth movement, represented by such influential figures as Mālik, Shāfiʻī, and Ibn Ḥanbal, was fiercely opposed to the Muʻtazilah. As the application of the term kalām to the discipline responsible for defending the articles of Islamic faith was first popularised by the Muʻtazilah, the designation mutikallim (student or practitioner of kalām) was used by the members of the Ahl al-Ḥadīth movement to refer exclusively to Muʻtazilī theologians. This reference was aided by the fact that the most contentious issue faced by the scholars and that indicated the rift between the Ahl al-Ḥadīth movement and Muʻtazilī theology was the precise nature of 'the word of God' (kalām allāh) as manifested in the Qur'an, and whether it was a temporal phenomenon created in time (ḥādith) or an eternal being (qadīm).
Furthermore, the Muʻtazilites were pejoratively labelled as 'ahl al-ahwā' literally meaning 'those given to whims and desires' because, all other things being equal, they considered the value of rational arguments as being on a par with the Qur'an and the hadith report corpus. Thus, the senior scholars in the 'scripturalist' school (ahl al-hadith) opposed them. This in fact is how the opposition to the science of kalām arose (Ibn Asākir 1404 AH:345). Additionally, the opposition of these jurists against the science of kalām and its advocates was not because they used neologisms such as 'attribute', 'substance', and 'corporeity'. Instead, it was because of [their belief] that they disseminated deviant beliefs which contravened [the teachings of] the Qur'an, the hadith report corpus, and plain reason (see Ibn Taymīyyah 1417 AH:vol. 1, 232-233).
In addition to all this, many of the major Sunni religious leaders, first and foremost among them being Abū Ḥanīfah, described kalām as 'the noblest of the sciences'. A number of them asserted that learning kalām in order to be able to respond to criticisms of Islamic doctrine is obligatory and teaching it to others is mandatory. In light of the inherent purpose of the discipline of kalām, which is to offer a rational account and defence of the doctrines of faith, a favourable stance toward the discipline of kalām as we understand it today is more reasonable and in greater agreement with the Qur'an, with Islamic tradition, and even with the expressed position of the most prominent Sunni jurisprudents and religious leaders, including those who denounce the practitioners of kalām as heretics. For, these seemingly unapologetic detractors were themselves, in a sense, practitioners of kalām, as they would argue against those beliefs and sects that in their view contradicted orthodox doctrine, and to this end they would engage in theological debates and write numerous texts of a theological character.
Conclusion
This study sought to answer the question as to why some of the well-known figures among Sunni scholarship anathemised (pronounced takfīr upon) the practitioners of kalām (mutikallimīn); and whether the subject of their anathemisation included everyone engaged in creedal theology or whether their criticism was limited to specific groups who harboured specific objectives in specific (and more limited) theological controversies.
In response to the above questions, the following points were established in this article: 1. The prohibition of kalām and the excommunication of its practitioners by scholars such as Abū Hanīfa Nuʻmān ibn Thābit, Mālik ibn Anas, Muhammad ibn Idrīs al-Shāfʻī, and Ahmad ibn Hanbal, and the like, has nothing to do with the later concept of kalām as it is commonly understood among Muslim scholarship. Rather, the main objective of those who advocated such prohibitions was to oppose a particular side in certain controversies such as the one over the createdness of Qur'an as the word of God, and/or targeting a specific group among the mutikallimīn (Kalam practitioners), who disseminated certain (believed to be heretic) beliefs of specific sects such as the Muʻtazilites, the Qādirīya, and the Jabrīya. The intention was not an all-encompassing ban on kalām as the science of understanding doctrinal beliefs and defending religious dogma. Kalām as a general category of that discipline was never a controversial issue. , and so on, consider the study of kalam as a legitimate practice and [some of them] have even issued religious verdicts (fatāwā) that learning the science of kalām for apologetic purposes, that is, defending the religion and its creeds in the face of disbelievers' intellectual attacks, is a religious obligation (wājib).
3. Objections and counterarguments such as 'that the Prophet and his companions did not practise Kalam' or the ones based on the negative consequences of studying it, and so on, were thoroughly examined. It was concluded that these objections do not serve as plausible grounds for the prohibition against kalam, and that the arguments for its legitimacy prove to be stronger and more logical, considering their compatibility with the verses of the Qur'an, the hadith report corpus, and the tradition-setting practice of the early companions of the Prophet, as well as the scholars in the formative period of Islam (ulamā-i salaf).
|
2021-11-11T16:20:44.452Z
|
2021-11-05T00:00:00.000
|
{
"year": 2021,
"sha1": "bf5ea5b9042db610721c82e7a6fd3f9ac81c6d72",
"oa_license": "CCBY",
"oa_url": "https://hts.org.za/index.php/hts/article/download/6917/20761",
"oa_status": "GOLD",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "512605443596623f11f8ff1011d9566ddd4fa76d",
"s2fieldsofstudy": [
"Philosophy"
],
"extfieldsofstudy": []
}
|
268635109
|
pes2o/s2orc
|
v3-fos-license
|
Is there a role for video-assisted parathyroidectomy in regions with high prevalence of goitre?
Minimally-invasive procedures for parathyroidectomy have revolutionized the surgical treatment of primary hyperparathyroidism (pHPT). Coexistence of goitre is considered a major contraindication for these approaches, especially if unilateral. A specific advantage of video-assisted parathyroidectomy (VAP) compared to other endoscopic techniques is the possibility to combine it with thyroidectomy when necessary and when the selection criteria for video-assisted thyroidectomy (VAT) are met. We evaluated the role of VAP in a region with a high prevalence of goitre. The medical records of all patients who underwent parathyroidectomy and concomitant thyroid resection in our Division, between May 1998 and June 2012, were reviewed. Patients who underwent VAP and concomitant VAT were included in this study. Overall, in this period, 615 patients were treated in our Division for pHPT and 227 patients (36.9%) underwent concomitant thyroid resection. Among these, 384 patients were selected for VAP and 124 (32.3%) underwent concomitant VAT (lobectomy in 26 cases, total thyroidectomy in 98). No conversion to conventional surgery was registered. Mean operative time was 66.6 ± 43.6 min. Transient hypocalcaemia was observed in 42 cases. A transient recurrent nerve lesion was registered in one case. No other complications occurred. Final histology showed parathyroid adenoma in all but two cases of parathyroid carcinoma, benign goitre in 119 cases and papillary thyroid carcinoma in the remaining 5 patients. After a mean follow-up of 33.2 months, no persistent or recurrent disease was observed. In our experience, a video-assisted approach for the treatment of synchronous thyroid and parathyroid diseases is feasible, effective and safe at least considering short-term follow-up.
Introduction
The last two decades have seen a revolution in the field of surgical treatment of sporadic primary hyperparathyroidism (pHPT), at least in part due to advances in technology 1 2 .The evolution of preoperative imaging studies, the availability of the quick parathyroid hormone assay (qPTHa) and the introduction of cervicos-copy gave great impulse to the application of several variants of minimally-invasive techniques for parathyroidectomy.To date, a minimally-invasive approach to parathyroidectomy emerged as a valid and validated option to treat selected cases of sporadic pHPT, at least in referral centres 3 , and are assuming an increasingly important role 4 .
Nonetheless, the diffusion of the minimally-invasive parathyroidectomy has resulted in several controversies regarding the indications for these approaches 5 .One of the criticisms regards the role of minimally-invasive parathyroid access in the treatment of a concomitant thyroid pathology [6][7][8] , which could represent a relevant limit for the diffusion of these approaches, especially in regions with a high prevalence of goitre.Indeed, in countries with endemic goitre, concomitant thyroid disease is found in 35-78% of patients with pHPT 7 .Miccoli 9 first described video-assisted parathyroidectomy (VAP), and in 1998 our department adopted the procedure 10 .Early after its first description, this technique encountered large, worldwide acceptance [10][11][12][13][14][15] as it is easy to reproduce in different surgical settings.Indeed, it reproduces all the steps of a conventional procedure, with the endoscope representing a tool that allows performing the same operation through a smaller skin incision 10 .Thanks to its central access, VAP, in selected cases, allows performing bilateral neck exploration and thyroid resection when necessary through the same central access 3 10 16 .The aim of this study was to evaluate the role of VAP in a region with a high prevalence of goitre.
Materials and methods
The medical records of all patients who underwent parathyroidectomy for pHPT and a concomitant thyroid lobectomy (TL) or total thyroidectomy (TT), between May 1998 and June 2012, at our institution were reviewed.Patients who underwent VAP and concomitant video-assisted thyroidectomy (VAT) were included in this study.Demographic, clinical, surgical, pathologic and follow-up data of these patients were evaluated.Statistical analysis was performed using a commercially available software package (SPSS 10.0 for Windows; SPSS Inc., Chicago, Ill).The χ 2 test was used for categorical variables, and a t test was used for continuous variables.A p value < 0.05 was considered statistically significant.
Preoperative work-up
Preoperative high-resolution ultrasound (US) and 99 mTcsestamibi (MIBI) scans were performed in all patients with sporadic pHPT to localize hyperfunctioning glands.All patients had normal renal function (serum creatinine value ranging from 0.7 to 1.2 mg/dl).
Indications
Ideal candidates for VAP are those with sporadic pHPT in whom a single adenoma is suspected basing on preoperative MIBI-scan and ultrasonography.Parathyroid adenomas larger than 3 cm in their maximum diameter should not be selected for VAP, because a difficult dissection can lead to dangerous capsular rupture.Patients with concomitant nodular goitre requiring surgical removal can be selected for VAP if the inclusion criteria for the video-assisted thyroidectomy (VAT) are respected 10 17 .Basing on the surgeons' experience, in selected cases, patients with previous contralateral neck surgery or intrathymic/retrosternal adenomas can be selected for VAP.In case of suspected multiple gland disease, a video-assisted bilateral exploration can be planned.
Intraoperative-PTH (IOPTH) monitoring
All patients underwent IOPTH monitoring.Blood samples were collected peripherally pre-incision (preoperative baseline concentration -PTH), pre-excision (after dissection and just before clamping the blood supply of the suspected affected gland -PTH) and at 10 min (PTH-10) and 20 min (PTH-20) after gland excision.A point of care chemiluminescence immunoassay system (Stat-IntraOperative-intact PTH -Future Diagnostics, Wijchen, The Netherlands) was used for IOPTH measurements and set up in the recovery room.Blood specimens were collected and analyzed following the manufacturer's indications 18 .
Surgical technique
The surgical technique we use has been previously described elsewhere in detail 19 .The patient, under general or loco-regional anaesthesia with cervical block 20 21 , is positioned supine with the neck in slight extension.A small (1.5 cm) horizontal skin incision is made between the cricoid cartilage and the sternal notch, in the midline.The cervical linea alba is opened as far as possible.The technique is completely gasless.The endoscope (5 mm, 30°) and the dedicated small surgical instruments are introduced through the skin incision without any trocar utilization (Fig. 1).After identifying the inferior laryngeal nerve in the involved side, a targeted exploration is usually carried out to identify the abnormal gland.The magnification (2-3 folds) of the endoscope permits a easy identification of the nerve and the parathyroid glands, if the principles of blunt and bloodless dissection are respected.In case of suspicion of multiglandular disease, bilateral parathyroid exploration can be accomplished.The gland should not be grasped, in order to avoid any capsular rupture.After cutting the pedicle, the adenoma is extracted through the skin incision.When video-assisted thyroidectomy is required, dissection of the thyroid gland is safely performed under endoscopic vision according to the technique previously described [17][18][19] .
Follow-up schedule
During follow-up, calcium levels combined with intact PTH (iPTH) values were examined at 1, 3 and 6 months after the operation, and the following classification criteria for postsurgical resection outcome at 6 months were used: group 1: cured patients/operative success: normal or low serum calcium levels for at least 6 months after parathyroidectomy; group 2 -patients with disease persistence/operative failure: persistent hypercalcaemia and elevated iPTH levels (> 65 pg/ml; 6.89 pmol/l) within 6 months after surgery.
Results
Between May 1998 and June 2012, at our institution 615 patients underwent surgery for primary HPT.Among these, 227 patients (36.9%) underwent concomitant thyroid resection.Three hundred and eighty four patients with PHPT who fulfilled the inclusion criteria were selected for VAP.Among these, concomitant VAT was carried out in 124 patients (32.Final histology showed parathyroid adenoma in all but two cases of parathyroid carcinomas.In particular, in these latter two cases, no signs and/or suspicions of malignancy were present on the basis of pre-operative evaluation; in both cases simultaneous videoassisted total thyroidectomy for a concomitant bilateral thyroid disease was performed.Final histology of thyroid specimens revealed benign goitre in 119 cases and papillary thyroid carcinoma in 5 cases.The preoperative cytological examination of the five cases of papillary thyroid carcinoma was consistent with indeterminate nodules.After a mean follow-up of 33.2 ± 20.0 months (range: 3-110), no persistent recurrent parathyroid or thyroid disease was observed.
Discussion
The results of VAP in terms of cure and complication rates are similar and rival those of conventional surgery.This has been extensively demonstrated in the international literature 3 4 16 , and it is also confirmed in our personal experience 10 18 19 .Some criticisms about the technical aspects for MIVAP have concerned the operating time.However, it has been demonstrated in large retrospective series 10 14 as well as in small comparative, randomized trials 3 22 that the operating time does not represent a limit.A prospective, randomized, small comparative study (level II evidence) showed that the operating time for MIVAP was significantly shorter than conventional bilateral exploration and similar to open minimally-invasive parathyroidectomy 22 .
Fig. 1.Video-assisted parathyroidectomy: two small conventional retractors are used to medially retract the thyroid lobe and laterally retract the strap muscles to maintain the operative space.The endoscope (5 mm, 30°) and dedicated small surgical instruments are then introduced through the skin incision without any trocar utilization.In addition, in our overall experience with exclusive VAP and IOPTH the mean operating time was 42.6 ± 18.1 min (range: 15-95).Besides its reproducibility, VAP also seems to offer significant advantages over conventional surgery in terms of patient satisfaction with the cosmetic result and postoperative recovery 3 .Moreover, it has been demonstrated that in selected cases of pHPT, VAP is curative and safe considering short-and medium-term outcome and complications 10-14 22 23 .The association between benign and malignant thyroid disease with pHPT has been previously described [5][6][7][8] .Indeed, the presence of concomitant thyroid disease has been reported in 15%-70% of patients with pHPT 24 25 .In a surgical series of 51 patients with pHPT, Kösem described 43 cases (84.3%) of coexistent thyroid diseases.Among these, 9 patients (17.6%) had papillary thyroid cancer 26 .The association between pHPT and goitre seems to be more much relevant in areas with iodine deficiency.It has been reported that in an endemic goitre region, 67 of 137 patients with PHPT (49%) had concomitant thyroid disease 7 .In a large retrospective analysis involving a population living in a geographical area with an average-mild iodine deficiency, such as the mixed Italian regions, a total of 124 of 241 patients (51.5%) with pHPT showed concomitant thyroid disorders 27 .It has been pointed out that a head and neck endocrine surgeon needs to be aware of the possible coexistence of thyroid and parathyroid disease so that, when encountered, they can be safely and effectively managed in a single procedure 24 25 .However, some authors have argued that concomitant thyroid disease requiring surgical management represents an exclusion criteria for minimally-invasive parathyroidectomy 5 6 .Perrier 5 reported that 17% of patients referred to parathyroidectomy were considered ineligible for localized parathyroid surgery because of a coexisting thyroid pathology.However, a relevant advantage of central access is the possibility to perform thyroid resection, even bilateral, when necessary.This introduces an important difference when compared to other minimally-invasive techniques, as conversion to a conventional approach is usually required when bilateral thyroid resection is needed 28 .Because of the high prevalence of multinodular goitre in some countries, the central approach of VAP can allow experienced surgeons to increase the number of patients eligible for a video-assisted procedure.Indeed, in our experience in a region of high prevalence of goitre in Italy, the video-assisted approach was successful accomplished in the treatment of both parathyroid and thyroid disease during the same procedure in a significant percentage of patients (32.3%).Moreover, concomitant thyroid diseases with pHPT may result in more difficult preoperative localization of the pathological parathyroid gland.Indeed, in these cases, one of the main problems is the accuracy of preoperative imaging studies.It has been reported that the coexistence of pHPT and thyroid disease may reduce the sensitivity of sestaMIBI from 81% to 75%, and of US from 73% to 55% 29 .However, the combined use of sestaMIBI and US seems to be more effective in patients with pHPT for localization of an enlarged parathyroid gland even in the case of concomitant thyroid disease 29 .This would make it possible to perform a focused parathyroidectomy in the most of patients suffering from pHPT even in an endemic goitre region 7 .Moreover, it might be stressed that even if preoperatively, well-documented single adenoma is a prerequisite for any type of focused parathyroidectomy 1 30 , the concept that preoperative localization studies are mandatory when performing a minimally-invasive approach remains valid only for procedures that imply a unilateral approach.Among these, the lateral approach (VAP-LA) described by Henry which is the most diffuse, has been shown to be effective and safe, with a minimal complication rate 31 .Nonetheless, due to its unilateral approach, the rate of contraindications for VAP-LA appears fairly high (43%) compared to that of large series of VAP 23 .This may depend on the need of stricter eligibility criteria, due to the unilaterality of the approach, which implies the strong demonstration of a single enlarged parathyroid gland on preoperative imaging studies and excludes bilateral thyroid disease.The main technical limitation of the technique is, indeed, that a unilateral approach prevents the possibility to accomplish bilateral exploration when necessary without conversion to an open conventional procedure 28 30 31 .On the other hand, a relevant merit of VAP is the possibility of performing bilateral neck exploration when necessary through the same central access.This characteristic in part explains the lack of conversion in the present series and the very low conversion rate generally reported for VAP 10 16 .Moreover, the possibility of performing a bilateral neck exploration produces two main effects on the restrictive inclusion criteria.First, at least from a theoretical point of view, VAP can be performed if IOPTH monitoring is not available or in the case of inadequate preoperative localization, given the ability to explore all parathyroid sites through this approach 10 23 32 .In our experience in an endemic goitre area, 62.4% of patients with PHPT were eligible for VAP.The possibilities given by central access allowed us to treat a significant percentage of patients with concomitant thyroid disease who fulfilled the indication criteria of VAT.No conversion to conventional surgery was necessary and no definitive complications occurred.After a mean follow-up of 33.2 months, no persistent or recurrent parathyroid or thyroid diseases was observed.
Conclusions
When selection criteria are followed, the treatment of synchronous thyroid and parathyroid disorders can be accomplished using a minimally-invasive procedure.In our experience, a video-assisted approach is feasible, effective and safe for the treatment of synchronous thyroid and parathyroid disorders, at least considering short-term follow-up.
3%): 105 females and 19 males, with a mean age of 60.8 ± 10.3 years (range: 38-84).Indications for concomitant thyroid resection were indeterminate nodule within bilateral goitre in 31 cases and multinodular goitre (MNG) with compressive symptoms in the remaining 93 cases.The mean diameter of the dominant nodule of the 93 cases of MNG was 20.3 ± 9.8 mm (range: 2-46 mm).Video-assisted TL was performed in 26 cases and video-assisted TT in 98 cases.No conversion to conventional surgery was necessary.The mean operative time of all procedures was 66.6 ± 43.6 min (range: 30-175).In particular, the mean operative time of VAP associated to video-assisted TL was 63.19 ± 38.1 min (range: 30-170); the parathyroid lesion was ipsilateral to the side of the thyroid lobectomy in 19 cases and contralateral in the remaining 7 cases.The mean operating time for VAP associated with video-assisted TT was 68.27 ± 32.5 min (range: 40-175).Transient hypocalcaemia was observed in 42 cases (32.8%), transient recurrent lesion in one case (0.8%).No other complications occurred.Mean postoperative hospital stay was 3.4 ± 1.7 days (range: 3-7).Most patients (93%) considered the cosmetic result as excellent as evaluated by a verbal response scale (Fig. 2a-b).
Fig. 2a .
Fig. 2a.Cosmetic result of a VAP at two weeks after surgery.
Fig. 2b .
Fig. 2b.Cosmetic result of a VAP at six months after surgery.
|
2016-05-18T11:21:30.367Z
|
2013-12-01T00:00:00.000
|
{
"year": 2013,
"sha1": "00d5173c70d551be59746fad362d198bcac3859d",
"oa_license": "CCBYNCND",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "00d5173c70d551be59746fad362d198bcac3859d",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
484474
|
pes2o/s2orc
|
v3-fos-license
|
Morphology and Ciliary Motion of Mucosa in the Eustachian Tube of Neonatal and Adult Gerbils
The Eustachian tube is a small canal that connects the tympanic cavity with the nasal part of the pharynx. The epithelial lining of the Eustachian tube contains a ciliated columnar epithelium at the tympanic cavity and a pseudostratified, ciliated columnar epithelium with goblet cells near the pharynx. The tube serves to equalize air pressure across the eardrum and drains mucus away from the middle ear into the nasopharynx. Blockage of the Eustachian tube is the most common cause of all forms of otitis media, which is common in children. In the present study, we examined the epithelial lining of the Eustachian tube in neonatal and adult gerbils, with a focus on the morphological and functional development of ciliated cells in the mucosa. The length of the tube is ∼8.8 mm in adult gerbils. Scanning electron microscopy showed that the mucosal member near the pharyngeal side contains a higher density of ciliated cells and goblet cells than that near the tympanic side. The cilia beat frequency is 11 Hz. During development, the length of the Eustachian tube increased significantly between postnatal day 1 (P1) and P18. Scanning electron microscopy showed that the mucosa contained a high density of ciliated cells with a few goblet cells at P1. The density of ciliated cells decreased while the density of goblet cells increased during development. At P18, the mucosa appeared to be adult-like. Interestingly, the ciliary beat frequency measured from ciliated cells at P1 was not statistically different from that measured from adult animals. Our study suggests that the Eustachian tube undergoes significant anatomical and histological changes between P1 and P18. The tube is morphologically and functionally mature at P18, when the auditory function (sensitivity and frequency selectivity) is mature in this species.
Introduction
The Eustachian tube (ET) is a small canal that connects the tympanic cavity with the nasal part of the pharynx. The ET acts as a passageway to ventilate the tympanic cavity and allows equalization of pressure between the middle ear and pharynx, which is necessary for normal hearing [1]. Additionally, it drains mucus away from the middle ear into the nasopharynx. Abnormal or impaired function(s) of the ET may cause pathological changes in the middle ear. Blockage of the ET is the most common cause of all forms of otitis media [2], which is common in children, partly because the ET is not completely developed.
Although the development of some parts of the murine middle ear has been described [3,4], to date, knowledge of the morphological and functional development of the ET mucociliary system is sketchy. In this study, our goal was to examine the epithelial lining of the ET in neonatal and adult gerbils, with a focus on the morphological and functional development of ciliated cells in the mucosa. We examined anatomical and ultrastructural changes of the ET and its epithelial lining, the mucosa, in neonatal and adult gerbils. Since ciliary motion of the mucosa is important for drainage of the ET and impaired mucociliary function of the ET and respiratory tract mucosa is associated with secretory otitis media [5], we also measured ciliary beat frequency of the ciliated cells at different locations along the ET in neonatal and adult gerbils using a photodiode-based displacement measurement system. Our study not only showed how the ET and its mucosa develop, but also provides a morphological and functional basis for future animal studies concerning the pathogenesis of otitis media in earlier stages using the murine model.
Animals
Neonatal gerbils aged between postnatal day 1 (P1) and P30 were used for the experiments. The animals were euthanized by an overdose of sodium pentobarbital (200 mg/kg, IP) followed by decapitation. The animals were used according to a protocol approved by the Institutional Animal Care and Use Committee of Creighton University.
Histological Examinations of the ET
Temporal bone was isolated at the ages of P1, P8, P12, P18, and P30. The tissue was fixed in 4% paraformaldehyde in PBS at 4uC overnight. After being washed with PBS, the tissue was embedded in celloidin. Serial cross sections (10 mm) were prepared with a microtome (CM3050, Leica, Nussloch, Germany) and collected on microslides. The sections were stained with hematoxylin and eosin. After washing with PBS solution, the slides were mounted with a glycerol-based mounting solution and examined under a microscope. The length of the ET was determined using ImageJ (version 1.47) from captured images.
Scanning Electron Microscopy
For scanning electron microscopy (SEM), the temporal bone tissues were fixed with 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) containing 2 mM CaCl 2 , washed in PBS, then post-fixed for 15 minutes with 1% OsO 4 in the same buffer and washed. The tissues were dehydrated in an ethanol series, critical point-dried from CO 2 , and sputter-coated with gold. The tissue was then examined using an FEI Quanta200 scanning electron microscope.
Measurements of Ciliary Motion
The ET was dissected out and sectioned along its longitudinal length. The preparation was bathed in L-15 medium (Invitrogen), containing 136 mM NaCl, 5.8 mM NaH 2 PO 4 , 5.4 mM KCl, 1.4 mM CaCl 2 , 0.9 mM MgCl 2 , 0.4 mM MgSO 4 , and 10 mM HEPES-NaOH (pH 7.4, 300 mmol/l) in an experimental chamber mounted on the stage of a Leica upright microscope. The tissue was firmly attached to the bottom of the chamber by the weight of two thin platinum rods (0.5 mm in diameter), with one of their ends anchored in two small droplets of vacuum grease on the bottom of the chamber. The tissue was mounted with the hair bundle facing upward toward the water-immersion objective. The cilia were imaged using a 636 water immersion objective (Leica) and magnified by an additional 206relay lens. Ciliary motion was measured and calibrated by a photodiode-based measurement system [6,7] mounted on the Leica upright microscope. The magnified image of the cilia was projected onto a photodiode through a rectangular slit. Spontaneous cilia motion modulated the light influx to the photodiode. The photocurrent response was calibrated to displacement units by moving the slit a fixed distance (0.5 mm) with the image of the cell in front of the photodiode. After amplification, the photocurrent signal was low pass-filtered by an antialiasing filter before being digitized by a 16-bit A/D board (Digidata 1322; Molecular Devices). The photodiode system had a cutoff (3 dB) frequency of 1,100 Hz. The motile responses were filtered at 200 Hz and digitized at 10 kHz. Spontaneous ciliary motion was acquired in a four-second window and five of such response segments were obtained in each run. The power spectrum of the response was averaged and analyzed in the frequency domain using Clampfit software. The experiments were done at room temperature (2262uC).
Morphology of the ET and Its Mucosal Epithelium in Adult Gerbils
The auditory system is mature after P20 in gerbils. We measured the length of the ET in the cross-section preparation ( Fig. 1) of the adult gerbil at P30. The length of the ET, measured between the tympanic and pharyngeal ends (marked by arrows in Fig. 1), was 0.8860.9 mm (n = 7). The ET is often divided into two portions: the osseous portion and the cartilage portion. The osseous portion constitutes approximately one-third of the entire length of the ET. We examined the histological architecture of the mucosal epithelium of the ET in the osseous portion near the tympanic opening and the cartilage portion near the pharyngeal opening from three animals using light microscopy. The mucosal epithelium near the pharyngeal opening is lined with pseudostratified ciliated columnar epithelium composed of ciliated and goblet cells. There is a gradual change to a mixture of ciliated, goblet, and squamous cells near the tympanic opening. The density of ciliated cells and goblet cells near the pharyngeal opening appears to be higher than that in the area near the tympanic opening.
We examined the ultrastructural change of the mucosal epithelium in two locations approximately 60 mm near the tympanic and pharyngeal openings from three animals using SEM. As illustrated in Fig. 2, the mucosal epithelium contains ciliated cells and goblet cells in both regions. However, the density of ciliated cells and goblet cells in the different regions are different. Consistent with the light microscopic observation, the mucosa in the region near the pharyngeal opening contains a higher density of ciliated cells and goblet cells. The region near the tympanic opening has fewer ciliated cells. We counted the total number of ciliated cells in a 150675 mm area in both regions from three ET preparations from three gerbils. The number of ciliated cells is 126616 and 61610 for the pharyngeal and tympanic regions, respectively. Thus the density of ciliated cell is significantly higher in the pharyngeal region than in the tympanic region (p, 0.01, student's t-test). More squamous cells were observed in this area. Squamous cells have short blunt microvilia. We also observed absorption of cilia in ciliated cells in both regions. In all three samples examined, the absorption of cilia all started from the center of the cilia (marked by arrows in Fig. 2D). However, we did not observe any signs of the growth of new cilia. The fate and identity of the ciliated cells after the cilia were absorbed are unclear.
Ciliary Beat Frequency of Ciliated Cells
Ciliary motion of the mucosal epithelium in the ET is important for clearance. We measured and compared ciliary beat frequency from ciliated cells located in the osseous and cartilage portions of the ET from five adult gerbils (P30). Fig. 3A shows three representative response waveforms of ciliary motions from three different ciliated cells near the pharyngeal orifice in the cartilage portion. As shown, different cells exhibit ciliary motions with different frequencies, magnitudes, and patterns. We analyzed the beat frequency using a fast Fourier transform. The frequency spectrum of the responses is presented in Fig. 3B. All three responses have a main frequency component between 8 to 15 Hz, with several harmonics at higher frequencies in the spectrum. Fig. 3C presents the means and standard deviations of the frequency of the main frequency component measured from 15 cells in the cartilage portion. For comparison, we also measured the frequency of the main frequency component from 15 ciliated cells in the osseous portion and 12 cells in the airway tissue [7]. As shown in Fig. 3C, the beat frequency of the ciliated cells in the osseous and cartilage portions is not statistically different (p.0.05). No statistical difference (p.0.05) in beat frequency was found between the ciliated cells in the ET and airway tissue, either.
Development of the ET and Its Mucosal Epithelium
Gerbils, like mice and rats, are altricial animals and their auditory system develops within the first three weeks after birth. We examined anatomical and morphological changes of the ET between P1 and P30 from five animals for each age group. Fig. 4 shows the development of length of the ET between P1 and P30.
As shown, the length of the ET grows significantly during this time window. The significant growth occurs after P8; by P18, the length of the ET is not statistically different from that of the adult animals (P.0.05).
We also examined the ultrastructural changes of the mucosal epithelium in the osseous and cartilage regions of the tube at P1, P8, P18, and P30 using SEM. Three animals for each age group were examined. As illustrated in Fig. 5, the mucosal epithelium undergoes significant change during this time window. At P1, both regions were covered by densely populated ciliated cells; only a few goblet cells were seen. The density of the ciliated cells decreased while the density of goblet cells increased during development in the cartilage region near the pharyngeal orifice. In the osseous portion, there was an increase in the density of squamous cells and a reduction in the density of the ciliated and goblet cells. We were unable to count the ciliated cells at P1 due to the fact that the mucosal epithelium was almost completely covered by cilia. However, we counted the total number of ciliated cells in the two regions at P8 and P18 from three ET preparations. In both regions, there was a significant reduction of the density of ciliated cell between P8 and P18 (p,0.05). By P18, the mucosal membrane reached adult appearance.
Ciliary Beat Frequency of Ciliated Cells in Neonatal Gerbils
Although ciliary beat frequency has been measured in adult animals, the development of ciliary motion has not been examined. We measured ciliary motion from ciliated cells in the cartilage region at different time points during development. Fig. 6A shows some representative response waveforms from three different cells at P1. Similar to the responses seen in adult animals, the ciliary motion in neonatal gerbils also exhibited different patterns, frequencies, and magnitudes. The frequency spectrum of the response is presented in Fig. 6B. Fig. 6C illustrates the mean and standard deviations of the beat frequency (main frequency component) measured from 12 cells in the cartilage portion at P1, P8, P18, and P30. As shown, the beat frequency at P1 is not statistically different from that of P8, P18, and P30. In other words, the ciliary motion is developed at birth.
Discussion
We used morphological and electrophysiological methodology to examine the mucosal epithelium in neonatal and adult gerbils. Gerbil is one of the most commonly used animal models for auditory research. Like other rodents, the middle ear and inner ear of gerbils develop after birth [8]. Previous studies using developing gerbils showed that cochlear microphonic responses were first recorded at P12, with thresholds exceeding 100 dB SPL, suggesting that the onset of hearing occurs around P12 [9]. Cochlear microphonic thresholds subsequently improved rapidly across the responsive frequency range, achieving adult levels by P18 [9]. Although other studies suggest that middle ear and inner ear functions (such as cochlear potentials and basilar membrane mechanics) continue to improve beyond P30 [10,11], it is generally believed that the gerbil auditory system is functionally and morphologically mature at P18. Although we did not examine gross morphology of the middle ear, our study shows that the development of the ET, which is considered part of the middle ear, occurs before P18. The most dramatic increase in ET length is observed between P8 and P18. This period is associated with the final stages of resorption of middle ear mesenchyme and ossicular ossification, and accounts for approximately a 25 dB loss in sensitivity due to middle ear immaturity [12]. As one of the major roles of the ET is to ventilate the tympanic cavity to allow equalization of pressure between the middle ear and pharynx, concurrent development of the ET is necessary for functional maturation of the auditory system in this species.
Using SEM, we show that there is a gradual change of the mucosal membrane along the ET in both neonatal and adult animals. The structure of the gerbil mucosa shows striking similarities to that of mice, rats, guinea pigs, dogs, and humans [13][14][15][16]. In the cartilage region near the pharyngeal opening, there is a higher density of ciliated cells and goblet cells. The density of ciliated cells is reduced significantly in the tympanic region. Such gradual change of the ciliated cell density in mucosal epithelium has been reported before. Interestingly, we observed a dynamic change of the cilia in both regions; such change is reflected by resorption of cilia (Fig. 2). It is not clear, however, whether such resorption of cilia is a sign of the beginning of the renewal of the cilia or the cell. Mature cells in the epithelium have a different life span; the turnover time of ciliated cells and goblet cells in guinea pigs is estimated to be six days [17].
Ultrastructural examination shows that the mucosal epithelium of the ET contains a higher density of ciliated cells at birth. Studies in rats and mice have shown that development of ciliated cells starts at the sixteenth gestation day [3,[18][19][20], one day earlier than the secretory cells appear in the ET and middle ear. The number of ciliated cells and secretory cells increases rapidly after birth. Although we did not examine the development of the mucosal epithelium before birth, we observed a higher density of ciliated cells in the ET in neonatal animals than in the adult animals; the high density of ciliated cells remained until P8. However, contrary to the study by Park and Lim in mice [20], we observed a significant reduction of ciliated cells between P8 and P18. In the area near the tympanic opening, the reduction was even more dramatic. The significant remodeling of the mucosal epithelium after P8 is more consistent with the ultrastructure of mucosa observed in adult animals.
Fluid and debris are continually being cleared and drained from the middle ear through the ET by mucociliary clearance [21,22]. The mucociliary transport system is a major defense mechanism of the middle ear. Direct measurement of mucociliary clearance in the ET of small animals such as gerbils has never been done. To measure this clearance in neonatal gerbils is more difficult since the middle ear is filled with mesenchyme before P10. Thus, the performance of mucous transportation in the middle ear and ET is often represented by measuring ciliary motion. Cilia beat frequency has been measured in animals [23] and humans [5,24,25] under normal and pathological conditions. The beat frequency ranges from 8 to 15 Hz. We demonstrated that the mean beat frequency of the ciliated cells in the ET was 11 Hz, consistent with that seen in biopsy tissue obtained from the middle ear of children [24]. Ciliary motion is known to be sensitive to temperature, pH, calcium, and ATP concentration [26][27][28]. Drugs (e.g., cocaine, adrenaline), smoking (nicotine), infections (viral or bacterial), and noxious fumes (e.g., sulfur, carbon monoxide) all inhibit the normal ciliary beat [28]. Under our experimental condition, the beat frequency of cilia in the different regions of the ET was similar, despite the fact that there was gradual change in cilia density along the length of the ET. We further showed that cilia beat in the ET was not significantly different from that seen in the airway [7]. The methods that have been used most frequently to measure ciliary beat frequency in the middle ear and airway tissues are the cinematograph, high-speed digital camera [29,30], and more recently, the optical flow technique [31] and optical coherence tomography [32]. We used the photodiode technique to measure cilia motion. This technique has some advantages over other techniques. While other techniques can quantitatively measure beat frequency, complex response waveforms with harmonic components are often missed. We showed, with power spectrum analysis, that the cilia motion was complex, with different frequencies, magnitudes, and patterns. The disadvantage of the photodiode technique is that the measurement can only be made from one cell at a time, whereas the optical flow technique and optical coherence tomography can capture motions of all cells in a larger area. Optical coherence tomography has recently been used to measure functional dynamics of cilia and mucus flow on the airway epithelium [32].
Cilia beat frequency of the ciliated cells from the middle ear has not been measured from developing animals. Therefore, it is unclear whether cilia motion is mature at birth or still undergoes development in the first three weeks after birth. We measured cilia motion at different time points at two different areas in the ET during development and showed that cilia beat frequency measured at birth was not statistically different from that measured from adult animals. The magnitude and pattern of motion are compatible among different age groups. Combined with the ultrastructural evidence using SEM, we conclude that ciliated cells are morphologically and functionally mature at birth.
In summary, we examined the morphology and function of the mucosa in the ET of neonatal and adult gerbils in the present study. We show that the ET undergoes significant anatomical and histological changes between P1 and P18. The tube is morphologically and functionally mature at P18, when the auditory function is mature in this species. Although the anatomy (in terms of length) and development of the ET in humans and gerbils are different, our study provides important information about ultrastructural and functional changes of the mucosa in the ET of neonatal and adult gerbils, which may help to understand the development and function of the ET in humans.
|
2018-01-22T13:44:05.837Z
|
2014-06-12T00:00:00.000
|
{
"year": 2014,
"sha1": "04160aa2ee9a4e6558612cd3f1201f43a31c4804",
"oa_license": "CCBY",
"oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0099840&type=printable",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "04160aa2ee9a4e6558612cd3f1201f43a31c4804",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
53677715
|
pes2o/s2orc
|
v3-fos-license
|
Deformations of minimal cohomology classes on abelian varieties
We show that the infinitesimal deformations of the Brill--Noether locus $W_d$ attached to a smooth non-hyperelliptic curve $C$ are in one-to-one correspondence with the deformations of $C$. As an application, we prove that if a Jacobian $J$ deforms together with a minimal cohomology class out the Jacobian locus, then $J$ is hyperelliptic. In particular, this provides an evidence to a conjecture of Debarre on the classification of ppavs carrying a minimal cohomology class. Finally, we also study simultaneous deformations of Fano surfaces of lines and intermediate Jacobians.
Introduction
Given a smooth complex curve C of genus g ≥ 2 with Jacobian J, we denote by C d (d ≥ 1) the d-fold symmetric product of C and by the Abel-Jacobi map defined up to the choice of a point Q ∈ C. In the papers [Ke2] and [Fa] the infinitesimal deformations of C d and f d are studied: these are in one-to-one correspondence with the deformations of C if and only if g ≥ 3. In particular, there are isomorphisms of functors of Artin rings On the other hand, the computation of infinitesimal deformations of the images of Abel-Jacobi maps, namely the Brill-Noether loci parameterizing degree d line bundles on C having at least one non-zero global section, is a problem that has not been studied yet in its full generality and has interesting relationships to a conjecture of Debarre (see Conjecture 1.3). Previous calculations of deformations of Brill-Noether loci have been performed only for Theta divisors Θ ≃ W g−1 of non-hyperelliptic Jacobians where the authors of [SV] prove that the firstorder deformations of C inject in those of W g−1 .
One of the main difficulties for the computation of deformations of Brill-Noether loci is that in general these spaces are singular. However, as shown by work of Kempf [Ke1], the singularities of W d are at most rational and a resolution of its singularities is provided by the Abel-Jacobi map u d : C d → W d which factorizes f d . By extending a construction of Wahl in [Wa,§1] for affine equisingular deformations, this allows us to define a "blowing-down deformation" morphism of functors of Artin rings u ′ d : Def C d −→ Def W d sending an infinitesimal deformation C d of C d over an Artinian local C-algebra A, to the infinitesimal deformation W d := W d , u d * O C d of W d over A (Proposition 2.9). In the following theorem we prove that the blowing-down morphism is an isomorphism of functors whenever C is non-hyperelliptic. We refer to §3.1 for its proof.
Theorem 1.1. If C is a smooth non-hyperelliptic curve of genus g ≥ 3, then for all 1 ≤ d < g − 1 the blowing-down morphism u ′ d : Def C d → Def W d is an isomorphism. In particular, Def W d ≃ Def C , W d is unobstructed, and Def W d is prorepresented by a formal power series in 3g − 3 variables.
We believe that the hypothesis of non-hyperellipticity is unnecessary at least in cases d = 2. However, for d = 2, we notice that the space W 2 , besides the deformations coming from the curve, may also acquire additional deformations coming from deformations of Fano surfaces of lines associated to smooth cubic threefolds. In fact, Collino in [Co] shows that the locus of hyperelliptic Jacobians of dimension five lies in the boundary of the locus of intermediate Jacobians of smooth cubic threefolds. Now we make some comments regarding the proof of Theorem 1.1. In case the exceptional locus of u d has codimension at least three in C d , then Theorem 1.1 easily follows from the general theory of blowing-down morphisms (see §2.3 and in particular Criterion 1.2 below) and does not rely on the special structure of Abel-Jacobi maps. In fact, the existence of blowing-down morphisms is not specific to the morphism u d itself, rather to any morphism f : X → Y between projective integral schemes such that Rf * O X ≃ O Y (Proposition 2.5). In §2.3, we give an explicit description to the differential of a blowing-down morphism f ′ : Def X → Def Y and, moreover, we find sufficient conditions on X, Y and f so that f ′ defines an isomorphism of functors. This leads to the following criterion whose proof can be found in Corollary 2.8. We refer to [Ran3] for related criteria regarding source-target-stability type problems.
Criterion 1.2. Let f : X → Y be a birational morphism of integral projective schemes over an algebraically closed field of characteristic zero such that the exceptional locus of f is of codimension at least three in X. Furthermore assume that Rf * O X ≃ O Y . If X is non-singular, unobstructed and h 0 (X, T X ) = 0, then the blowing-down morphism f ′ : Def X → Def Y is an isomorphism of functors of Artin rings.
Hence the difficult case of Theorem 1.1 is precisely when the exceptional locus of u d is of codimension two (the case of codimension one is excluded as we are supposing C non-hyperelliptic). For this case we carry out an ad-hoc argument specific to Abel-Jacobi maps. The main point is to prove that the differential du ′ d of the blowing-down morphism is an isomorphism even in this case. To this end, first of all we notice that the kernel and cokernel of du ′ d are identified to the groups Ext 1 , respectively, where P is the cone of the following composition of morphisms of complexes: where the first map is the truncation morphism and the second is the natural morphism between sheaves of Kähler differentials. On the other hand, an application of Grothendieck-Verdier's duality shows that the vanishings of the above mentioned Ext-groups hold as soon as the support of the higher direct image sheaf R 1 f d * (Ω C d /W d ⊗ ω C d ) has sufficiently high codimension in J, namely at least five (see Propositions 3.5 and 3.6). But this is ensured by Ein's computations of the Castelnuovo-Mumford regularity of the dual of the normal bundle to the fibers of f d ( [Ein]). Finally, the passage from first-order deformations to arbitrary infinitesimal deformations follows as C d is unobstructed.
As an application, we compute the infinitesimal deformations of the Albanese map where both the domain, the codomain, and the closed immersion are allowed to deform (Sernesi in [Se,Example 3.4.24 (iii)] solves the case d = 1). The importance of this problem relies on a conjecture of Debarre pointing to a classification of d-codimensional subvarieties X of ppavs (A, Θ) representing a minimal cohomology class, i.e.
over an Artinian local C-algebra A such that the restriction to the closed point is the closed embedding ι d : W d ֒→ J will tell us in which directions the Jacobian J is allowed to deform as a ppav containing a subvariety representing a minimal class. More precisely, this study will suggest us along which type of Jacobians there might be an irreducible component of C g,d (different from J g ) that passes through them. The main result of this paper in this direction is an evidence to Conjecture 1.3 mainly saying that non-hyperelliptic Jacobians, seen as elements in C g,d , deform along the expected directions. Less informally, if denotes the natural forgetful morphism, we have then: Theorem 1.4. If C is a smooth non-hyperelliptic curve of genus g ≥ 3, then for any 1 ≤ d < g − 1 the forgetful morphism p W d : Def ι d → Def W d is an isomorphism. In particular, ι d is unobstructed and Def ι d is prorepresented by a formal power series in 3g − 3 variables.
Combining with Theorem 1.1 we obtain: Corollary 1.5. Under the hypotheses of Theorem 1.4 there exists an isomorphism of functors of Artin rings Def ι d ≃ Def C for every 1 ≤ d < g − 1. Hence, if J is a non-hyperelliptic Jacobian, then any infinitesimal deformation of J together with an infinitesimal deformation of the minimal class W d deforms J along the Jacobian locus.
The proof of Theorem 1.4 still relies on the use of blowing-down morphisms. More precisely, we prove that not only deformations of schemes with rational singularities can be blown-down, but also deformations of morphisms between them (Proposition 2.9). Thus there is a well-defined morphism of functors F : Def f d −→ Def ι d which we prove to be an isomorphism, by means of Theorem 1.1 and Ran's formalism of deformations of morphisms recalled in details in §2. Finally, as by work of Kempf [Ke2] the forgetful morphism Def f d → Def C d is an isomorphism, we deduce that so is p W d .
Problem 1.6. Pareschi-Popa in [PP, Conjecture A] suggest that d-dimensional subvarieties X of g-dimensional ppavs (A, Θ) representing a minimal cohomology class should be characterized as those for which the twisted ideal sheaf I X (Θ) is GV (we recall that a sheaf F on an abelian variety . In fact, one of the main results of [PP] is that the GV condition on X implies that X has minimal class. It would be interesting to check whether this property is stable under infinitesimal deformations in order to get information concerning the geometry of the corresponding loci in A g for all d. Steps in this direction are again due to Pareschi-Popa as they prove, without appealing to deformation theory but using the technique of Fourier-Mukai transforms, that for d = 1 and d = g − 2 this locus coincide precisely with the Jacobian locus in A g ( [PP, Theorem C]).
Notation. In these notes a scheme is a separated scheme of finite type defined over an algebraically closed field k of characteristic zero, unless otherwise specified. We denote by Ω X the sheaf of Kähler differentials and by T X the tangent sheaf of a scheme X.
Support. This collaboration started when ST visited the University of Illinois at Chicago supported by the AWM-NSF Mentoring Grant (NSF award number DMS-0839954), while LL was a graduate student there. Both authors are grateful to UIC for the nice working environment and the kind hospitality. LL was supported by the SFB/TR45 "Periods, moduli spaces, and arithmetic of algebraic varieties" of the DFG (German Research Foundation).
Deformations of morphisms
We begin by recalling Ran's theory on deformations of morphisms between compact complex spaces extending works of Horikawa in the smooth case (cf. [Ran2,Ran3,Ran4]). The deformations considered by Ran allow to deform both the domain and the codomain of a morphism. For the purposes of this work we present Ran's theory for the category of schemes defined over an algebraically closed field k of characteristic zero. Let X be a reduced projective k-scheme. We define the spaces and denote by Def X the functor of Artin rings of deformations of X up to isomorphism. We recall that T 1 X is the tangent space to Def X . Under this identification a first-order deformation π : X → Spec k[t]/(t 2 ) of X is sent to the extension class determined by the conormal sequence of the closed immersion X ⊂ X : (the fact that π is flat implies that O X is the conormal bundle of X ⊂ X , while the fact that X is reduced implies that the conormal sequence is exact also on the left). Moreover, if X is a locally complete intersection, then T 2 X is an obstruction space (cf. [Se,Theorem 2.4.1 and Proposition 2.4.8]).
Let Y be another reduced projective k-scheme, and let f : X → Y be a morphism. A deformation of f : X → Y over an Artinian local k-algebra A with residue field k is a diagram of commutative squares and triangles such that X and Y are deformations of X and Y over A respectively andf restricts to f when pulling-back to Spec k. We say that a deformation f : are the isomorphisms determined by X and X ′ respectively, and similarly for Y, Y ′ , and φ. We denote by Def f the functor of Artin rings of deformations of f up to isomorphism. The functors Def X and Def f satisfy Schlessinger's conditions (H 0 ), (H 1 ), (H 2 ) and (H 3 ) when both X and Y are projective schemes (cf. [S]).
2.1. Tangent space. One of the central results in [Ran2] is that the first-order deformations of a morphism f : X → Y are controlled by a certain space T 1 f defined as follows. Let be the natural morphisms induced by f and let be the morphism induced by δ 0 via adjunction. Then we define T 1 f to be the abelian group consisting of isomorphism classes of triples (e X , e Y , γ) such that e X and e Y are classes in determined by the conormal sequences of some deformations X and Y of X and Y respectively: and γ : f * Ω Y|Y → Ω X |X is a morphism such that the following diagram commutes. For future reference we recall [Ran2,Proposition 3.1] revealing the role of T 1 f .
Proposition 2.1. Let X and Y be projective reduced k-schemes and let f : X → Y be a morphism.
Then T 1 f is the tangent space to Def f .
Proof. As already pointed out earlier, the datum of two extension classes e X ∈ T 1 X and e Y ∈ T 1 Y is equivalent to giving two Cartesian diagrams such that both g and h are flat morphisms. On the other hand, as it is shown in [BE,Theorem 1.6], the existence of a morphismf : X → Y such that the top square of (2) commutes is equivalent to the existence of a morphism γ : f * Ω Y|Y → Ω X |X such that the right square of (5) commutes. Finally, it is not hard to prove that the commutativity of the rightmost triangle in (2) is equivalent to the commutativity of the left square of (5).
2.2.
Ran's exact sequence. In order to study the space T 1 f usually one appeals to an exact sequence relating T 1 f to the tangent spaces T 1 X and T 1 Y . This sequence turns out to be extremely useful to study stability and co-stability properties of a morphism, and furthermore, in some situations, it suffices to determine the group T 1 f itself (cf. [Ran3]).
Let T 0 f be the group consisting of pairs of morphisms such that the following diagram where the morphisms without names are the obvious ones, while the others are defined as follows.
Given a pair of morphisms as in (6) Finally, in order to define λ 2 , we introduce some additional notation. Let be the truncation morphisms (see [Huy,Exercise 2.32]) and set We denote by λ 1 2 and λ 2 2 the components of λ 2 and we set λ 2 (e X , e Y ) = λ 1 2 (e X ) − λ 2 2 (e Y ) where e X and e Y are extension classes as in (4), so we only need to define λ 1 2 and λ 2 2 . Thinking of the extensions e X ∈ T 1 X and e Y ∈ T 1 Y as morphisms of complexes α : [Ha1,Corollary 5.11]. More concretely, we have Proposition 2.2. If f : X → Y is a morphisms of reduced k-schemes such that f (X) is not contained in the singular locus of Y , then the sequence (7) is exact.
Proof. We show exactness only at the term T 1 X ⊕ T 1 Y , exactness at the other terms follows easily from the definition of our objects and maps. First of all we show that the composition Let e X and e Y be two extension classes as in (4) which we think of as morphisms of complexes α : Ω X → O X [1] and β : Ω Y → O Y [1], and suppose that there exists a morphism γ : f * Ω Y|Y → O X |X such that (5) commutes. By our assumption on f (X), together with generic freeness and base change, we see that the sheaf L −1 f * Ω Y is torsion on X. This in particular yields the exactness of the sequence Therefore the commutativity of (5) implies the commutativity of the following diagram of distinguished triangles whereγ denotes the composition → Ω X |X . Hence in particular the commutativity of the right-most square tells us that •Lf * β which implies the commutativity of (11). By taking cohomology in degree 0 and by the fact that (10) is exact, we conclude that (5) is commutative too. Therefore the triple (e X , e Y , γ) lies in the image of λ 1 .
Remark 2.3. The functor Def f comes equipped with two forgetful morphisms p X : Def f → Def X and p Y : Def f → Def Y obtained by the fact that a deformation of f determines both a deformation of X and one of Y . The differential dp X is identified to the morphism T 1 f → T 1 X of the sequence (7), and similarly for the differential dp Y .
Remark 2.4. Obstruction spaces to the functor Def f are studied in [Ran2] in complete generality. However, we are able to follow Ran's argument only under the additional hypothesis that the involved spaces are locally complete intersections. As in this paper we will be dealing with varieties which are not locally complete intersections, we refrain to give a systematic description of the obstructions, but rather we refer to [Ran2] and [Ran4] to get a flavor of this theory.
2.3. Blowing-down deformations. We recall that a resolution of singularities X of a variety Y with at most rational singularities induces a morphism of functors of Artin rings Def X → Def Y (cf. [Wa] for the affine case, and [Hui, Proposition 2.1] and [Sa,Corollary 2.13] for the projective case). We extend this fact by relaxing the hypotheses on X.
Proposition 2.5. Let X and Y be projective integral k-schemes and let f : X → Y be a morphism such that Rf * O X ≃ O Y . Then f defines a morphism of functors where to a deformation X of X over a local Artinian k-algebra A with residue field k associates the Proof. Let A be a local Artinian k-algebra as in the statement and let be a deformation of X over A. Moreover denote byh : Y → Spec k the structure morphism of Y .
Thenḡ =h • f and Y admits a morphism h to Spec A determined by the following morphism of k-algebras We only need to prove that h is a flat morphism. To this end we can suppose that Y is affine.
This easily follows by taking cohomology from the following chain of isomorphisms (and from the fact that we are assuming Y affine) Now as g is a flat morphism we can apply the push-pull formula of [Ku,Lemma 2.22 and Corollary 2.23] to (13) to have and therefore by Nakayama's lemma we would get the contradiction R i 0 g * O X = 0. We conclude that i 0 = 0 and moreover that From this we see that for any i > 0 we have Moreover, since the functor of global sections is exact on affine spaces, we finally get R i f * O X = 0 for all i > 0.
In order to show that h is flat, we will prove that for any coherent sheaf F on Spec A we have L i h * F = 0 for all i < 0. But this follows from the projection formula ([Ha1, Proposition 5.6]) as the following chain of isomorphisms yields We conclude that L i h * F = 0 for all i < 0 since the derived push-forward of a sheaf lives in non-negative degrees.
Proposition 2.7. The differential df ′ to f ′ in (12) can be described as the composition where the first map is obtained by applying the functor Ext 1 O X (−, O X ) to the morphismδ 1 : Lf * Ω Y → Ω X , and the second by adjunction formula [Ha1,Corollary 5.11]. Moreover, if f is birational and the exceptional locus of f in X has codimension at least 3, then df ′ is an isomorphism.
Proof. We refer to [Wa] for the description of df ′ in the affine case. In the global case this is obtained as follows; we continue using notation of Proposition 2.5 and its proof. Let X be a first-order deformation of X and let Y = (Y, f * O X ) be the deformation determined by f ′ . As f * O X = O Y , we get a morphism f : X → Y such that the top square of the diagram (2) commutes. Therefore, as shown in the proof of Proposition 2.1, the right square of (5) commutes and hence, which determines the deformation X , is sent to the triangle Consequently, via adjunction, this last triangle is sent to the exact sequence 0 → O Y → Ω Y|Y → Ω Y → 0 which is the sequence determining the first-order deformation Y of Y . Therefore the morphism defined in (14) takes X to Y which is what we needed to show.
For the second statement we completeδ 1 to a distinguished triangle: We note that since f is an isomorphism outside the exceptional locus, the supports of the cohomology sheaves H i (Q) of Q have codimension ≥ 3. Moreover, H i (Q) = 0 for i ≥ 1 as Lf * Ω Y lives in non-positive degrees. Then Ext j O X (H i (Q), O X ) = 0 for all j ≤ 2 and all i, and therefore the spectral sequence E j,i The previous proposition leads to a criterion for the blowing-down morphism to be an isomorphism. This proves Criterion 1.2 of the Introduction.
Corollary 2.8. Let f : X → Y be a birational morphism of integral projective k-schemes such that the exceptional locus of f is of codimension at least three in X and Rf * O X ≃ O Y . If X is nonsingular, unobstructed ( e.g. h 2 (X, T X ) = 0), and h 0 (X, T X ) = 0, then the blowing-down morphism f ′ : Def X → Def Y is an isomorphism.
Proof. By [Se, Corollary 2.6.4 and Corollary 2.4.7] the functor Def X is prorepresentable and smooth. Moreover, as f is a small resolution, there is an isomorphism f * T X ≃ T Y whose proof can be found in [SV,Lemma 21]. Then H 0 (Y, T Y ) ≃ H 0 (X, T X ) = 0 and Def Y is prorepresentable as well. At this point the corollary is a consequence of Proposition 2.7 and the following criterion [Se, Remark 2.3.8]: if γ : G → G ′ is a morphism of functors of Artin rings such that the differential dγ is an isomorphism and both G and G ′ are prorepresentable with G smooth, then γ is an isomorphism and G ′ is smooth as well.
In a similar fashion, we also show that it is possible to blow-down deformations of morphisms.
Proposition 2.9. Let f : X → Y be a resolution of singularities of a projective integral normal k-scheme Y such that Rf * O X ≃ O Y . Moreover fix a smooth projective variety Z together with two morphisms f 1 : X → Z and f 2 : Y → Z such that f 1 = f 2 • f . Then f defines a morphism of functors F : Def f 1 → Def f 2 such that the following diagram (15) commutes where p and p ′ are forgetful morphisms and f ′ is the blowing-down morphism defined in Proposition 2.5.
Proof. By definition, a deformation f 1 : X → Z of f 1 over a local Artinian k-algebra A with residue field k determines a deformation X of X and a deformation s : Z → Spec A of Z. Furthermore, by Proposition 2.5, X defines a deformation Y = (Y, f * O X ) of Y . As f 1 determines a morphism of sheaves of k-algebras f At this point in order to check that f 2 is a deformation of f 2 we only need to show that the composition Y f 2 → Z s → Spec A is flat. This is equivalent to proving that for any coherent sheaf F on Spec A the higher cohomology of L(s • f 2 ) * F vanish. But since s • f 1 is flat, by projection formula we have that for any index i < 0 (we use the symbol H i to denote the i-th cohomology of a complex): The commutativity of (15) follows from the definitions of all involved morphisms and functors.
Deformations of W d (C)
Let C be a complex smooth curve of genus g ≥ 3. We denote by the Brill-Noether loci parameterizing degree d line bundles on C having at least one non-zero global section. We recall that it is possible to put a scheme structure on W d (C) by means of Fitting ideals so that W d (C) is an irreducible, normal, Cohen-Macaulay scheme of dimension d ( [ACGH,Corollary 4.5
]). A resolution of singularities of W d (C) is provided by an Abel-Jacobi map
where C d is the d-fold symmetric product of C. Note that a fiber of u d over a point [L] ∈ W d (C) is nothing else than the linear series |L| associated to L. Finally, by fundamental results of Kempf ([Ke1]), we have that W d (C) has at most rational singularities so that the following isomorphisms hold: By Proposition 2.5 there is then a well-defined blowing-down morphism The goal of this section is to prove the following: Theorem 3.1. If C is a smooth non-hyperelliptic curve of genus g ≥ 3, then the blowing-down morphism u ′ d : The proof of the previous theorem requires a few technical results on the supports of higherdirect image sheaves of type We will collect these facts in the following subsection and we will show the proof of Theorem 3.1 in §3.2.
It is worth noticing that Fantechi ([Fa]), by extending previous work of Kempf ([Ke2]), proved the following: Theorem 3.2 (Fantechi). Let C be a smooth curve of genus g ≥ 2 and let d ≥ 2 be an integer. Then the quotient morphism C d → C d induces an isomorphism of functors of Artin rings Def C d ≃ Def C if and only if g ≥ 3.
Combining with Theorem 3.1 we obtain Corollary 3.3. If C is a smooth non-hyperelliptic curve, then for all 1 ≤ d < g − 1 there are isomorphisms of functors Def C ≃ Def W d (C) .
It follows from the previous corollary that Def W d (C) is unobstructed and is prorepresented by a formal power series in 3g − 3 variables as so is Def C ([Se, Proposition 2.4.8 and Corollary 2.6.6]).
3.1. Supports of special higher direct image sheaves. We denote by the Brill-Noether loci parameterizing degree d line bundles on C having at least i + 1 linearly independent global sections, and by the loci parameterizing degree d effective divisors on C whose associated linear series is of dimension at least i. Note that u −1 d (W i d (C)) = C i d . We start by remarking a general fact concerning the supports of higher direct image sheaves under Abel-Jacobi maps.
Proof. There are fiber product diagrams where ν and µ are open immersions andū d is the restriction of [Ha2,Corollary 11.2]. Therefore by base change we find µ * R j u d * F = 0.
We now give more precise information regarding the supports of two specific higher direct image sheaves: is the sheaf of relative Kähler differentials. The main tool we use towards this study is Ein's cohomological computations of the dual of the normal bundle to the fibers of u d ( [Ein]).
Proof. By Proposition 3.4 we have that supp , so we only need to show that the stalks . From now on we fix an element [L] ∈ W 1 d (C)\W 2 d (C). Recall that, as shown in [Ein,Theorem 1.1], the normal bundle N of the fiber P L := u −1 d ([L]) ≃ P 1 at [L] sits in an exact sequence of the form From this we easily deduce that det N ∨ ≃ O P 1 (g − d + 1), and moreover that We apply the theorem on formal functions to get the vanishing of the above mentioned stalks. Denote by I the ideal sheaf defining E = E 1 := P L in C d and let E n be the subscheme defined by I n . We have exact sequences By the theorem on formal functions we obtain isomorphisms so that it is enough to check the vanishing of cohomology groups on the RHS. By tensoring (18) by F, and by using the isomorphisms I n /I n+1 ≃ Sym n N ∨ , we deduce new exact sequences where we denote by K n the kernel of ψ. We are interested in the vanishing of We proceed by induction on n. The base step n = 0 is easily proved as Now we show that if H 1 (E n , F n ) = 0, then also H 1 (E n+1 , F n+1 ) = 0. First of all we note that all we need is the vanishing of In fact, by denoting by Q n the cokernel of ψ, the inductive hypothesis tells us that H 1 (E n+1 , F n+1 ) vanishes as soon as H 1 (E n+1 , Q n ) does. But this is the case if H 1 (E, Sym n N ∨ ⊗ F |E ) = 0 as H 2 (E n+1 , K n ) = 0 (recall that dim E n+1 = 1).
Finally, in order to get the vanishing of the RHS of (19), we note that dualizing the sequence (17) we get surjections Sym n (H 1 (C, O C ) ∨ ⊗ O P 1 ) ։ Sym n N ∨ for all n ≥ 1. Therefore there are surjections from which one easily deduces the vanishing of (19) as the LHS of (20) is zero.
Proof. We follow the strategy of Proposition 3.5 so that we only need to check the vanishings of for all n ≥ 1. We recall the isomorphism Ω C d /W d (C) |E ≃ ω E ≃ O P 1 (−2) (cf. e.g. [Ha2,Proposition 8.10]) and that ω C d |E ≃ O P 1 (g − d − 1). Therefore For the second set of groups in (21) we note the isomorphisms and the surjections deduced from (17). As the groups on the LHS of the previous surjections vanish, so the groups in (22) do.
Remark 3.7. The previous two propositions can be extended to all higher direct images to yield inclusions for all j ≥ 1 (the latter holds for d ≤ g − 2). While the proof of the first set of inclusions do not require any additional tools, for the latter we need to involve Bott's formula to check that 3.2. Proof of Theorem 3.1. To prove Theorem 3.1 we use the criterion [Se,Remark 2.3.8] which we have already recalled in Corollary 2.8.
In our setting, the functor Def C d is prorepresentable and unobstructed by Theorem 3.2. Therefore we only need to prove that Def W d (C) is prorepresentable and that the differential to u ′ d is an isomorphism. A sufficient condition for the prorepresentability of Def W d (C) is the vanishing of H 0 (W d (C), T W d (C) ) ( [Se,Corollary 2.6.4]). On the other hand, as u d is a small resolution (as we are supposing that C is non-hyperelliptic), we obtain an isomorphism u d * T C d ≃ T W d (C) ( [SV,Lemma 21]) which immediately yields by the Künneth decomposition (here σ d denotes the d-symmetric group).
We now prove that the differential du ′ d is an isomorphism. This is slightly more difficult and it will take the rest of the subsection. To begin with, we complete the morphisms ζ : (8) and (9) to distinguished triangles: Moreover, sinceδ 1 = δ 1 • ζ where δ 1 : u * d Ω W d (C) → Ω C d is the natural morphism, these triangles fit in the following commutative diagram: where N is the cone of δ 1 . Therefore, from the description of du ′ d as in Proposition 2.7, in order to prove that du ′ d is an isomorphism, it is enough to prove that Ext 1 The key ingredient to prove these vanishings is the Grothendieck-Verdier duality which reduces calculations from C d to the Jacobian J(C) of C. We fix a point Q ∈ C and denote by ι d the closed immersion Moreover we define the composition Then applications of Grothendieck-Verdier duality ([Ha1, ). At this point the proof that du ′ d is an isomorphism follows from the following Propositions 3.8 and 3.9.
Proposition 3.8. If C is a smooth non-hyperelliptic curve of genus g, then for j = 2, 3 we have Proposition 3.9. If C is a smooth non-hyperelliptic curve of genus g, then for j = 1, 2 we have Proof of Proposition 3.8. Our assertion is equivalent to Ext j+g−d O J (C) (Rf d * (M ⊗ ω C d ), O J(C) ) = 0 for j = 1, 2. First of all we note that To see this we first tensorize the top distinguished triangle of (23) by ω C d , and then we apply the functor Ru d * . Hence projection formula ([Ha1, Proposition 5.6]), together with the isomorphism (16), yields an exact sequence Finally the statement in (25) follows as ι d is a closed immersion.
We now point out that, for all j ≥ 1, there are isomorphisms (23). These isomorphisms, together with Propositions 3.5 and 3.4, yield Consider now the spectral sequence ( [Huy,p. 58 We are interested in the vanishing of the terms on the lines p + q = 1 + g − d and p + q = 2 + g − d, and therefore in the vanishing of the terms E 1+g−d−q,q 2 and E 2+g−d−q,q 2 for q ≤ 1. However this easily follows from the general fact that Ext k O J (C) (F, O J(C) ) = 0 for any coherent sheaf F on J(C) such that codim supp F > k, and from Martens' Theorem saying that dim W j d (C) ≤ d − 2j − 1 if C is non-hyperelliptic ([ACGH, Theorem 5.1]). In fact, in this way, we obtain E p,q 2 = 0 for couples (p, q) such that either p ≤ g − d − 2q and q ≤ −2, or p ≤ 4 + g − d and q = −1, or p ≤ 3 + g − d and q = −1, 0.
Proof of Proposition 3.9. We consider the spectral sequence by Propositions 3.4 and 3.6 we find At this point to compute the groups Ext j+g−d (Rf d * (N ⊗ ω C d ), O J(C) ) for j = 1, 2 we use the spectral sequence and we argue as in Proposition 3.8.
Simultaneous deformations of W d (C) and J(C)
In this section we aim to prove Theorem 1.4. We start by proving some general facts regarding the closed immersion ι d : W d (C) ֒→ J(C).
Proposition 4.1. Let C be a smooth curve of genus g ≥ 2 and let 1 ≤ d ≤ g be an integer. Then Proof. The proof of (i) can be found in [Mac]. Nonetheless we present here a proof for the reader's ease. Denote by π d : C d → C d the quotient morphism realizing the symmetric product C d as quotient of the d-fold product C d under the action of the symmetric group σ d . Therefore Künneth's decomposition yields Now we turn to the proof of (ii). By recalling the definition of f d = ι d • u d : C d → J(C) in (24), we get isomorphisms f * d H j (J(C), O J(C) ) ≃ ∧ j f * d H 1 (J(C), O J(C) ) ≃ ∧ j H 1 (C, O C ) ≃ H j (C d , O C d ) thanks to the universal property of J(C) and by (i). Moreover, since Ru d * O C d ≃ O W d (C) , we obtain isomorphisms ). These immediately yield (ii) once one looks at the long exact sequence in cohomology induced by the short exact sequence Finally, for the last point it is enough to apply RHom O J (C) (Ω J(C) , −) to the sequence (27) and to use (ii). This yields the claimed isomorphisms for j ≤ d − 1 together with an injection Ext d O J (C) (Ω J(C) , O J(C) ) ֒→ Ext d O J (C) (Ω J(C) , ι d * O W d (C) ) which is an isomorphism for dimensional reasons.
The deformations of Abel-Jacobi maps f d : C d → J(C) have been studied by Kempf. In particular,in [Ke2], Kempf shows that these deformations are all induced by those of C d in case C is non-hyperelliptic. However, in view of Theorem 3.2, Kempf's result extends to all smooth curves of genus g ≥ 3. In the following proposition we present a slightly different proof of this fact in the case d ≤ g by means of the theory of deformations of holomorphic maps developed by Namba in [Na]. Moreover, we include a statement regarding the deformations of the closed immersion ι d : W d (C) ֒→ J(C). In combination with Theorem 1.1, this in particular proves Theorem 1.4 of the Introduction. an isomorphism ( [Se,Remark 2.3.8]). Finally, as both p and u d are isomorphisms, this yields that p ′ is an isomorphism as well.
Deformations of Fano surfaces of lines
In this section we prove some deformation-theoretic statements regarding the Fano surface of lines. Let Y ⊂ P 4 be a smooth cubic hypersurface and let F (Y ) be the Fano scheme parameterizing lines on Y . Then F (Y ) is a smooth surface which embeds in the intermediate Jacobian J(Y ). We denote by ι : F (Y ) ֒→ J(Y ) this embedding and we recall that the tangent space to Def F (Y ) has dimension 10 while the obstruction space dimension 40.
Proposition 5.1. The surface F (Y ) is co-stable in J(Y ), i.e. the forgetful morphism p F (Y ) : Def ι → Def F (Y ) is smooth. Moreover, dp F (Y ) is an isomorphism and Def ι is less obstructed than Def F (Y ) .
Proof. By [CG,Theorem 11.19] (cf. [LT,Theorem 4.1] for a different proof) there is an isomorphism J(Y ) ≃ Alb(F (Y )) between the intermediate Jacobian J(Y ) and the Albanese variety of F (Y ) from which we get an isomorphism H 1 F (Y ), O F (Y ) ) ≃ H 1 (J(Y ), O J(Y ) . Moreover, by [CG,(12.1)] we deduce a further isomorphism But the vanishing of these groups for i = 2 is a sufficient condition for co-stability ( [Se,Proposition 3.4.23]). On the other hand, the vanishing of all of them allow us to reason as in Proposition 4.2 to obtain the other statements.
|
2014-10-29T14:16:36.000Z
|
2014-10-29T00:00:00.000
|
{
"year": 2014,
"sha1": "68e35f51001240a6fdc7d517a9754723db773b3e",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/1410.7986",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "5b01ecad590686a7373e4ea84776c8e1ab34fa60",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
238024657
|
pes2o/s2orc
|
v3-fos-license
|
Comparison and Validation of Hydrodynamic Theories for Wave Energy Converter Modelling
: Dynamic Wave Energy Converter (WEC) models utilize a wide variety of fundamental hydrodynamic theories. When incorporating novel hydrodynamic theories into numerical models, there are distinct impacts on WEC rigid body motions, cable dynamics, and final power production. This paper focuses on developing an understanding of the influence several refined hydrodynamic theories have on WEC dynamics, including weakly nonlinear Froude-Krylov and hydrostatic forces, body-to-body interactions, and dynamic cable modelling. All theories have evolved from simpler approaches and are of importance to a wide array of WEC archetypes. This study quantifies the impact these theories have on modelling accuracy through a WEC case study. Theoretical differences are first explored in a regular sea state. Subsequently, numerical validation efforts are performed against field data following wave reconstruction techniques. Comparisons of significance are WEC motion and cable tension. It is shown that weakly nonlinear Froude-Krylov and hydrostatic force calculations and dynamic cable modelling both significantly improve simulated WEC dynamics. However, body-to-body interactions are not found to impact simulated WEC dynamics.
Introduction
Numerical modelling tools expedite the development of the wave energy sector by aiding in the design, analysis, and optimization of wave energy converter (WEC) devices. These tools aid in capturing the complex interactions between rigid bodies dynamics, hydrodynamics, hydrostatics, power take-off (PTO) units, and control systems present in WECs. With design decisions based heavily on predictions from simulation tools, it is of vital importance to verify and validate WEC numerical software [1]. Users have the choice to include numerous refined numerical theories within WEC software that may increase the accuracy of simulated dynamics.
Numerical theories have evolved with the intention of improving the numerical representation of the inherently complex dynamics of floating offshore systems. For instance, floating body hydrostatic and Froude-Krylov force calculations traditionally concern a static submerged portion of the body [2]. However, many floating bodies have dynamic wetted surface areas. Weakly nonlinear Froude-Krylov and hydrostatic force calculations pertain to incorporating this changing wetted surface area into WEC dynamics [3]. Additionally, a rigid body usually has 6 degrees of freedom corresponding to the particular body's translational and rotational motion [2]. However, rigid bodies in close proximity to one another impact forces on each other, and body-to-body interactions concern integrating these cross-body forces into a numerical model [4]. Additionally, when marine cables attach to a rigid body, forces cause structural deformation of the cable, and marine cable models, which have evolved from simple mass-spring-dampers to dynamic cable models, simulate such structural responses [5]. The three aforementioned modelling techniques are all theoretical refinements that may pertain to modelling a particular WEC's dynamics more accurately.
Several studies have focused on the implementation of an refined hydrodynamic theories. Combourieu et al. determined that including body-to-body interactions is significant for a floating oscillating flap device [1]. Wendt et al. focused on the importance of nonlinear Froude-Krylov and hydrostatic force calculations for a semi-submerged sphere and a heaving float, and determined that nonlinear effects are most significant for large motions in large waves [3]. Paduano et al. compared and validated different cable models for moored floating WECs, and found close agreement between a quasi-static and dynamic cable models [6]. All of these comparisons aid a particular aspect of the WEC design process. However, numerous theories are often pertinent for a singular WEC, and the present lack of a comprehensive review of the associated theoretical impacts creates uncertainty and an inability to cross validate prior research efforts.
The objectives of this study are two-fold: (1) determine how weakly nonlinear Froude-Krylov and hydrostatic force calculations, body-to-body interactions, and/or dynamic cable models impact a WEC's simulated dynamics, and (2) to validate numerical models against field measurement data. Section 2 lays out various WEC numerical modelling theories. Section 3 discusses the software and numerical model implementation of this study. Section 4 presents the particular WEC this study focuses upon. Section 5 shows the results and discussion for the code-to-code and code-to-field comparisons. Section 6 discusses this study's conclusions and the opportunities for future work.
Wave Energy Converter Numerical Modelling Theory
WEC modelling varies in terms of fidelity of the simulation environment and the WEC model's inherent assumptions. Many WEC models aim to approximate Newton's 2nd law of motion. Figure 1 provides A broad layout of different WEC modelling approaches. The lowest fidelity WEC models employ analytical approximations of these forces, as in Zhao et al., whom investigate the hydrodynamics of multi-pontoon floating platforms that work as an oscillating water column WECs [7]. However, these low-fidelity approaches rely on many assumptions that simplify WEC dynamics greatly. A slightly more robust approach performs calculations solely in the frequency domain to predict WEC behavior, as in Pastor and Liu, whom model a heaving point absorber WEC in the frequency domain [8]. Time-domain multibody dynamic modelling is a more robust approach, which this study incorporates. This approach employs coefficients found in the frequency domain following linear potential flow theory, and specifies additional forces in the time-domain through a multibody dynamic solver [1,9,10]. The highest fidelity numerical models provide the most robust and computationally expensive approach to approximating Newton's 2nd law of motion numerically, and concern computational fluid dynamics (CFD) and smoothed particle hydrodynamics (SPH). CFD and SPH differ in that CFD treats a fluid in an Eulerian reference frame, whereas SPH is inherently Lagrangian [11,12]. CFD and SPH are the most capable of approximating the Navier-Stokes equations for WEC dynamics. Active practice in the wave energy sector employs numerical modelling with varying levels of fidelity. However, the wave energy sector is highly reliant on time-domain multibody dynamic models, and so this study restricts itself to that approach.
Baseline WEC Modelling Theory
Time-domain representations of WEC dynamics enable the inclusion of nonlinear and higher order numerical schemes. The baseline time-domain representation of the WEC and its associated forces are shown in Equation (1): ..
In Equation (1) the left-hand side concerns the inertial force, which includes the dry mass of the body, m, the added mass at the infinite frequency, m a∞ , and the rigid body acceleration, ..
x. The right-hand side of Equation (1) includes all remaining pertinent forces present in the WEC system: F exc is the excitation wave force, F HP is the hydrostatic restoring force, F Rad is the radiation force (composed of radiation damping and frequency-dependent added mass), F pto is the PTO force, F D is the drag force, and F c is the overall cable force [1].
Excitation Force
The excitation force, F exc , is composed of the Froude-Krylov force, F FK , and diffraction force, F diff , and results from the unsteady pressure field due to waves. Particularly, F FK concerns the incident wave force, and is represented by: where S w is the wetted surface area, ρ is the fluid density, g is gravitational acceleration, k is the wave number, z is the vertical position, and η is the free surface elevation [2]. The diffraction force, F diff , results from the alteration of a wave field due to the presence of a rigid body, as shown in Equation (4): where φ diff is the diffraction wave potential, and n is the body's normal vector. The upper boundary for S w in Equations (2)-(4) is the free-surface [2].
Hydrostatic Force
The hydrostatic force, F HP , concerns the pressure loading of a liquid on a submerged rigid body. F HP , is represented by: The upper boundary for S w in Equations (2)-(4) is the free-surface [2].
Radiation Force
The radiation force, F Rad , represents the rigid body's response to a wave impulse. F Rad , is represented by: In Equation (5) K r is the radiation impulse response function determined from frequency-dependent added mass, m a , and damping coefficients, C rad . Rigid body velocities are specified by .
Power Take-Off (PTO) Force
The PTO influences a WEC's motions and the resulting power output, thereby impacting WEC performance [9]. PTO systems are designed in the time-domain, and are inherently unique for a particular WEC. The WEC's power take-off force, F pto , generally concerns a hydraulic PTO system [14], a pneumatic air turbine transfer PTO system [10,14], or a mechanical PTO system [15]. F pto calculations are dependent on the type of PTO system and the particular WEC characteristics; therefore, there is no uniform F pto equation [15].
Viscous Drag Force
Viscous drag is the resistive force acting from the fluid on a rigid body, and the drag force, F D , formulates from the fluid's tendency to withstand deformation. Equation (6) characterizes F D : x .
In Equation (6) C D is the structure's specific quadratic drag coefficient, ρ is the fluid density, A D is the characteristic area, and .
x is the rigid body's velocity [16].
Cable Force
The cable force, F c , pertains to the force marine cables induce to secure floating systems to a fixed location or to connect rigid bodies. Cable forces in their most simplistic form may be calculated as mass-spring-dampers: x.
In Equation (7), K c and C c are cable stiffness and damping coefficient matrices, and x and .
x are the relative displacement and velocity of the body, respectively [17].
Refined Numerical Theories
WEC modelling approaches are constantly evolving, and novel hydrodynamic theories reflect advances in understanding of WEC dynamic behavior. Each novel numerical theory must be continuously verified and validated. In this work, the theories of focus are weakly nonlinear Froude-Krylov and hydrostatic calculations, body-to-body (B2B) interactions, and dynamic cable models. These theories are chosen because they are pertinent to many WEC archetypes, and there is a present lack in literature on the verification and validation for systems that concern multiple of the aforementioned theories.
Weakly Nonlinear Froude-Krylov and Hydrostatic Forces
The standard practice for time-domain multibody dynamic modelling Froude-Krylov and hydrostatic forces are approximated from linear potential flow theory. These forces are calculated from the equilibrium position of the body using frequency-domain terms, which are found through a boundary element method BEM solver such as WAMIT [18] or NEMOH [19]. For time-domain implementation, F FK coefficients are only dependent on wave frequency, and F HP coefficients are constant.
Weakly nonlinear models extend linear potential flow theory by incorporating weakly nonlinear terms, which are induced by instantaneous water surface elevation and the exact position of the rigid body's geometry rather than mean surface elevation as in linear models [3,20,21]. Weakly nonlinear terms are only of importance for rigid bodies with a dynamic wetted surface area, and are dependent upon the portion of a geometric mesh above the still water line, as shown in Figure 2b [16,17]. This differs from linear models, where the submerged mesh is static, as displayed in Figure 2a. For WEC cases concerning rigid body(s) with significant dynamic wetted surface areas, the inclusion of weakly nonlinear calculations may considerably improve a WEC's simulated power capture assessment [3,22,23]. For calculation of instantaneous water surface elevation and associated pressure fields, linear wave theory is extended for a given computational mesh at the surface. Initial calculations of flow velocity and pressure show unrealistically large values for submerged mesh panels above still water level. Time-varying instantaneous surface elevation may be found by employing Wheeler wave stretching, where the instantaneous vertical position, z * , is represented by [24]: In Equation (8) h is the mean water depth, and η is the instantaneous water surface elevation [20,21,24,25].
Body-to-Body Interactions
When oscillating rigid bodies are in close proximity, hydrodynamic interactions can be modelled with additional components. In multibody systems, body-to-body interactions allow for the motion of one body to impart a force on nearby rigid bodies [18].
B2B interactions may be influential in the correct characterization of WEC performance. B2B hydrodynamics may reduce WEC float's motion, thereby, decreasing energy production [26]. If cross-coupled coefficients are important, then a rigid body's radiation force, F Rad , is extended. For example, in a two-body heaving system where cross-coupled terms are important, the total radiation force of rigid body 1, F Rad 1 , is [4]: In Equation (9) F Rad 11 is the radiation force on rigid body 1 due to the motion of rigid body 1, and F Rad 12 is the radiation force on rigid body 1 due to the motion of rigid body 2.
B2B interactions are inevitable in multibody systems; however, the importance of the inclusion of B2B interactions is not well-established. Depending upon the relative magnitude of a rigid body's self-vs. cross-coupled frequency coefficients, it may be acceptable to neglect B2B interactions for a multibody WEC model, for in such cases the added computational expense has an insignificant impact on WEC dynamics [4,27]. For example, following Starrett et al., B2B interactions for an array of 5 single-body WECs situated approximately 50 m apart were found insignificant [4]. The importance of B2B interactions corresponds to the distance between rigid bodies [27]. In order to show how B2B interactions impact the system's dynamics more clearly, it is important to verify and validate this extension to the calculated hydrodynamics of the system.
Dynamic Cables
When a cable attaches to a rigid body, reaction loads corresponding to the cable's elastic deformation apply as external forces on the rigid body [28]. Marine cable dynamic modelling is governed by nonlinear partial differential equations in terms of a space and time [5]. A solution to these equations may be obtained by linearizing the equations' dynamics about an equilibrium condition or employing numerical models in order to gain an approximation of the equations [29]. Lumped mass cable models define a cable by forces due to the weight, buoyancy, drag, and added mass at specified node positions [30][31][32][33][34]. These models account for cable elasticity through a series of point masses in a series of linear visco-elastic elements [34]. Cable loads may be expressed by [29]: In Equation (10), F c is the force vector of the specified cable, m c is the mass of the cable element, K ε is the axial stiffness of the cable, K κ is the cable bending stiffness, K τ is the torsional stiffness, W is the combined cable element weight and buoyancy, and H is the hydrodynamic load in terms of the global reference frame. The cable has n nodes, and a cable element is connected by two nodes. The term r (i)T refers to the position of the ith node, and r (i)T refers to that same node's curvature, where i = 0, · · · , n. Dynamic cable models presently vary as to the degree to which nonlinearity or high-order polynomials are included in the cable discretization scheme [35]. Linear lumped mass models neglect curvature and corresponding nonlinearities such as bending stiffness and torsion [29]. The MooDy model applies high-order polynomials for a cable's spatial discretization, which makes the numerical model well-suited for snap-loaded design [36]. Furthermore, it has been shown that including nonlinear terms such as bending stiffness when modelling cable dynamics helps prevent model instability [35,37]. To facilitate this, Buckham et al. present a method to incorporate curvature variables present in the dynamics of a continuous cable with cubic cable elements using finite element methods [38]. With some WEC designs, such as a PTO tethered system, the necessity of dynamic and robust finite element cable models for acceptable representation of the system's dynamics is unclear and, hence, an objective of this study.
Numerical Theory Implementations
Numerical models of WECs are an integral part of designing these floating energy absorption systems. The computational tools investigated involve the geometric build of a WEC to time-domain representation of the WEC's equation of motion. Two common WEC simulation tools are ProteusDS [39] and WEC-Sim [40]. With these codes, this study compares the implementation of the fundamental hydrodynamic theories, and validates simulation accuracy against experimental testing data.
The numerical modelling process follows that presented by Bailey et al. [9], which implements a wave-to-wire methodology and shows the necessary pre-processing to model a WEC in the time-domain. As shown in Figure 3, the numerical process begins with a geometric build formed with a CAD model in SOLIDWORKS [41] or similar packages. Then, the WEC geometry is discretized in Rhinoceros3D [42] or similar meshing software. The mesh is then passed through a frequency-domain Boundary Element Method (BEM) software, such as WAMIT [18], ShipMo3D [43], or NEMOH [44]. The BEM software resolves the velocity potential and fluid pressure for the entire submerged mesh, and outputs frequency dependent F exc , F HP and F Rad coefficients. Time-domain multibody dynamic software such as ProteusDS or WEC-Sim implements these frequency-dependent force coefficients as all other forces listed in Equation (1) into the time-domain.
ProteusDS
ProteusDS is a commercial time-domain analysis software package that tests virtual prototypes of many marine applications. The software focuses on determining the dynamic response of structures subject to currents, waves, and/or wind. Rigid body connections in ProteusDS are represented through the development of an articulated body algorithm (ABA) connection tool, which decreases the complexity of connections by reducing the degrees of freedom of the downstream rigid body to the number of degrees of freedom of an upstream joint [1,39]. Furthermore, ProteusDS has the capability to link with MATLAB/Simulink through an Application Programming Interface, allowing for control system customization [45]. In the time-domain, F exc , F HP and F Rad are calculated from BEM software. With weakly nonlinear terms, F exc and F HP coefficients depend on the instantaneous surface elevation, and utilize the Wheeler wave stretching method [24]. F Rad is only computed for a rigid body's 6 degrees of freedom in ProteusDS, for ProteusDS does not account for B2B interactions. F pto is calculated with specified joint damping and stiffness coefficients with the ABA connection. F D is calculated with Equation (6) about the center of pressure acting on the submerged mesh. F c is calculated in ProteusDS utilizing approach of Buckham et al., with material properties specified for the cable following Equation (10) [38,39].
WEC-Sim
The Wave Energy Converter SIMulator (WEC-Sim) numerical tool is an open-source software developed by Sandia National Laboratory and the National Renewable Energy Laboratory with the U.S. Department of Energy. WEC-Sim is developed in the MAT-LAB/SIMULINK environment using the multibody dynamics solver SimMechanics, allowing for control system and model customization. WEC-Sim models are comprised of rigid bodies, motion constraints, PTO units, and marine cable systems. Simulations are time-domain approximations to the WEC's governing equations of motion [1].
As in ProteusDS, F exc , F HP , and F Rad are computed in BEM software unless simulations include weakly nonlinear forces, for which scenarios F exc and F HP rely upon the instantaneous surface elevation, and employ the Wheeler wave stretching method [24]. F Rad calculation in WEC-Sim are dependent upon whether or not simulations enable B2B interactions [40]. F pto is calculated with specified joint damping and stiffness coefficients with a WEC-Sim SIMULINK joint [40]. F D is calculated with Equation (6) about each rigid body's specified reference areas [40]. This study performed simulations with WEC-Sim v4.2, which is coupled with MoorDyn C, a dynamic cable model [46]. However, MoorDyn C only permits a cable to connect to a rigid body to the seafloor, and so multibody cable connections are prohibited [40,46]. To circumnavigate this matter, a pseudo-cable representation as a massless rod is implemented, where F c is modelled as a simple spring for, as in Equation (7) [40].
Case Study WEC and Field Campaign
This study aims to validate the refined numerical theories against field deployment data (in addition to measuring the relative impact these theories have on the WEC's simulated environment). This study takes measurements from the Wave Energy Buoy that Self-deploys (WEBS) device for model validation.
Wave Energy Buoy That Self-Deploys (WEBS)
WEBS is a multibody point absorber developed by researchers from the University of Washington's Applied Physics Laboratory and C-Power Technologies. WEBS consists of a three-body surface float connected by a tether to a submerged heave plate as shown in Figure 4. The aft float is rigidly connected to the nacelle, which houses a gear box and rotary generator. The fore float rotates about the nacelle by means of a thru-axle on the nacelle to drive the generator. The tether is 3 4 " Samson Tenex, which has a breaking strength of 90 kN. The heave plate is a hexagonal cone located 60 m below the surface. This multibody device is designed for a sea state concerning a significant wave height of 1.8 m and an energy period of 7.9 s [47]. The relevant performance metrics of WEBS are the relative float angular velocity, tether tension, and heave plate position. The float angular velocity (with respect to the nacelle) is derived from an encoder's position, and, when paired with knowledge of the PTO, this metric allows for the calculation of power generation. Tether tension is a key indicator of loads that the WEC experiences, and was directly measured by load cells located along the tether. The heave plate position is derived from pressure sensors on the WEC heave plate, and the heave plate's operational mobility is a critical measure of design performance.
Heave plate heave displacement measurements include surface elevation changes, however, numerical values for heave plate depth are measured from the mean sea level and do not take instantaneous surface elevation into account. This is corrected by superimposing the surface elevation time-series on the numerical heave plate depth values. Drag coefficients are specified from experimental data found in Rusch et al. and listed in Table 1 [47]. The PTO system for this study is a mechanical PTO, where energy is generated as a torque through a rotary generator between oscillating bodies: In Equation (11), K mech is the mechanical spring stiffness, relative pitch orientation is denoted (θ 2 − θ 1 ), the PTO damping coefficient is given by C gen , and relative pitch angular velocity is denoted by . θ 2 − . θ 1 . The PTO system concerns a rotational joint with a stiffness coefficient value of 668 N/m and damping coefficient of 2400 Nms/rad [47].
Field Campaign Details
WEBS was deployed 9 nautical miles west of Moss Landing in Monterey Bay, CA, USA on 17 and 18 August 2016. Surface Wave Instrument Float with Tracking (SWIFT) buoys deployed alongside WEBS provide wind and wave measurements throughout deployment [48]. Data was collected over an 8-h period. During deployment, SWIFT buoy measurements found an average wave height of 1.33 m and period of 9.66 s [47].
Wave Reconstruction Methods
Wave spectra models and statistical techniques have been the norm for representing environmental conditions in numerical WEC models and provide statistical complementarity. Wave reconstruction is a novel concept determined to find predictions of instantaneous wave energy, and allows for direct time-domain comparison. There are two standard approaches to wave reconstruction: (1) the fixed-time approach, and (2) the fixed-point approach. The fixed-time approach takes wave elevation measurements via snapshots over a fixed area at a single instant, viewing the wavefield with a Eulerian reference frame [49]. The fixed-point approach creates a time series of fixed points to find wave elevation measurements, which is an inherently Lagrangian approach [49]. This study follows the fixed-point approach, where heave data is collected at 25 Hz in 448 s bursts every 9 min onboard SWIFT buoys [47,50]. This novel method employs deterministic wave prediction in WEC models such that developers may have near-optimal wave-by-wave control of a WEC in irregular waves [49].
Wave reconstruction is found to improve instantaneous heave predictions of experimental data by 63% relative to wave spectra forecasts [50]. ProteusDS allows for wave elevation reconstruction through custom wave segments, for which a user specifies a specific regular wave height, wave number, wave period, wave phase, and wave heading [39]. The number of wave segments is dependent on the number of frequencies specified in the SWIFT dataset. These regular wave segments are then superimposed in the time domain in order to generate the reconstructed wave field. This study applies 448 s of simulated waves, corresponding to one burst of SWIFT buoy data. Wave direction is assumed constant in this study. Spectral energy density from the SWIFT buoy data determines segment wave height.
WEC-Sim is capable to import a wave time series or a frequency dependent wave spectrum for wave reconstruction [39]. For comparative consistency within numerical models, ProteusDS and WEC-Sim use the same wave reconstructed time series. WEC-Sim applies the reconstructed wave elevation time series developed in ProteusDS to ensure compatible wave fields.
Results and Discussion
The objectives of this study are two-fold: (1) to show how the incorporation of weakly nonlinear Froude-Krylov and hydrostatic calculations, B2B interactions, and/or dynamic cable computations impact a WEC simulated dynamics, and (2) to validate the performance of these numerical simulations against field measurement data. As such, this study performs a code-to-code comparison and a code-to-field campaign comparison.
The code-to-code comparison helps clarify the relative impact of incorporating the refined numerical theories discussed in Section 2.2. Regular wave tests are performed over a range of wave frequencies and a 1 m wave height. Hydrodynamic theory implementation within different scenarios are listed in Table 2. The code-to-field campaign comparison uses wave elevation data from SWIFT buoys for numerically simulating WEBS and validating model outputs for the various scenarios listed in Table 2. Table 2. Refined numerical theories scenarios (WSM: WEC-Sim, PDS: ProteusDS; X = Theory in-cluded).
Weakly Nonlinear (NL) Body-to-Body Interactions (B2B) Dynamic Cables (DM)
WSM Base WSM NL X WSM B2B X WSM B2B and NL X X PDS DM X PDS DM and NL X X
Code-to-Code Comparison
A code-to-code regular wave comparison highlights the impact of hydrodynamic theory differences in a known and predictable sea state. Combourieu et al. provided a baseline understanding of a WEC's simulated response in ProteusDS and WEC-Sim by comparing and verifying similar responses of a floating three-body oscillating flap device in different time-domain multibody dynamic modelling software, including WEC-Sim and ProteusDS [1]. Furthermore, Wendt et al. compared and validated the dynamic response of a floating sphere and regular a floating cylinder between numerous WEC numerical modelling software, including WEC-Sim and ProteusDS [3]. The aforementioned studies showed that WEC-Sim and ProteusDS produce highly comparable dynamic responses when utilizing common baseline theories (as described in Section 2.1). As such, this study focuses on the impacts of refined numerical theories through a code-to-code comparison for regular wave scenarios listed in Table 2. Comparison quantities of interest for WEBS are the relative float pitch velocity, the PTO tether tension, and heave plate response.
Relative float angular velocity concerns the WEBS fore float pitching about the nacelle, which houses the PTO, is directly correlated to energy generation. Figure 5 shows the relative float angular velocity oscillation magnitude for the various simulation scenarios.
Including weakly nonlinear Froude-Krylov and hydrostatic force calculations has a clear impact on float response. Specifically, when weakly nonlinear calculations are included, the relative float angular velocity increases. This corresponds to the greater phase difference between the nacelle and fore float with weakly nonlinear terms. Such a phenomenon occurs because the fore float's dynamic wetted surface area increases variation in F exc and F HP calculations. When comparing WSM Base and WSM NL, the inclusion of weakly nonlinear Froude-Krylov and Hydrostatic force calculations is found to increase the relative float angular velocity by an average of 29% over the specified wave frequency range. These findings align with Michele et al. [22], whom find that with the implementation of weakly nonlinear forces, extractable energy increases due to increased float motion. B2B interactions do not have a significant impact on WEBS dynamics. When comparing WSM Base to WSM B2B or WSM NL to WSM B2B and NL, the relative float angular velocity magnitude changes only slightly. When comparing WSM Base and WSM B2B, the inclusion of B2B interactions alters the relative float angular velocity by an average of 1% over the specified wave frequency range. The small impact of B2B interactions may be due to the joint connection between bodies for the WEBS three-part float. Therefore, B2B interactions may still be of importance for rigid bodies in an array as discussed in Starrett et al. [27].
Cable model fidelity has a direct impact on WEC dynamics. Figure 5 shows that all scenarios modelling the tether with a simplified spring have an inverse relationship with wave period. A heave plate helps a WEC capture resonance within a particular frequency range [51]. The simplified spring cable's unable to capture resonance, for motion only increases with decreasing wave period. Therefore, the heave plate's load is not being transferred to the WEC's float through the tether. When the tether is more realistically modelled with a dynamic cable model, the float angular velocity oscillation magnitude increases toward a resonance condition and then decreases. Therefore, the heave plate load is transferred to the WEC's float. The trend difference hinders the determination of a clear quantitative value that a dynamic cable model has on the relative float angular velocity. However, dynamic cable modelling does improve the model's ability to capture peak energy extraction, for the resonance condition matches the design sea state's energy period of 7.9 s. Dynamic cable calculations are shown to directly correspond with the WEC's ability to capture resonance around the design sea state.
The PTO tether connects the hydrodynamically active surface float to the heave plate and plays an important role in the hydrodynamic modelling and the design process. Tension range displays the WEC's load variability due to wave forcing.
As shown in Figure 6 the inclusion of weakly nonlinear calculations reduces tether tension range. This differs from WEBS float motion magnitude, as seen in Figure 5. Recall from Figure 2, weakly nonlinear calculations are dependent on both wave frequency and the extent to which the mesh is submerged (rather than wave frequency alone as with linear calculations). From Equation (2) and Equation (4), F exc and F HP terms are smaller with increased rigid body submergence, and these smaller terms correspond to smaller WEC loads. For WEBS, the nacelle, which attaches to the PTO tether, is continually submerged, so weakly nonlinear calculations formulate a smaller WEC load and therefore a smaller tether tension range as shown in Figure 6. The inclusion of weakly nonlinear Froude-Krylov and hydrostatic force calculations is found to decrease tether tension range by an average of 48% through comparisons between WSM Base to WSM NL. B2B interactions are not shown to impact tether tension. Cable tension is directly related to WEC loads. Figure 6 displays that B2B interactions have little impact on tether tension loads, which agrees with Figure 5 that showed B2B interactions have little impact on WEBS dynamics. Specifically, when comparing WSM Base to WSM B2B, a 0.3% increase in tension range is computed.
When implementing dynamic cable theories, WEBS experiences a significantly larger tension range. Tension with a simplified spring model follows Hooke's law, and has a fairly small and consistent range across varying wave periods. By implementing a dynamic cable theory, tether tension range varies significantly. When comparing WSM Base to PDS DM, modelling with a dynamic cable model increases tether tension range by an average of 240%. The larger range in dynamic cable model scenarios is in part due to the numerical model's inclusion of an axial stiffness term. Axial stiffness is directly related to strain, and tension grows linearly with strain [39]. Additionally, with dynamic cable models, tether tension ranges in Figure 6 follow similar trends with wave period as shown for the WEC's float motion in Figure 5. As such, the larger tension ranges in dynamic cable models correspond to larger WEC loads.
Heave plates play a key role in WEC dynamics and PTO efficiency, and provide a reaction force from the relatively stationary water deep below the WEC [39,44]. The heave displacement of a heave plate is directly correlated to how well the heave plate is performing as an artificial seafloor.
Weakly nonlinear Froude-Krylov and hydrostatic calculations and not implemented for the heave plate for any scenario, for it is 60 m deep and does not have a dynamic wetted surface area. However, weakly nonlinear terms at the surface do influence heave plate displacement. They are shown to decrease heave plate displacement when coupled with dynamic cable models, but increase heave plate displacement when the tether is modelled with a simplified spring. Recall from Figure 6 that tether tension does not vary significantly when utilizing a spring cable model. Thus, focusing on scenarios PDS DM and PDS DM and NL, the heave plate displacement magnitude decreases due with weakly nonlinear Froude-Krylov and hydrostatic calculations. Comparisons between PDS DM and PDS DM and NL show that the inclusion of weakly nonlinear Froude-Krylov and hydrostatic force calculations decreases heave plate displacement by an average of 4%. This finding confirms Figure 6, which showed tether loads decreased with the inclusion of weakly nonlinear terms. B2B interactions are shown to not play a significant role in heave plate displacement. Comparisons between WSM Base and WSM B2B show that the inclusion of B2B interactions alters heave plate displacement by an average of 1%. This result is consistent with B2B theory, for the heave plate is far away from any other rigid body, so B2B interactions are presumably negligible.
The utilization of dynamic cable models significantly affects the heave plate response. When included, displacement increases with wave period. This is because with an increasing wave period there is a decreasing phase difference between the heave plate's and wave elevation's oscillations, which increases overall oscillation size. When the tether is modelled with a simplified spring cable, the heave plate's oscillation does not change with wave period, and is due to the inaccuracy in the transfer of loads from the surface to the heave plate. As in Figure 5, the trend difference with cable modelling approaches prevent clear quantitative heave plate depth in Figure 7.
Code-to-Field Campaign Comparison
The code-to-field campaign comparison uses the 448 s wave data from SWIFT buoys for simulating WEBS numerically and validating simulations. Wave reconstruction methods as discussed in Section 4.2 are implemented in this code-to-field comparison. However, weakly nonlinear calculations are not permissible when importing a time series, for weakly nonlinear calculations require the wave induced pressure at each wetted mesh panel [40]. Therefore, an imported time-series in WEC-Sim scenarios that utilize weakly nonlinear Froude-Krylov and hydrostatic calculations would not work, as the rigid bodies begin at the mean water surface elevation as shown in Figure 2a, and there would be disparities present in rigid body position relative to the instantaneous surface elevation [40]. PDS DM and NL is capable of simulating the wave-reconstructed wavefield, for the wave elevation time-series is developed from the frequency domain in the ProteusDS solver. Therefore, only the WSM Base, WSM B2B, PDS DM, and PDS DM and NL scenarios are simulated for wave reconstruction simulations. Comparisons between PDS DM and PDS DM and NL scenarios will show the impact weakly nonlinear Froude-Krylov and hydrostatic calculations has on WEC dynamics, and so this study is still able to validate the relative impact of all the refined numerical theories discussed in Section 2.2.
The WEBS field data is represented by the shaded green region in Figures 8-10 and is based on the probability of occurrence for each 448 s segment of the 8 h WEBS field data. Float angular velocity is directly proportional to WEBS extractable energy calculations. It is clear from Figure 8 that the maximum float angular velocity is underpredicted regardless of refined numerical theory implemented. The underpredictions can be attributed to: (1) inaccuracies in modelling calculation, or (2) the inability of the reconstructed wave field to formulate waves as large as those experienced during the WEBS field campaign due to the wave reconstruction's reliance on regular wave superposition. Particularly, the regular wave superposition approximation involved with the chosen wave reconstruction methodology is unable to produce steep, higher order waves that occur in the field, as per Whitmer et al. [52].
Simulated float angular velocity increases with weakly nonlinear Froude-Krylov and hydrostatic terms, and more closely resembles the float angular velocity experienced in the field. Furthermore, the shifting trend past 5 deg/s for the PDS DM and NL scenario is due to the increased relative rotational velocity between the nacelle and the fore float when employing weakly nonlinear Froude-Krylov and hydrostatics to the numerical model. These findings validate the code-to-code results discussed in Section 5.1. B2B interactions slightly increase float angular velocity, but this difference is small and could be attributed to slight computational differences. Dynamic cable modelling is not shown to play a significant role in simulation accuracy for the surface float's motion, as Figure 8 displays that WSM Base and PDS DM have similar float angular velocity probability ranges. The representation used in Figure 8 underrepresents the differences due to dynamic cable modelling. Figure 5 shows that relative float angular velocity differences due to cable modelling scheme depend significantly on wave frequency. Specific wave conditions are not shown in Figure 8, and so the impact due to dynamic cable modelling is not clearly represented.
Tether tension indicates how well the numerical methods model captures the WEC loads. As with float angular velocity, tether tension is underpredicted regardless of numerical approach, as shown in Figure 9. PDS DM experiences the largest simulated tension range with extreme tension values under 2 kN and over 6 kN. In the field campaign, the 5th and 95th tether tension percentiles are 0.5 kN and 6.6 kN, respectively [47]. Therefore, the majority of the tensions experienced during the WEBS field campaign are met, but the more extreme loads are not captured. The tether tension underprediction may be a result of loading heave plate hydrodynamics from BEM software, for this method provides smaller hydrodynamic resistance than experimental data, as per Rusch et al. [53].
The inclusion of weakly nonlinear Froude-Krylov and hydrostatic terms slightly impacts the tether range. The linear approach with PDS DM experiences a slightly larger range than PDS DM and NL, which confirms Figure 6. Additionally, the inclusion of B2B interactions does not impact tether tension, for WSM B2B completely overlaps WSM Base as shown in Figure 6. It is clear from Figure 9 that the static spring cable model has a much smaller tension range than experienced by dynamic mooring calculations. This is expected, as Figure 6 showed that with the static spring approach tether tension range is invariant with wave frequency. Dynamic cable models significantly increase the fidelity in capturing the cable's structural dynamics, and provide a better representation of tension and final power production.
Heave plate displacement indicates how well the numerical scenarios capture the WEC dynamics far away from the surface, which is 60 m for WEBS. In Figure 10, the numerical model heave displacement data is superimposed with surface elevation to ensure consistency with the field data.
Weakly nonlinear Froude-Krylov and hydrostatic forces do not have a clear impact on the heave plate displacement, for displacement probabilities in PDS DM and PDS DM and NL consistently overlap one another. However, the linear calculations of PDS DM has a 2% larger probability of a minimum displacement of 0 m than PDS DM and NL. In Figures 5 and 8, we found an increase in float angular velocity with the inclusion of weakly nonlinear terms. This increase in the WEC's float motion corresponds to a more dynamic heave plate response. Therefore, the 2% difference due to the inclusion of weakly nonlinear terms is expected. B2B interactions are shown to not impact heave plate dynamics in Figure 10. This insignificance, matches what we saw in Figure 7 and found in Section 5.1. The heave plate is far away from any other rigid body, thereby making inter-body coefficients miniscule or nonexistent. Finally, dynamic mooring calculations are not shown to significantly impact heave plate displacement accuracy. Regardless of chosen numerical theories implementation, the rigid body dynamics appear well-captured.
Conclusions
This study's objectives concern: quantifying how the incorporation of weakly nonlinear Froude-Krylov and hydrostatic calculations, body-to-body interactions, and/or dynamic cable computations impact a WEC's dynamics; and validating the performance of these numerical simulations against field measurement data.
Utilizing a case study WEC and field deployment data from the WEBS field campaign and SWIFT buoy data, these refined hydrodynamic theories were verified and validated. It is well understood that multibody WEC dynamics are highly intermingled and interrelated. Therefore, this study quantifies the impact of each refined hydrodynamic theory to better understand theoretical significance in WEC dynamics.
Weakly nonlinear Froude-Krylov and hydrostatic calculations are shown to significantly increase the float motion of a WEC, and therefore increase extractable energy. For the regular wave tests performed in the WEBS case study, the inclusion of weakly nonlinear Froude-Krylov and hydrostatic terms increase the fore float's relative angular velocity about the nacelle by a maximum of 29% when the simulations also use a static spring cable model, and by a maximum of 54% with dynamic cable models. Weakly nonlinear Froude-Krylov and hydrostatic calculations have a clear impact on WEC dynamics, and the relative impact of this theory is dependent on additional refined hydrodynamic modelling techniques.
B2B interactions were negligible in this study. For the regular wave tests performed in the WEBS case study, the inclusion of B2B interactions changed the fore float's relative angular velocity about the nacelle by less than 5%. However, the small impact of B2B interactions may be due to the particular WEC's joint, which connects rigid bodies at the surface.
Dynamic cable modelling is shown to significantly impact the overall performance of a WEC. Particularly, dynamic cable models are shown to improve the simulated capacity to capture resonance around the design sea state as discussed in Section 5.1. Furthermore, with dynamic cable models, tether tension range increases by 300%, thereby showing significantly more variability in simulating WEC loads due to wave forcing.
This research highlights the continued need for future investigation on the impact of differing hydrodynamic theories, particularly, for floating wave energy converters. Specifically, the impact of B2B interactions may be examined for WEC systems with numerous disconnected but closely situated rigid bodies. Furthermore, incorporating more realistic waves into the numerical wave field through wave reconstruction or nonlinear wave methods provide an improved ability to do code-to-field data comparative fidelity.
Both rigid body hydrodynamics and cable dynamics must be well-represented to ensure accurate WEC modelling. Implementing refined hydrodynamic theories into a WEC numerical model may significantly improve simulation accuracy. This study found: (1) that simulating body-to-body interactions did not alter WEC dynamics, and (2) that the inclusion of weakly nonlinear Froude-Krylov and hydrostatic force terms and dynamic cable models within a numerical model significantly improve simulated WEC dynamics. reconstruction techniques performed. M. Leary would also like to acknowledge Bret Bosma for modelling guidance and support.
Conflicts of Interest:
The authors declare no conflict of interest.
|
2021-09-28T23:13:44.748Z
|
2021-07-01T00:00:00.000
|
{
"year": 2021,
"sha1": "685879aa784c17380144f03810c6a182139253e9",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/1996-1073/14/13/3959/pdf",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "348e466dfef3804e4f719a77c51e1fa13d7f41f1",
"s2fieldsofstudy": [
"Engineering"
],
"extfieldsofstudy": [
"Computer Science"
]
}
|
55654210
|
pes2o/s2orc
|
v3-fos-license
|
Preparation of Porous Materials by Magnesium Phosphate Cement with High Permeability
High permeability and strength magnesium phosphate cement (MPC) with porosity, average pore size, and compressive strength varied from 63.2% to 74%, 138.7 μm to 284.7 μm and 2.3MPa to 4.7MPa, respectively, were successfully prepared by combining the physical foaming method and chemically entrained gas method at room temperature. )e effects of borax content, chemical foaming agent content, zinc powder content and W/S ratio on the porosity, pore size distribution, compressive strength, and permeability of theMPCwere investigated.)e results indicate that the chemical foaming agent content tends to have little impact on the porosity and compressive strength, and the zinc powder content has the most significant influence on the average pore size of MPC.)e air pores distribution and connectivity of MPCwere mainly controlled by the borax content, W/S ratio, and chemical foaming agent content. Zinc powder played a destructive role in the pores formed by the early physical foaming and led to an increase in pore size and a large number of through pores, which increased the permeability of the materials.
Introduction
Cement-based foam material has high heat capacity, excellent fire resistance, and low cost and is usually used in building energy efficient materials because of its lightweight and thermal insulation properties [1,2].It has also been widely applied in acoustic insulation [3,4], electromagnetic wave absorbing material [5], and safety block material [6].In recent years, magnesium phosphate cement has also been developed to prepare porous materials to obtain foamed concrete for cast-in-situ construction and high-temperature resistance of cement-based porous materials [7].
Phosphate cement is fabricated by an acid-based solution reaction between a divalent or trivalent oxide and an acid phosphate or phosphoric acid [8][9][10].e phosphate used in this system is potassium dihydrogen phosphate, sodium dihydrogen phosphate, or ammonium dihydrogen phosphate.
e metal oxides often used are magnesium, aluminum, zinc, and calcium oxides, and magnesium oxide is the most common oxide to prepare magnesium phosphate cement (MPC) [11].In this reaction, a phosphate gel was formed as a precursor of the ceramics since the metal oxide dissolved to release cations reacted with the hydrolytic phosphate ions, and the main final phase of hydration product is struvite, but many other phases also existed during or after hydration reaction, such as dittmarite (NH 4 MgPO 4 •H 2 O), schertelite ((NH 4 ) 2 Mg(HPO 4 ) 2 •4H 2 O), newberyite (MgHPO 4 •3H 2 O), and magnesium phosphate hydrate Mg 3 (PO 4 ) 2 •4H 2 O [12,13].With the in-depth research of MPC, MPC has received an astonishing amount of attention since it was discovered in the nineteenth century [14,15].Due to its series of advantages, such as rapid setting, high early strength, strong bonding strength, excellent biocompatibility, and low drying shrinkage, it has been widely used in biomedical fields, civil engineering repair, and stabilization of nuclear or heavy metal [16][17][18][19][20][21].Recently, the preparation of porous MPC was reported in some studies such as that of Li and Chen, who prepared a new type of MPC with a density ranging from 210 to 380 kg/m 3 and compressive strength ranging from 1.0 to 2.8 MPa by a prefoaming method [22].Liu et al. also fabricated MPC with a maximum compressive strength of 0.30 ± 0.05 MPa and porosity of 83.75% by the physical foaming process [23].Ma and Chen [7] also obtained a novel foamed concrete with the characteristics of quick setting and high early-strength by using sodium bicarbonate as a foaming agent.However, imperviousness and low strength are the dominant properties of MPC fabricated by the preforming method mentioned above.Recently, the application of cement-based porous materials for hazardous wastewater, such as heavy metals and the radioactive nucleus has attracted more and more attention [24,25].Unlike the applications mentioned above, adsorbent materials require good permeability and high strength to form self-supporting systems, thus achieving good service performance.
In this study, magnesium phosphate cement is used as the base material, and porous materials with high water permeability and strength will be prepared by combining the physical foaming method and chemically entrained gas method, using a chemical foaming agent (CF) and zinc powder as the compound foaming agent.Additionally, the effects of borax content, CF content, zinc powder content, and the water to solid ratio (W/S) on the pore size distribution, connectivity, and compressive strength of the MPC were systematically investigated.
Materials and Experiments
In the current work, the raw materials for the MPC formation were dead-burnt magnesia (MgO, Liaoning Xinrong Mining Co., Ltd., China) obtained by calcinating magnesium carbonate at 1700 °C, and analytically pure ammonium dihydrogen phosphate (ADP, AR-grade, Jinshan Chemical Reagent Co., Ltd., Chengdu, China) and quartz powders were prepared by ball-milling quartz for 30 min.e chemical foaming agent (CF, the primary chemical composition is sodium dodecyl sulfate, alkyl amido betaine, and citric acid) having weak acidity was prepared by our lab, which was combined with zinc powder (Zn) for use as a compound foaming agent.Borax (Na 2 B 4 O 7 •10H 2 O) and analytically pure citric acid monohydrate (CAM) were used as a retarder in this study.As characterized by a laser particle size analyzer, the mean particle diameters of the MgO, ADP, and quartz powders are 29 μm, 60 μm, and 20 μm, respectively.e formulas for fabricating MPC are listed in Table 1, where M/P, M/Q, and Zn/CAM represent the mass ratio of MgO to ADP, MgO to quartz powder, and zinc powder to CAM, and the value is 1, 0.5, and 1, respectively.e CF, zinc powder, and CAM were weighed by the mass of all the solid powders, and the mass of borax weighed the MgO.Additionally, all the water used in this work was tap water from the laboratory.
First, the raw materials including MgO powder, ADP powder, quartz powder, and borax were mixed for 1 min in a vertical-axis planetary mixer according to Table 1.Secondly, the CF, CAM, and water were added and stirred for 1 min, and then zinc powder was mixed with them and stirred rapidly for 90 s. irdly, the slurry after mixing was cast into steel moulds with a size of 40 mm × 40 mm × 160 mm.Finally, the specimens of MPC were demolded after 2 h and cured in the lab at a temperature of 20 ± 2 °C and a relative humidity of 60 ± 5%.It should be pointed out that the CF and CAM were dissolved in water in advance and then added into the mixer.
e sample for SEM is prepared by cast slurry into Φ50 mm × 150 mm column steel moulds, cured at a temperature of 20 ± 2 °C and a relative humidity of 60 ± 5% for 28 d, and cut horizontally into five test blocks for analyzing the pore distribution by SEM pictures.
Before testing porosity, the samples should be dried under the temperature of 60 °C for 24 h to obtain a constant weight.e porosity of material was calculated using P � (1 − (ρ apparent /ρ th )) × 100% with an apparent density of the dried material and theoretical density [26].
e compressive strength of 28 d was applied by a microcomputer control universal testing machine (CMT5105, Shenzhen SANS Testing Machine Co., Ltd., China) with a loading rate of 2.4 kN per second according to standard Chinese GB17671-1999, each sample taken six specimens tested, and the compressive strength is the average value of six samples.Crystal phases were determined by X-ray diffraction using a D/Max-RB (Rigaku, Japan) powder X-ray diffractometer using CuKα irradiation generated at 60 mA and 35 kV; the scanning rate was 8 °/min from 3 to 80 °. e microstructure of the MPC was observed by scanning electron microscopy (TM1000, Hitachi, Japan), and the distribution of the pore size was determined by analyzing SEM pictures using a soft image analyzer (Nano measurer, China).
Results and Discussion
e porosity, average pore size, and compressive strength of the porous MPC fabricated according to Table 1 were measured and the results are listed in Table 2.It can be seen from Table 2 that the porosity increased from 66% to 71.6%, the compressive strength reduced from 4.7 MPa to 2.6 MPa, and the average pore size ranged from 138.7 μm to 160.01 μm as the borax content increased from 4% (M I-1 ) to 10% (M I-3 ).
is result can be attributed to the retarding effect of borax, which can delay the setting time of MPC slurry.As a result, there is enough time for the air bubbles to expand and migrate in the slurry.As the content of CF increased from As the content of zinc powder and CAM increased from 0.5% to 1.5%, the porosity and average pore size increased from 65.2% to 74% and from 151.9 μm to 284.7 μm, respectively, while the compressive strength decreased from 4.6 MPa to 2.3 MPa.ese results may be explained by the reaction between the zinc powder and hydrogen ions which were jointly supplied by the ADP and CAM. e more zinc powder and CAM were added, the more air bubbles were obtained, which led to the higher porosity and lower compressive strength.Furthermore, as the W/S ratio increased from 0.14 to 0.18, the compressive strength reduced from 4.3 MPa to 2.5 MPa due to the porosity and pore diameter size rising from 63.2% to 72%, and from 141.8 μm to 200.4 μm, respectively.e higher W/S ratio makes the consistency of the slurry and foaming resistance lower, which led to a large number of air pores left in the samples and the high porosity obtained.
Figure 1 shows the effect of borax content, CF content, zinc powder content (or CAM content), and W/S ratio on the pore size distribution of the MPC.As shown in Figure 1(a), with the borax content (B%), increased from 4%, 7% to 10%, the pore distribution became wider, ranging from 40 μm to 550 μm, from 80 μm to 1150 μm, and from 70 μm to 1280 μm, respectively.Besides, the majority of pores uniformly ranged from 40 to 300 μm, and the accumulative frequency exceeded 70%.As the content of CF (CF %) increased from 1.0%, 1.5% to 2.0%, the air pore size distribution varied about from 80 μm to 780 μm, 80 μm to 950 μm, and 90 μm to 1280 μm, respectively, as observed from Figure 1(b).However, the CF content has little influence on the smaller air pores, which range in size from 0 to 150 μm.It can be seen from Figure 1(c) that the air pore size distribution ranged from about 0 to 700 μm, from 50 μm to 1300 μm, and from 100 μm to 1350 μm, respectively, as the zinc powder (Zn%) increased from 0.5%, 1.0% to 1.5%.Additionally, a distinct difference can be found that there are about 25% air pores whose size distribution varied from 300 μm to 1350 μm with 1.5% zinc powder in the samples.As seen in Figure 1(d), as the W/S increased from 0.14, 0.16 to 0.18, the size distribution of the air pores varied from 40 to 600 μm, from 50 μm to 1300 μm and from 40 to 750 μm, respectively.Besides, the size of the pores distribution became wider for the specimens with the W/S ratio of 0.16 compared to the other two ratios.
By analyzing the data in Table 2 and Figure 1, there is a certain relationship between the strength of the phosphate cement porous materials and the porosity, pore size of asprepared materials.In general, the greater the porosity, the lower the strength, and the larger the average pore size, the smaller the strength.But comparing the porosity and average pore size, the effect of porosity is higher than the effect of pore size on the strength; for M II-1 , M II-2 , and M II-3 , the porosity is almost the same, and the average pore size increases from 140 μm to 180 μm, but their strength is maintained at about 3.2 MPa; for M II-1 and M IV-1 with almost the same average pore size of 140 μm, the porosity of M II-1 and M IV-1 is 70.4% and 63.2%, the strength of M II-1 and M IV-1 is 3.2 MPa and 4.3 MPa, and the results indicate that the porosity has a huge influence on the strength.Of course, the foaming agent will also affect the strength.In this paper, zinc powder is used as a chemical foaming material, and the strength of different zinc powders is analyzed, such as M III-1 , M III-2 , and M III-3 in Table 1, and the strength decreased from 4.6 MPa to 2.3 MPa with increasing amount of Zn. is is due to the increase in the amount of zinc powder, the increase in porosity and the average pore size, and also due to the side effects of zinc on the hydration process [27].
erefore, for the strength of phosphate cement porous material, the effects are multifaceted, and the influence of porosity plays a leading role than other factors.
Figure 2 shows SEM micrographs of the MPC formulated with different borax content, CF content, zinc powder content, and W/S ratio.As shown in Figures 2(a)-2(c), the size of air pores and the number of connected pores increased obviously when the borax content increased from 4%, 7% to 10%, and the air pores were distributed more uniformly.e pore distribution uniformity and the size of the samples increased slightly as the content of CF increased from 1.0%, 1.5% to 2.0%, but the specimen having the most connected macropores was that with the foaming agent content of 1.5%, as presented in Figures 2(d)-2(f ).As can be seen in Figures 2(g)-2(i), the size and connectivity of the air pores increased significantly as the zinc powder content increased from 0.5 wt.% to 1.5 wt.%.However, the pores were distributed less uniformly as higher contents of zinc powder were used in the samples.As seen in Figures 2(j)-2(l), the size and uniformity of the pores increased obviously when the W/S ratio increased from 0.14 to 0.18.As shown by the consequence of Figure 2, the size and uniformity of the pores were mainly controlled by the borax content, zinc powder content, and W/S ratio, while the CF content mainly controlled the connectivity of the air pores.
We selected the sample M III-2 for XRD analysis shown in Figure 3. Figure 3 shows that the primary phase is SiO 2 , MgO, and MgNH 4 •6H 2 O, of which the SiO 2 peak is the strongest, MgO second, and MgNH 4 •6H 2 O the weakest.e main reason is that SiO 2 mainly acts as a filler and does not participate in the hydration process.Due to the large amount of unreactive SiO 2 , the di raction peaks are sharp.Of course, the incorporation of silica also has more e ect.On the one hand, it can prolong the setting time, which is bene cial to the foaming e ect of Zn powder and forms more pores, on the other hand, silica can also reduce the hydration heat.e reduction of hydration heat avoids the release of ammonia gas during the reaction, while also avoiding the decomposition of hydration products which leads to the strength loss of the materials [28].
Figure 4 shows the water permeability of MPC fabricated by combining the physical foaming and chemically entrained gas methods.It can be seen that water can pass quickly from the surface to underlying layers of the MPC through the abundant connected pore channels that exist in its interior.In general, permeability is mainly a ected by pore size and connectivity of samples.e porous materials obtained by composite foaming technology have a large number of connected pores and large pore size, which can ensure wastewater to easily pass through when used as adsorbent materials.
is paper mainly uses zinc powders as the foaming agent.e previous study also showed that the use of zinc powder could form a better closed-hole structure which has been discussed more detailed, and zinc powder content, water to cement ratio, and borax content play a role on the formation and growth of foams [29].e chemical medium used in this paper is a weak acid foaming agent, which is to reduce its e ect on the slurry of magnesium phosphate cement due to the acidic environment of the initial Advances in Materials Science and Engineering hydration and to ensure that its strength has no significant loss.In the initial stage of molding, a large number of micropores were formed during the stirring process due to the action of the foaming agent CF.At this time, the effect of zinc powder foaming was not apparent.In the static stage, due to the acidic environment at the initial stage of the phosphate cement reaction, together with the low viscosity in the early stage of hydration, the silica fillers prolonged the hydration hardening time, and the zinc foaming effect was exerted.However, at this time, the zinc powder is mainly located on the pore wall formed in the earlier stage.erefore, during the foaming process, the pores formed in the early stage were damaged and enlarged, and a large number of through-holes are formed in the structure so that the water permeability of the material is significantly increased.
Conclusion
In the present work, magnesium phosphate cement (MPC) with high permeability and strength was successfully fabricated by combining the physical foaming method and chemically entrained gas method at room temperature.e porosity, average pore size, and compressive strength of little influence on the porosity and compressive strength of specimens.e pore size distribution is significantly influenced by the zinc powder content and W/S ratio.e compressive strength is affected by the borax content, zinc powder content, and W/S ratio.Additionally, MPC with higher porosity and larger pore diameter shows high permeability due to the mass of through pores existing in the samples.As a result, the porous MPC has enormous potential for filtration applications, such as heavy metals, radioactive waste, exhaust fumes, and could also be used for bio-scaffolds in tissue engineering.
Figure 1 :
Figure 1: Pore size distribution of porous MPC fabricated with di erent borax content in (a), varying foaming agent content in (b), varying zinc powder content in (c), and W/S ratios in (d).
Table 1 :
e formulas for fabricating MPC.
II-1 ) to 2.0% (M II-3 ), the average pore size rose from 140.6 μm to 180.11 μm, but the porosity and compressive strength remained at about 69% and 3.2 MPa, respectively.
Table 2 :
Porosity, average pore size, and compressive strength of porous MPC.
|
2018-12-08T15:14:15.952Z
|
2018-09-09T00:00:00.000
|
{
"year": 2018,
"sha1": "663fb1582b9e6f61fd435161b7d0c11e7704c9a8",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1155/2018/5910560",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "663fb1582b9e6f61fd435161b7d0c11e7704c9a8",
"s2fieldsofstudy": [
"Materials Science"
],
"extfieldsofstudy": [
"Materials Science"
]
}
|
15385049
|
pes2o/s2orc
|
v3-fos-license
|
On Hahn-Banach type theorems for Hilbert C*-modules
We show three Hahn-Banach type extension criteria for (sets of) bounded C*-linear maps of Hilbert C*-modules to the underlying C*-algebras of coefficients. One criterion establishes an alternative description of the property of (AW*-) C*-algebras to be monotone complete or additively complete.
Introduction
The Hahn-Banach theorem belongs to the central results of Banach space theory, together with the open mapping theorem, the closed graph theorem, and others. When investigating the generalized Hahn-Banach extension problem for Banach C*-modules, their C*-submodules and bounded C*-linear maps thereof to the C*algebra of coefficients, we must resort to case studies since a general answer is out of reach at present. The goal of the present paper is to find criteria for the particular class of Hilbert C*-modules, i.e. of Banach modules over C*-algebras A that admit an A-valued inner product, [22,18]. We aim to preserve the C*-algebra of coefficients as the codomain of the extended maps.
In comparison to the particular case of Hilbert spaces the problem is non-trivial since Hilbert C*-submodules of Hilbert C*-modules are in general not direct summands, and bounded module maps of C*-submodules to the C*-algebra of coefficients may be non-extendable to bounded module maps on the entire Hilbert C*-module. To give an example, let A = C([0, 1]) denote the C*-algebra of all continuous functions on the unit interval [0,1] and I = C 0 ((0, 1]) the subset of all continuous functions vanishing at zero. The corresponding A-valued inner product is defined by f, g A = f g * for f, g ∈ A. The map f ∈ I −→ g(t) = sin(1/t) · f (t) ∈ I, t ∈ [0, 1], is a bounded module map on I which cannot be continued to a bounded A-linear map on A preserving the codomain C*-algebra A. Moreover, the property of a Hilbert C*-submodule N of a Hilbert C*-module M of being an orthogonal direct summand thereof sometimes depends on the choice of the C*-valued inner product on M, even if the derived norms are equivalent, [9]. In other words, we are faced by a difference between the properties of C*-submodules of being an orthogonal or merely a topological direct summand, cf. Example 3.4.
Hahn-Banach extension problems for bounded module maps have several origins and applications. For Hilbert C*-modules the problem to extend bounded module maps was first touched in W. L. Paschke's papers [22,23]. Knowing that the analogue of Riesz representation theorem for bounded module functionals may fail, in general, W. L. Paschke analyzed the conditions under which the A-valued inner product on a Hilbert A-module {M, ., . } over a fixed C*-algebra A can be continued to the Banach A-module M ′ of all bounded A-linear maps r : M → A turning it into a Hilbert A-module, among other similar extension problems. A particular reinterpretation of this problem is to ask for the existence of isometric extension of bounded module maps r on M to those on M ′ preserving the values of r if it is restricted to the canonically isometrically embedded copy of M in M ′ . Later on H. Lin [20], M. Hamana [13] and the author [5,6] continued the study of extension problems of bounded module maps and C*-valued inner products on Hilbert C*-modules and realized the key role played by the order structure of the positive cone of the underlying C*-algebra of coefficients. In particular, these results mark the departure from a topological point of view on Hahn-Banach extension problems for bounded module maps on pairs of Hilbert C*-modules since there exists a commutative AW*-algebra ( [15]) described by E. E. Floyd [3] for which there does not exist any Frechét topology with respect to which both the partial order structure of its positive cone (which is monotone complete) and its linear structure are continuous at once. The reader is encouraged to consult the publications [13,5,6] and, especially, to study H. Lin's work [20] for more detailed information on the achieved module map extension results and their proofs. We refer to them at the appropriate places of the present paper.
A second root and motivation have been results by G. Vincent-Smith [31], G. Wittstock [33,34], Ching-Yun Suen [30], P. S. Muhly and Qiyuan Na [21], A. M. Sinclair and R. R. Smith [29] on Hahn-Banach type theorems for completely bounded module maps on matricial normed Banach C*-(bi-)modules (i.e. operator (bi-)modules over C*-algebras) if the range of the extensions is allowed to grow inside a (larger) injective C*-algebra, for example the injective envelope of the C*algebra of coefficients ( [12]). Hilbert C*-modules have the advantage that they can always be represented as operator modules on Hilbert spaces, and that bounded C*linear maps between them are automatically completely bounded, [34,Prop. 2.7,2.8]. We give more details on this point of view in the last section.
To formulate the Hahn-Banach type criteria we obtained the reader should be aware that a Hilbert A-module {M, ., . }, its A-dual Banach A-module M ′ and its A-bidual Banach A-module M ′′ are connected by two canonical isometric embeddings M ֒→ M ′′ ֒→ M ′ , where surjectivity may fail at any position and any possible interrelation may be realized by examples, cf. [22,23]. Moreover, the process taking A-duals stabilizes since the A-dual of M ′′ equals the A-dual of M. Also, the A-valued inner product on M can be always continued to an A-valued inner product on M ′′ , cf. [23,Th. 2.4]. To define some properties of C*-algebras that come into the play we say that a C*-algebra A is monotone complete if every bounded increasingly directed net {a α : α ∈ I} of self-adjoint elements of A has a least upper bound a = sup{a α : α ∈ I} inside A. It is said to be additively complete if every norm-bounded infinite sum of positive elements possesses a least upper bound inside A. For more details we refer to the next section were the preliminaries will be explained. The implications (iii)→(ii) and (iii)→(i) have been established by Huaxin Lin [20,Th. 3.8]. The equivalence (iii)↔(iv) has been discovered by K. Saitô and J. D. M. Wright [28, §3]. The goal of the present paper is the equivalence (i)↔(ii) and the implication (i)/(ii)→(iii)/(iv). During the course of proof, we obtain a result valid for particular pairs of some Hilbert C*-module and one of its Hilbert C*-submodules without any restrictions to the properties of the C*-algebra of coefficients: Then every bounded A-linear mapping r : N → A can be continued to a bounded A-linear mapping r ′ : M → A so that (a) r = r ′ , (b) r ′ restricted to N equals r and (c) the extended mappings of N ′ form a Banach A-submodule of M ′ , and (d) the extensions {r ′ n : n ∈ N } of the standardly embedded mappings {r n = ., n : n ∈ N } ⊆ M ′ coincide with the latter, if and only if every bounded A-linear mapping r : N → A can be continued to a bounded A-linear mapping r ′ : M → A so that (a) r = r ′ , (b) r ′ restricted to N equals r and (c) r ′ restricted to N ⊥ ⊆ M equals zero.
This happens if and only if the A-bidual Hilbert A-module M ′′ of M is the orthogonal direct sum of the orthogonal and the biorthogonal complement of N inside M ′′ with respect to the continued A-valued inner product ., . , i.e. M ′′ = N ⊥⊥ ⊕ N ⊥ . Moreover, the A-bidual Banach A-modules N ′′ and (N ⊥⊥ ) ′′ coincide isometrically.
To weaken the additional assumption on the embedding φ : N ′ → M ′ to be isometric we can suppose that it has to be merely bounded and A-linear with closed range, cf. Example 3.4. Going with these assumptions we can formulate the following result: (ii) N ≡ N ⊥⊥ ⊆ M and there exists a bounded A-linear injective mapping φ : N ′ → M ′ such that φ(r)[n] = r(n) for every r ∈ N ′ , every n ∈ N .
Without the assumption of M being A-reflexive (ii) does not imply (i), in general.
The next section contains some definitions and facts from the literature that have to be stated for proving. The third section is concerned with metric aspects of the proof, whereas the forth section deals with its inner product aspects. The proof of the theorems is divided into a number of propositions which are of independent interest. We close the considerations with a remark on the structure of particular extensions of C*-valued functionals on Hilbert C*-modules and on the relation of our results to Hahn-Banach type problems for operator modules.
Preliminaries
In this section we give definitions and basic facts about Hilbert C*-modules and C*-algebras necessary for the proofs of the theorems. The papers [22,16,4,19,20,9], some chapters in [14,32], and the books by E. C. Lance [18] and by I. Raeburn and D. P. Williams [27] are used as standard sources of reference on Hilbert C*module theory. We make the convention that all C*-modules of the present paper are left modules by definition. A pre-Hilbert A-module over a C*-algebra A is an A-module M equipped with an A-valued mapping ., . : M × M → A which is A-linear in the first argument and has the properties: x, y = y, x * , x, x ≥ 0 with equality iff x = 0 .
The mapping ., . is said to be the A-valued inner product on M. A pre-Hilbert A-module {M, ., . } is Hilbert if and only if it is complete with respect to the norm . = ., .
1/2
A . We always assume that the linear structures of A and M are compatible. Two Hilbert A-modules are isomorphic if they are isometrically isomorphic as Banach A-modules, if and only if they are unitarily isomorphic, [18]. Moreover, Banach A-modules can carry unitarily non-isomorphic A-valued inner products that induce equivalent norms to the given one, nevertheless, [9].
Lemma 2.1. Let A be a C*-algebra acting on a Banach A-module M in such a way that A•M is norm-dense in M.
Then every element x ∈ M can be decomposed as x = ay for some y ∈ M and some positive a ∈ A with a ≤ 1. Moreover, for any given ε > 0 an estimate x − y < ε can be observed.
If M is a Hilbert A-module then by construction the element a can be selected from the smallest norm-closed two-sided ideal of A containing the value x, x .
As an immediate consequence we obtain that any The next statement characterizes monotone complete and additively complete C*-algebras by alternative conditions. For example, W*-algebras and commutative C*-algebras A = C(X) with stonean compact spaces X are monotone complete, cf. [15,20,13,5].
C*-algebra A is monotone complete if and only if A is additively complete, if and only if for every Hilbert
As a proving technique we need a basic construction to switch from a given Hilbert A-module M to a bigger Hilbert A * * -module M # while preserving many useful properties and guaranteeing the existence and uniqueness of extended operators and A-(A * * -)valued inner products, (cf. H. Lin [19,Def. 1.3], [22, §4]). For example, the C*-valued inner product on a Hilbert W*-module M over a W*algebra B can always be continued to an C*-valued inner product on its C*-dual Banach W*-module M ′ turning it into a self-dual Hilbert W*-module, [22,Th. 3.2]. Therefore, for Hilbert W*-modules we have M ′′ ≡ M ′ . Moreover, taking the biorthogonal complement N ⊥⊥ of a C*-submodule N of a self-dual Hilbert W*module M and forming its C*-dual Banach W*-module N ′ gives the same subset of M under the canonical indentifications usually made. Also, the C*-linear hull of the completed with respect to the topology {|f ( ., x )| : x ∈ N , f ∈ B * } unit ball of N may be identified with both N ⊥⊥ and N ′ inside M in this situation, [4,Th. 3.2]. We refer to [22,19,1,4] for more details.
Let {M, ., . } be a left pre-Hilbert A-module over a fixed C*-algebra A. The algebraic tensor product A * * ⊗ M becomes a left A * * -module defining the action of A * * on its elementary tensors by the formula Consider the unique extension of T to a bounded A * * -linear operator on the self-dual Hilbert A * * -module (L # ) ′ , also denoted by T . Taking advantage of the special properties of Hilbert W*-modules and of the extension operation # as described above we obtain that both ker(T ) # and (ker(T ) ⊥⊥ ) # are contained in the biorthogonal complement of ker(T ) ⊆ M ֒→ (L # ) ′ estimated with respect to (L # ) ′ . What is more, the latter is part of the kernel of the extended to (L # ) ′ operator T since T becomes adjointable there and admits a polar decomposition, [32,Prop. 15.3.7] and [22]. We arrive at a contradiction.
Since Banach C*-modules may admit different C*-valued inner products that induce equivalent norms to the given one and, nevertheless, are not unitarily equivalent (cf. [9]) one further comment is necessary: the arguments do not depend on the choice of the A-valued inner products on M and N within their respective unitary equivalence classes since on (L # ) ′ any two A * * -valued inner products are unitarily equivalent by self-duality and the kernel of T is always an orthogonal direct summand there, cf. [22].
The metric aspects of the proof
This section is devoted to the various Banach C*-modules that are related to a pair {M, N ⊆ M} consisting of a Hilbert C*-module {M, ., . } and one of its Hilbert C*-submodules N , and to their interrelation. The main emphasis is put on pairwise containment relations, isomorphisms and direct sum decompositions. The orthogonality of C*-submodules is merely used for their definition and for their identification with other C*-submodules. We start with two simple observations concerning C*-reflexive Hilbert C*-modules and their properties. Since M is C*-reflexive and N ⊥⊥ is a Hilbert C*-submodule of M we have the equality π(r) N ⊥⊥ = π(r) M ′′ . However, π(r) acts on N ⊥⊥ ֒→ M ′ as r acts on N ⊥⊥ ֒→ (N ⊥⊥ ) ′ , and therefore π(r) N ⊥⊥ = r (N ⊥⊥ ) ′′ for any r ∈ (N ⊥⊥ ) ′′ . So the map π is isometric. Moreover, the map π is onto since The isometric inclusion N ′′ ֒→ (N ⊥⊥ ) ′′ ≡ N ⊥⊥ follows from Lemma 3. , n for n ∈ N ). Then φ has the property that for every r ∈ N ′ the element φ(r) ∈ M ′ equals zero on N ⊥ ⊆ M.
Proof. By Lemma 3.5 of [20] the map φ is uniquely defined by the condition that it extends the canonical embedding of N into M ′ . Therefore, if A is a W*-algebra we can refer to [22,Prop. 3.6] for the properties of φ, and the desired property can be obtained from there.
In case A is an arbitrary C*-algebra we turn to the bidual situation and use the construction described in the preliminaries, i.e. switch from the two Hilbert To show the converse conclusion (ii)→(i) consider an element x ∈ M ′′ i ֒→ M ′ . The element x ∈ M ′′ applied to M ′ defines a bounded A-linear functional on the subset φ(N ′ ) ⊆ M ′ that belongs to N ′′ . By Lemma 3.2 the biorthogonal complement N ⊥⊥ of N with respect to M ′′ can be identified with N ′′ preserving the canonical A-valued inner products. Therefore, there exists an element y x ∈ N ′′ ≡ N ⊥⊥ ⊆ M ′′ so that φ(r)[y x ] = φ(r)[x] for any r ∈ N ′ . That is, x = y x + (x − y x ), and the summand (x − y x ) ∈ M ′′ vanishes on φ(N ′ ) ⊆ M ′ . The map x ∈ M ′′ → y x ∈ N ′′ is bounded and A-linear by the assumed properties of φ. Consequently, M ′′ admits a topologically direct sum decomposition as +Ñ , whereÑ is the set {x ∈ M ′′ : φ(r)[x] = 0 for r ∈ N ′ }. This is condition (i).
The additional assumption of M being A-reflexive can in general not be dropped. V. M. Manuilov gave an example of a Hilbert W*-submodule N in a non-self-dual Hilbert W*-module M that coincides with its biorthogonal complement with respect to M and, nevertheless, is not a topological direct summand of M. We realize that N additionally has the property (ii) required by Theorem 1.3.
To be more precise, let A = l ∞ be the W*-algebra of all bounded complex-valued sequences, c 0 be the ideal of all sequences converging to zero and select the Hilbert A-module M = l 2 (c 0 ) ⊕ l 2 (l ∞ ) with its standard A-valued inner product. Fix the injective operator T ∈ End A (l 2 (c 0 )) defined by T ({a k } k ) = {1/k · a k } k and take the Hilbert A-submodule Note, that N coincides with its biorthogonal complement N ⊥⊥ with respect to M.
The Hilbert A-submodule N can be realized as the kernel of a bounded nonadjointable module operator on M.
However, N is not a topological direct summand of M, whereas N ′ is an orthogonal direct summand of M ′ . To see this, assume the existence of a topologically direct decomposition M = N . + L for some Hilbert A-submodule L of M. The resulting A-linear bounded idempotent operator P : M → N can be described by two operators S 1 ∈ End A (l 2 (c 0 )) and S 2 ∈ End A (l 2 (l ∞ ), l 2 (c 0 )) via the formula P : M → N , P ((x, y)) = (T S 1 (x) + T S 2 (y), S 1 (x) + S 2 (y)) .
Relying on the property (id M −P ) 2 = (id M −P ) we consider the action of (id M −P ) on the special subsets {(0, y) : y ∈ l 2 (l ∞ )} and {(x, 0) : x ∈ l 2 (c 0 )} of M and obtain the operator equalities By the injectivity of T they can be transformed to the operator equalities which are valid on l 2 (l ∞ ) and on l 2 (c 0 ), respectively. Consequently, the operator (S 1 T +S 2 −id) equals zero on l 2 (c 0 ) = ran(S 1 +S 2 ) and hence, on its A-dual Hilbert A-module l 2 (l ∞ ) ′ by [22,Prop. 3.6]. However, S 2 = id + S 1 T on l 2 (l ∞ ) ⊂ l 2 (l ∞ ) ′ since equality would cause S 1 to be unbounded by the range restriction to S 2 and T . This contradicts the assumption of the existence of a bounded A-linear idempotent operator P : M → N .
The inner product aspects of the proof
To continue the proofs of Theorem 1.1 and 1.2 we have to take into account aspects of Hilbert C*-modules related to orthogonality of elements, and of C*submodules. So we leave the sphere of considerations on the Banach C*-module level and deal with orthogonal direct sum decompositions and orthogonal complements.
First, we show that the statements (i) and (ii) of Theorem 1.1 are fulfilled for C*-algebras A that possess a monotone complete multiplier C*-algebra M (A). In particular, the result reproduces [34,Th. 4.1] for the special set of operator C*modules that consists of all Hilbert C*-modules. However, the restriction to the Hilbert C*-modules enables us to enlarge the class of well-behaved (in the sense of a Hahn-Banach theorem) coefficient C*-algebras beyond the set of injective C*algebras because this restriction rules out most of the existing more complicated operator C*-modules. To show the converse we should give some facts about A-reflexive Hilbert C*modules over C*-algebras A fulfilling assertion (ii) of Theorem 1.1: Proof. Let {L, ., . L } be another Hilbert A-module containing M as a Banach Asubmodule, i.e. housing a bicontinuously embedded copy of the Banach A-module M. We show that both M and its biorthogonal complement M ⊥⊥ with respect to L possess the same A-dual Banach A-module M ′ . Indeed, consider an element r of the isometrically embedded copy of M ′ inside L ′ that exists by Theorem 1.1,(ii) and has the properties claimed there. Restrict r to the submodule M ⊥⊥ ⊆ L. We obtain an element r ′ of the isometrically embedded copy of (M ⊥⊥ ) ′ inside L ′ . The difference (r − r ′ ) vanishes on M since r and r ′ coincide there, and it vanishes on M ⊥ ⊆ L by the supposed properties of the embedded A-dual Banach A-modules formulated at Theorem 1.1,(ii). By Lemma 2.4 the element (r − r ′ ) has to vanish on M ⊥⊥ ⊆ L, too. This proves our claim. Investigating the pair of Hilbert A-modules M ⊆ M ⊥⊥ we know that M is A-reflexive and that the Hilbert A-modules share the same A-dual Banach A-module. These facts together with Lemma 3.1 force them to coincide.
Consider an arbitrary element x ∈ L and the restriction of the A-valued bounded functional ., x L to M. By supposition condition (ii) of Theorem 1.1 holds. So there exists another A-valued bounded functional r x ∈ L ′ with the properties that r x restricted to M ⊆ L coincides with ., x L and that r x restricted to the orthogonal complement M ⊥ of M inside L equals the zero mapping. We want to find a Hilbert A-module containing both M isometrically as a Hilbert A-submodule and a copy of r x . Let us show that the value r x , r x has a meaning and belongs in fact to A. For this aim consider the self-dual Hilbert A * * -module (L # ) ′ which is constructed from L in the canonical way described in the preliminaries. The two bounded Alinear functionals ., x L and r x can be continued to elements of (L # ) ′ , cf. [22,19]. The lifting of the A-valued inner product on L to an A * * -valued inner product on (L # ) ′ allows to define inner product values for any two elements of L ′ ⊆ (L # ) ′ , in particular for ., x and r x . The A * * -valued functional ( ., x −r x ) vanishes on M by construction. The extension of ( ., x − r x ) from L to (L # ) ′ preserves this property and extends its validity to the biorthogonal complement of M taken with respect to the self-dual Hilbert A * * -module (L # ) ′ , cf. Lemma 2.4. However, by construction and by self-duality the biorthogonal complement of M taken with respect to the self-dual Hilbert A * * -module (L # ) ′ contains r x , [22,4]. We obtain the equality applying the functional to the particular element r x . In other words, the value r x , r x (L # ) ′ = r x , x (L # ) ′ = r x (x) * is an element of A. Consequently, the Avalued inner product on M can be continued to an A-valued inner product on the Banach A-submodule of M ′ generated by M and r x , reducing the A * * -valued inner product on (L # ) ′ first to the embedded subset M ′ and than to the Banach A-module of interest. That way the range of this inner product reduces to a subset of A, cf. [22,19]. Now, by Lemma 3.1 the assumption r x ∈ M ֒→ M ′ leads to a contradiction. Indeed, in this case we have constructed a Hilbert A-module containing M as a proper Hilbert A-submodule and possessing the same A-dual Banach A-module M ′ by construction, which contradicts the supposed A-reflexivity of M.
Summing up, for every element x ∈ L there exists an element r x ∈ M ⊆ L so that the orthogonal decomposition x = r x ⊕ (x − r x ) holds inside L. Hence, M is a direct summand of L, and M is orthogonally complementary as a Hilbert A-module by definition. To see that any bounded module operator on M admits an adjoint consider M as a Hilbert M, M -module and apply Proposition 2.2. x, x ∈ A}. The set N is an A-module. Indeed, the condition x, x ∈ A allows a representation of x ∈ M as x = ay for some a ∈ A, y ∈ M by Lemma 2.1 and, therefore, for x 1 , x 2 ∈ N with x i = a i y i for a i ∈ A, y i ∈ M, i = 1, 2 we obtain = a 1 y 1 + a 2 y 2 , a 1 y 1 + a 2 y 2 = a 1 y 1 , y 1 a * 1 + a 2 y 2 , y 1 a * 1 + a 1 y 1 , y 2 a * 2 + a 2 y 2 , y 2 a * 2 ∈ A . Hence, the A-module axioms are fulfilled by the rules for C*-valued inner products and ideals. Furthermore, N is complete with respect to the norm on M.
We want to demonstrate that the A-dual The second assertion on the converse conclusion follows from Lemma 3.1 if we continue bounded module mappings r ∈ (N ⊥⊥ ) ′ to bounded module mappings r ′ ∈ M ′ by simply setting r ′ = r • P , where P : M ′′ → N ⊥⊥ is the projection existing by assumption.
To give an example of a situation where N is not a direct summand of M ≡ M ′′ , but its biorthogonal complement N ⊥⊥ is, consider the W*-algebra A = M of all bounded linear operators on a separable Hilbert space as a Hilbert A-module over itself and set I = N to be the Hilbert A-submodule of all compact operators. For this pair the generalized Hahn-Banach theorem holds.
As an example for the case M ≡ M ′′ consider the Hilbert A-module M = A ⊕ I equipped with its canonical A-valued inner product for the same W*-algebra A. The Hilbert A-submodule N = {(i, i) : i ∈ I} coincides with its biorthogonal complement taken with respect to M, but it is not an orthogonal direct summand of M. By contrast its biorthogonal complement N ⊥⊥ = {(a, a) : a ∈ A} in the A-bidual Hilbert A-module M ′′ = A ⊕ A of M certainly is an orthogonal direct summand.
The following example shows that the bidual Banach C*-modules of a Hilbert Proof of Theorem 1.1 and 1.2. Proposition 3.3 demonstrates that the first set of conditions stated at Theorem 1.2 implies the second one. To show the converse implication consider two elements r 1 , r 2 ∈ N ′ , an element a ∈ A and the extended to M ′ elements s := (r 1 + r 2 ) ′ − r ′ 1 − r ′ 2 , t := a(r ′ ) − (ar) ′ . By assumption s, t ∈ M ′ vanish on both N and on N ⊥ ⊆ M. Since the orthogonal complement of N ⊕ N ⊥ with respect to M is trivial Lemma 2.4 applies, and s = t = 0 on M. So by assumption the subset of M ′ consisting of all extended elements of N ′ has the structure of a Banach A-module inherited from that of the Banach A-module N ′ . Moreover, for x ∈ N the element ( ., x − ( ., x ) ′ ) ∈ M ′ vanishes on both N and N ⊥ , too, by assumption. The same argument as above yields ., x ≡ ( ., x ) ′ for any x ∈ N .
By the way we have obtained the equivalence of the assertions (i) and (ii) of Theorem 1.
Final remarks
Hahn-Banach type problems for Hilbert C*-modules are closely related to those for general operator modules as treated by G. Vincent-Smith [31], G. Wittstock [33,34], Ching-Yun Suen [30], P. S. Muhly and Qiyuan Na [21], A. M. Sinclair and R. R. Smith [29], and others. Hilbert C*-modules serve as a special class of examples of operator modules, [34,Prop. 2.7,2.8]. All these Hahn-Banach type theorems for completely bounded module maps on matricial normed Banach C*-(bi-)modules (i.e. operator (bi-)modules over C*-algebras) are formulated under the assumption that the the range of the extensions is allowed to grow inside a (larger) injective C*-algebra, for example the injective envelope of the C*-algebra of coefficients ( [12]). Naturally, order aspects play a mayor role in these investigations, cf. [21]. All these theorems give only sparse information about situations with sharper codomain restrictions, so Hilbert C*-modules could serve as a model to develop more general Hahn-Banach type criteria for operator (bi-)modules. Our presented case study demonstrates that monotone complete C*-algebras serve as an upper bound for attempts to extend these results to a larger class of C*-algebras of coefficients and codomains. However, the sharpest restriction has been obtained in collaboration with V. I. Paulsen during a research stay of the author in Houston in 1998. The example belongs to another class of operator bimodules: taking an arbitrary C*-algebra A and its injective envelope I(A), a monotone complete C*-algebra ( [12]), so this pair of operator A-bimodules fulfills a Hahn-Banach type extension theorem for completely bounded A-linear maps into A if and only if the C*-algebra A is injective itself (and coincides with its injective envelope). So the Wittstock extension theorems for operator (bi-)modules turn out to be actually already criteria. Furthermore, after a talk of the author given at a meeting on operator spaces at CIRM in Marseille in July 1998 M. Junge formulated the idea that for an arbitrary pair of an operator bimodule and one of its operator subbimodules over a given C*-algebra A any completely bounded module map of the submodule to A admits a completely bounded extension to the entire operator bimodule preserving the completely boundedness norm and taking values at most in the bidual to A W*-algebra A * * , if and only if A has the weak expectation property in the sense of E. C. Lance [17]. The proof was found during discussions of M. Junge with the author at that meeting. These results will be published elsewhere since they lead beyond the topic of the presented paper, see [10], [11].
As a second addition we want to explain another equivalent condition to the assertions of Theorem 1.1 which is much more technical. We get an idea of how particular mappings r ∈ N ′ may be isometrically extended to mappings r ′ ∈ M ′ failing to be zero on N ⊥ ⊆ M (cf. Example 3.4): Proposition 5.1. Let A be a C*-algebra with the property that for every Hilbert Amodule M and every Hilbert A-submodule N ⊆ M every bounded A-linear mapping r : N → A can be continued to a bounded A-linear mapping r ′ : M → A possessing the same norm so that the restriction of r ′ to N equals r. If additionally the equality r, r = r ′ , r ′ ∈ A * * is supposed to hold for the A * * -valued inner product on the self-dual Hilbert A * * -module (M # ) ′ , then the restriction of each extension r ′ to N ⊥ ⊆ M equals zero and hence, A has the property (ii) of Theorem 1.1.
The point is that the metric equality r = r ′ = r, r + r 2 , r 2
1/2
A can in general be valid for (existing) extensions r ′ of r even if r 2 = 0. We have to conclude that the consideration of any possible extensions r ′ ∈ M ′ of a given bounded Alinear mapping r : N → A (if there is any at all) has to be made separately in every particular situation.
For example, consider the C*-algebra A = C([0, 1]) of all continuous functions on the unit interval [0,1] as a Hilbert A-module M over itself. Set N = {f ∈ A : f ≡ 0 on [1/2, 1]}. The embedding map i : N → A is an element of N ′ which does not belong to the subset { ., n : n ∈ N } ⊂ N ′ . Every function f i ∈ A taking the value 1 on the interval [0,1/2] and satisfying the equality f i = 1 is an extension of i ∈ N ′ inside M ′ = A. However, N ⊥⊥ is not a direct summand of A = M and there does not exist any extension f i of i in M whose restriction to N ⊥ equals zero.
Problem 5.2. Denote by M and N ⊆ M a Hilbert C*-module and one of its Hilbert C*-submodules. For which kind of C*-algebras A is the biorthogonal complement N ⊥⊥ of N inside M or inside its A-bidual Hilbert A-module M ′′ , respectively, always a topological or orthogonal direct summand?
|
2014-10-01T00:00:00.000Z
|
1996-09-30T00:00:00.000
|
{
"year": 1996,
"sha1": "3f96a61167863574b8bed0d1c5e87ed3df29f0a6",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/funct-an/9609003v2.pdf",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "69f70bb3c306fc16dbaa1127220ba6a5ca9e5e32",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
268792239
|
pes2o/s2orc
|
v3-fos-license
|
Radial artery pseudoaneurysm a rare complication after coronary angiography: A case report and systematic review of the reported cases
Key Clinical Message This is a case of 83 years old male who had radial artery pseudoaneurysm after cardiac catheterization. The diagnosis was through Doppler ultrasound and the patient was treated with thrombin injection and reported good outcomes. The literature also included 41 cases of pseudoaneurysm after catheterization. The mean age of patients was 68.5 years with a male prevalence of 49%. Onset of pseudoaneurysm ranged from 0 days (directly after the catheterization) to 150 days with a median of 5 days. The treatment of patients was mainly surgical (19 cases) followed by compression (either manual or TR band) (12 cases), thrombin injection (four cases), compression then surgery (three cases), compression then thrombin injection (one case), percutaneous endovascular repair using a covered stent (one case) and not reported in one case. All cases recovered well.
| INTRODUCTION
Since the discovery of cardiac catheterization, huge improvement has been noticed regarding the mortality, hospital stay, and discharge in many cardiac conditions. 1 Although many access sites can be used for the application of catheterization, access through the radial artery became the most preferred site until now.Such observation stems from the lower risk of complications such as bleeding, morbidity, mortality, and hospital length of stay. 2 Many complications arose from the transradial (TR) access such as radial artery occlusion, spasm, perforation, arteriovenous fistula, granuloma, bleeding, pseudoaneurysm, neurological injury, and compartment syndrome. 3The risk of pseudoaneurysm after the TR access seems to be low and estimated to be less than 1%. 4,5The increasing number of puncture attempts, patients with high bleeding tendency, inadequate compression after catheterization, and the use of large sized sheath are reported to be risk factors for pseudoaneurysm development. 6Pain and swelling along with the radial access site are considered the one of the manifestations of pseudoaneurysm which later confirmed by Doppler ultrasound examination. 7,8erein, this report indicated our experience with one case of pseudoaneurysm development after cardiac catheterization in accordance with a literature review of the existed case reports.
An 83-years-old-male, who is known to have hypertension, hyperlipidemia, benign prostatic hypertrophy, and severe aortic stenosis awaiting trans-catheter aortic valve implantation (TAVI), came to our hospital for elective coronary angiogram as a part of pre TAVI assessment.The examination was normal regarding: temperature, heart rate, blood pressure, peripheral oxygen saturation, and neurological examination.Respiratory examination demonstrated minimal bibasilar crackles heard bilaterally.Cardiovascular examination showed normal 1st and 2nd heart sounds, S4 gallop, crescendo-decrescendo harsh systolic ejection murmur heard at the right 2nd intercostal space radiating to the carotids bilaterally, and minimal pitting lower limb edema.All the laboratory data were normal including complete blood picture, renal and liver function tests and troponin levels; except for a high Pro-BNP value (1161.0pg/mL).The patient was also negative for human immunodefieincy virus and hepatitis C and B viruses.
After the injection lidocaine 1% local anesthesia and the intravenous injection of 5000 unit heparin, a 6Fr sheath was placed in the right radial artery using modified Salinger technique.Selective coronary arteriography was then performed using a 5Fr JR4 diagnostic catheter to engage the right coronary artery and 5Fr JL3.5 diagnostic catheter to engage the left coronary artery.Right and left coronary angiographies were performed in multiple views by hand injections of contrast.Following the procedure, the sheath was removed and hemostasis was achieved by using TR Band at 10 cc of air.
Two days post angiography the patient was presented to our hospital complaining from small bulge in the wrist associated with pain.
| METHODS
Local hand examination showed a right wrist swelling associated with redness and pain, intact pulses with no coldness or paresthesia.Sonography indicated an oval heterogeneous structure in the right lateral wrist measuring 3.31 × 1.58 cm, demonstrated swirling flow of PS of 83 cm/s as evident by different color signal within the lesion (yin yang sign) in Doppler US with a narrow neck measuring 0.15 cm communicates with the distal radial artery (Figure 1).The findings were consistent with partially thrombosed pseudoaneurysm arising from the distal radial artery.Using ultrasound guidance, sterile technique and local anesthetic, a 25-gauge needle was advanced into the right radial artery pseudoaneurysm, targeting the anterior portion, 0.2 mL of 1000 units of thrombin mixed with normal saline was injected.
| CONCLUSION AND RESULTS
Post injection imaging showed no color flow of the pseudoaneurysm.The patient was discharged and followed for 1 month with no apparent complications.
| DISCUSSION
A literature search was performed in five databases until 28th July 2023: Virtual Health Library (59), PubMed (54), Google Scholar (467), Web of Science (43) and Scopus (63).The following keywords were used for the search: "radial artery pseudoaneurysm," "post catheterization" "post cardiac catheter" "post percutaneous coronary intervention" "post PCI" "post coronary intervention" "catheterization" "catheter."Any case report that reported pseudoaneurysm of the radial artery following catheterization was included.Exclusion criteria: reviews, abstract only articles and nonrelevant articles were excluded.The literature included 41 patients from 32 case reports (Figure 2 and Table 1). 6, Themean age of patients was 68.5 years with a male prevalence of 49%.Onset of pseudoaneurysm ranged from 0 days (directly after the catheterization) to 150 days with a median of 5 days.The treatment of patients was mainly surgical (19 cases) followed by compression (either manual or TR band) (12 cases), thrombin injection (four cases), compression then surgery (three cases), compression then thrombin injection (one case), percutaneous endovascular repair using a covered stent (one case) and not reported in one case.All cases recovered well.This is the first reported case in the Middle East region that developed pseudoaneurysm after cardiac catheterization.Few case reports had discussed the development of radial artery pseudoaneurysm post cardiac catheterization in the current literature.Most of the reported cases came from high, upper, or lower middle income countries which indicated a high quality of care towards patients as well as performing more interventional procedure compared to resource limited settings presented in low income countries where surgery remains the main therapeutic strategy. 39Of the reported 41 patients from the literature and by including our case report, pseudoaneurysm prevalence was equal in both males and females.Such observation is crucial for health care professional as vascular complications were reported to be higher in females rather than males. 40he onset of presentation of pseudoaneurysm is variable among the existed case reports with a median of 5 days from the 41 reports.In our case, the onset of development was 2 days post cardiac catheterization.The case report of Cauchi et al, demonstrated that pseudoaneurysm developed in a 45 years old male directly after finishing the procedure and presented with sharp pain. 8However, the onset may be delayed to reach 4 months post interventional cardiac procedure that was reported by Sharma et al, in a 77 years old female. 33Interestingly, the 56 years old male that was reported by Korabathina et al, had developed pseudoaneurysm after 5 months of performing cardiac stenting through the radial access. 25Per se, follow up of the patient for local complications in the access site is still of paramount interest by the clinicians as the vascular complications may be presented months after the intervention.For the best of our knowledge, such delay should be under investigation in the future studies where precipitating factors may play a role in pseudoaneurysm presentation months after the procedure. 25,33Carrying a heavy object was a triggering factor for inducing pseudoaneurysm in a 32 years old male after 2 months of the catheterization. 21Though, detailed history taking could solve the mystery of why pseudoaneurysm may develop months after the catheterization.
Surgery was the main therapeutic strategy in treating pseudoaneurysm in almost half of the reported cases.This was followed by manual compression or TR band compression as the second most common reported therapy.However, few cases reported the use of thrombin for treating such complications. 29,30Our patient was successfully treated with thrombin injection and favorable outcome was achieved.Additionally, thrombin injection was reported as a backup treatment after failing of compression technique in the case report of Bauer et al. 18 The choice of choosing thrombin injection in the treatment of pseudoaneurysm should be considered as an alternative approach to the surgical technique.While weighting the risks and benefits of applying this treatment strategy should be used wisely by the treating physicians especially in patients with a high risk of bleeding tendency.
In all the reported case reports, no patient had developed severe complications after pseudoaneurysm treatment by different treatment strategies.Indeed, one patient had suffered from paresthesia postsurgical treatment. 6And another patient had asymptomatic radial occlusion after using compression technique as a treatment strategy. 38This indicates the good outcomes of pseudoaneurysm when treated with many therapeutic techniques.
Pseudoaneurysm of the radial artery after the cardiac catheter is considered a rare event.Curious follow up of the patients especially those who reported local symptoms in the access site is recommended.Furthermore, choosing the best technique for treating such event should be wise according to the general health status of each patient.
1 2 Surgery
Representing the reported cases in the literature.Breast cancer, chronic hepatitis-C-virus infection, hypertension and permanent atrial fibrillation MI, persistent atrial fibrillation, hypertension, and moderately depressed left ventricular function artery disease, atrial fibrillation treated with apixaban, diabetes mellitus type II, hyperlipidemia, chronic kidney disease stage III
|
2024-04-01T05:07:26.372Z
|
2024-03-29T00:00:00.000
|
{
"year": 2024,
"sha1": "a7beadd40daf8cf065eafa0d70d029225aa03cc2",
"oa_license": "CCBYNCND",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "a7beadd40daf8cf065eafa0d70d029225aa03cc2",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
73673318
|
pes2o/s2orc
|
v3-fos-license
|
Confession Validity in Civil Right in Iranian Law and International Conventions
Confession importance in retribution and even in civil affairs is so that law scientists call it as key reasons. Its importance is that is always in claims such as the most important reason to it. In principal in punishment affairs, confession accused without there is investigation on his/her validity confession and there is documentaries on its affirm, there is not subjectivism and maybe there is from knowledge and presumption from court located on accused crime identification, therefore there is not worth confession that it obtains from compulsion, threat and torment and it has no way to it and taking confession to this way is perfect breach for civil law in law examination. There is one of sensible antitype retribution justice, observe civil law in judgmental courts there is the most important pillar in observation civil right, security and freedom in judgement sessions and especially confession controversy is the same subject. Guarantee civil law and judgement execution all of judgement procedures including crime discovery, pursuit, investigation, punishment execution or security and cultivation actions are basic assignments of retribution justice. In this study there was paid to specify confession place in judgements and volume its credit with confirm on civil rights.
Introduction
Obtain reason is the radial law subjects and especially retribution justice.We examine two subjects in obtain reason, first there are using reasons such as tool and affirmation resource or ascribe, are there observe lawful in obtain that items?In other words, is there including validity principle for obtain reason or not?Second, has been gathered reasons in legal way, could be ascribe crime to accused man in standard way?It is maybe judge ascribe crime to accused man with regard to documents and circumstances, but this judgment breach in order lack of reason sufficiency in resource.(Farakhsheh, 2014).
There is one of the most tension controversial subjects in every court is confession from everybody for his/her tort and/or testimonial against other man.Whether there is order his/her intake rights or other's right confirms or override in court.Confession and testimony subjects in law rules and of course more seriously in retribution problems is focus from law and attorney.From civil right aspect in third edition of civil law means claim affirmation reasons subject, it was paid to confession in civil claims in first book and there was paid to attention to conditions and results of confession.There was subject in 168 and 169 articles of Islamic punishment law: « (168) article-there is pervasive confession that confessor is to be wise, full grown, peon and free during confession.(169) article-the confession that obtained under mandatory, torment and/or physical or mental tease, has not credit and worth and court should re-investigation from accused man again.»Therefore there are the most important points in confession about judge that have perfect and expanded attention is validation confession conditions.Authority and intention are basic pillars in everybody that without those there is not created any retribution responsibility in minimal according Islamic punishment law for actual everybody.What more sentences in right courts are issue according testimonial .
Citizenship Concept
Citizenship concept in social dimension was idea that born in west in Europe and has been published simultaneous with expanding in 16 th century.Accept laws and legal and political duties due to citizenship place is basic and substantial idea this concept.
Political identity, loyalty to political geographic territory, respect to law, careful attention to other's rights and is being to know to privileges and its altitude and finally obey from duties are due to citizenship and enjoying from citizenship right that was created citizen concept in nowadays meaning .
In 14 clause of constitution third principium has been specified to «supply full perfect rights including woman and man and square judgmental security» for all of people and «public equality against law».Return to past of laws (169 principium), acquaintance principium (37 principium), Judgmental supply principium (32 principium) using of attorney right principium (35 principium), and prohibition torment and obtain confession with threat principium (38 principium) aboveboard judgement (165 principium) and compliance given principium (34 principium) are including citizenship rights that has been considered in constitution of Islamic republic in Iran.
Confession Concept and Its Elements and Conditions
Confession is that proving action with his/her saying in word, expressing word in explicit and obvious manner and in term in general there is news and express right for other profit and against self in retribution law that often using for accept execution crime by accused man.
Elements Confession Definition
News: there is kind of news, mean from past till now or informing for future a kind of something and confessor informs he/she has right for another man to him/her.Else profit: third element "being profit for else one.Non including actual or legal person and even there is confession to action that is pensionable Islamic punishment or excuse punishment, although this kind is not profit for else other man, but we could know it this kind of confession for profit society with some neglect.
Confessor loss: confession must pensionable induct loss and stroke to physical body (in retribution law) and/or harm or damage to financial cattle (confessor).Therefore if confession has not harm for confessor and has other harm it is considered as testimony.
1) Confessor or domicile
2) Beneficiary from confession or domicile 3) Confession subject or in domicile 4) Concubine or confession word
Domicile Conditions
1) Maturity: confession from individuals are less than 18 years old is not incisive, although they confession.
2) Wise: confessor should to be wise therefore confession from madman is not incisive.If confessor to be period madman and he/she does confession in mental health and lack of madness this kind of confession is incisive.
3) Growth: confession from daffy men is not effective; of course confession from daffy men is incisive at affairs that are financial.For example, confession daffy man to execution crime that has Islamic punishment or nemesis or excuse punishment such as adultery or accusing man to adultery or buggery is effective.
4) Intention: confessor has intension for informing during confession and words and actions that are accomplish without intention has not legal effect, then confession man is not effective during sleep, unintelligent or in drunken.
5) Authority: confessor should have authority and has not mandatory for it .
Basic Validity and Augment Confession
We refer to 1258 clause civil law in beginning that in first its clause has been considered confession from proving reasons for claims meanwhile confession definition has been mentioned in 1259 clause: «confession is that right informs for else one and in loss for self» although confession manner is extensive and general because according 1260 clause civil law «confession accomplish for every word that implies to it» word every saying in this clause is absolute law it is including every kind of saying for confession although confession should to be explicit and clear.
Because there was mentioned on 1268 clause in civil law «suspected confession is not effective» but whole of proven and denies in confession subject are about confession validity.There was mentioned on (168) clause in Islamic punishment law: «confession is effective that confessor to be wise, mature, peon and free during confession» therefore there are important points that we consider to it with perfect and extended attention about judge, are confession conditions.Authority and intention are two basic pillars in every confession that without two them there is not created retribution responsibility at minimal according Islamic punishment law for any legal person.Whatever many of sentences in right courts are issue according testimony, but is there everybody could testimony with every condition and in every situation for every subject?Exactly no, testimonial has other conditions than domicile (confessor).According (177) clause Islamic punishment law «legal testimonial should have these conditions during testimony: a-maturity, b-wise, c-faith, d-justice, e-purity, f-not beneficial at subject, g-not having hostility with parties or one of them, h-not occupant in pauperize, i-not to be tramper.»so that with attention to these conditions judge could issue sentence according testimonies whatever after being these mentioned conditions again testimonial should attest for proving his/her loyalty .
Confession Validity in Retribution Affairs
Mean of «else» are one or another men except confessor (domicile).
Mean of « on loss self» it means on confessor.It is better to explain it, informing subjects that we are saying or Have informing aspect and they were considered in ordinary talks or therein to law claim and retribution dossier.First kind is not related to out subject but second kind has three situations: a-We declare a thing to benefit for another man and loss self that is confession.
b-We declare a thing on benefit self and other loss that is claim c-We declare a thing on other benefit and another man loss that is testimony or evidence.
According 38 principium at constitution «there is forbidden every kind torment for taking confession and/or obtain information, there is forbidden to constraint to testimony or oath and this kind of testimony or confession and oath has not validity and credibility, offender from this legal principium has punishment.
The important point in this retribution confession is different than civil confession, according 360 clauses of retribution judgmental law related to 2014,«whenever accused man confession to his/her crime exactly, so that there is no doubt and misgiving in confession and also doubt in validity and authority for it, the court is issue sentence according this confession»
Forbidden Torment
Some of Islamic lawyers have been explained torment including every kind tease and torment and there is separator between torment and tease and harassment that is continuity for tease and pain in torment.So that Tousi sheikh explained torment in this way: Torment, is obtain harassment for its owner and this torment is continues for that one.
With deduction to nowadays rules it is could definite torment in this way: every kind of tease and physical harassment that governmental agents or official and public authority owners do those actions during their duties or according their job for taking confession from accused man.Of course in this definition, mental pillar for torment was limited to taking confession from accused man .
In constitution that other rules are to be coincident with it in (38) principium self to affirmation confession sentence due to torment and mentions it: constraint person to testimony, oath or confession is not permissible and this testimony and confession and oath has not credit and value.Offender from this principium has punishment according law.This principium absolutely and without any exception, it declared forbidden every kind torment including physical or mental and with any tool and at every way with perfect precision and it is not credit.Following this following legal principium, it declares confession sentence due to torment: The 9 clause of constitution considered respect to legal freedoms and maintain civil rights: every kind of torment accused man for taking confession and/or constraint him/her to other affairs are forbidden and these confessions have not legal argument.
The 4 clause forbidden torment law: whole of confessions and admissions that took without observing these clauses related to this law and accused man denies those, if there are not reasons and acceptable documents, has not validity and those are not valid.
In convention torment forbidden, this convention knows torment as privative civil and social rights and definite that in this way: First clause: this convection definite torment in this way: every action that accomplish due to pain or intense physical and mental harassment against person for obtain information or taking confession from him/her or third party it calls torment.(also) torment a person such as action that he/she or third party has been executed and/or it is likely he/she executes, with threat and constraint and according discrimination in every kind and when this harassment and tease and/or with in provocation and stimulus and/or with satisfaction and lack of resistance governmental agent and/or other authority owner first ranked person is considered as torment.
Holy Quran
There is predicting verses from holy Quran for validity confession that we are referring to that in following:
First verse
(AL emran, 81 verse) (remembering that when the god got seriously treaty from messengers and their followers) that whenever I gave to you holy Quran and science, then the messenger who came to you that who confirms whatever I gave you, faith to and help him, then (god) said to them: do confession to this subject and contract seriously to that?
They said (yes) we confession, (god told them) (on this treaty) to be witness, I am witness together you.
In this verse, god got confession from people that treaty and promise that they accepted ordered treaty and promise and then put on this promise self and those angels on this confession.However we could conclude from this verse that god put people for accepting their promise and answerable to their treaty and its results due to acceptance, but we conclude anther point from this verse that god considered confessions people to acceptance obligations due to their treaty with self as validity and tool for proving their possibility breaching treaty.
Second verse
(Toubeh, 102 verse) and some of them has confession to their sin (concision) that did good and bad actions, we are hoping that god accept their sin that of course god is forgiving and kind.In this verse some of Arab people confession to their sin and their actions are mixed good and bad actions and from this point they hope to forgiving their sin from god and this verse express this situation (Tafsir al-mizan, 1996, p.376).
Tradition
In controversy from documents for confession in tradition and holy prophet method and stories and news from holy messengers to this following, there has been argument for validity confession: Mohammad Ibn Yaghob is one of our friends … he is one of surgeon and one of son of abi adeallah (greeting to him) said: I don't accept testimony of lecher man unless to his/her tongue».Argument aspect is that lecher is not accepting his/her testimony against another man for lack of justice.Therefore accepting his/her testimony against self is ignorance of testimony reasons and we should searching another basic for his/her testimony validity against self and this affair is not possible unless we know it including confession and validity base.In other word, correct testimony of lecher man against self is to depend on acceptance lecher individuals against self.
Consensus
One of reasons that was predict for validity confession on it, consensus Islam lawyers on validity this base (Tousi, 2000, p.3) (Javaher al-kalam, 1989, p.3) (Riaz al-masael, 2015, p.238) (Ibn roshd, p.471) but with attention to text, we could not to consensus as independent reason for argument in other word consensus is document and it has not validity.
The Wise Men Method
The wise men from a nation with attention to culture level and thinking and their civilization and without considering religious and way and their thinking religious, they believe on whole of items that confession of every wise intentioned and independent man against his/her loss self is valid and obligatory.In addition the judge has denied this way and it is enough for validity confession.Even some of authors go beyond this limit and they know validity and importance confession from a wise man against self is necessary sentence from all of religious and nations.
Compulsion to Confession, Wastage of Citizenship Right
In rules of citizenship right, there was emphasis on observing perfect rights on investigations and forbidden torment and harassment for confession to crime.Judge should not compulsion accused man to ignorance when he/she want to gathering reasons and crime documents for completing dossier, it is mean he/she should not put him/her in horrible space than he/she does confession because constitution says agents should say to accused man why arresting him/her and he/she knows what does he/she do.According 1277 clause civil law, this is confession that is done from fun, kidding and/or compliment.In other word we could say that in exactly in all of items that it is obvious that the domicile has no intention and satisfaction, it causes confession is not valid.Also confession from ignorance to subject or sentence caused confession has not validity basic conditions and it could not has validity .
From Islam retribution punishment even if crime could not discovered and accused man remains without punishment, it is better that accused man did confession because he/she has afraid and in result lack of knowing to his/her right and/or under pressure and he/she punishment.Accused man has silence right and Islam gave this right to him/her that could silent against questions, it is obvious that Islam has more attention to rights and citizenship freedom for every individuals and it not to emphasis to proving all of crime certainly and all of criminals receive punishment.Wastage accused man right, torment, to put pressure accused man for confession to crime, compulsion to confession and doing illegal actions for proving and discovery crime, are items that in constitution, citizenship rights , human rights and human conventions and citizenship and Islamic sentences has emphasis to illegal and its forbidden and there was mentioned as the most important breach items for citizenship right for individuals in judgements and wastage accused man rights and observing square investigation and judgment.
Common Points for Lawmaker in Iran, UN and EU
1) Forbidden using of torment and punishment for accused man principle 2) Not validity for inputs as result of obtain with torment 3) Using of possibility ways for preventing and fighting with torment 4) Training executive governmental agents and supervision on their behaviors 5) Torment definition and its pillar and elements, it is obvious consensus for them as mental and physical pillar of torment.Purposely hurt and severe mental-spiritual torment and interference governmental agents as material pillar and obtain reason or intimidation and severely punishment accused man as mental pillar is including this treaty.
6) Removing hurt from torment about balance judgmental trend with UN and human rights resources is necessary mention, although it is not denied that there are deficiencies in judgmental laws and rules and judgmental department with helping parliament and state should employing with seriously and attempt human rights despite insular of some of persons in perfect supply in rights and freedoms, but news associations, state with approach expanded improves in recent years than meanwhile making clearance Islamic republic of Iran position against important subjects about human rights, civil and political right for citizens than supply it in favorite.
Forbidden torment convention and other behaviors and cruelly punishments and non-human behaviors confirmed public association of UN and (39.46) article at 10 December 2006 in three sections and including 33 articles that provide substantial approaches for states for preventing torment.
Citizenship Observing Right in Iran and International Conventions
International treaty for civil and political right in 9 clause declares it is not arrested anyone headstrong or detained that person, it is not foreclose freedom from anyone unless according aspects and judgmental rules considered to constitution.According statue of international retribution punishment, suspected person to commitment one of crimes under authority judgmental, bureau has right for informed before investigation during trial a reason on to be criminal or considered as his/her innocence, right of enjoying of legal assistances and investigation right at especial attorney unless that criminal has ignored from this right.According 69 clause statues, breaching these procedures causes, bureau not including reasons those obtained result of torment from reasons category.Decisions related to constitution, urgent receive right and immediately information related to reasons for arresting and charges are subjected to him/her and trial right without delay are rights for suspected who has considered statue of crime.
Suspected person has right to refer to unaligned judgmental authority for re-investigation to necessary or lack of necessary detain.This is same affair that at common love and to compliance right according calling detainee arrangement.According one of accepted and proven principals in human rights international rules, detain before trial, when it not exceeds limit rational time, it is permissible law .
Assumption innocence accused man when his/her commitment not proven, it constitutes base and foundation retribution judgmental trend, therefore unduly delay in investigation and/or justifiable detain for suspected, is bullet that targeting heart of judgmental trend.Human rights American convention has mentioned similar rules to 7 clause and African charter for human has referred to necessaries in 6 clause too.
In Iranian Law
Constitution in Iran in principals, was expressed dominant rules on pursuit accused men and in 32 principal was considered « it is not arrested anybody unless the law defines its arrangements, if the person was arrested charge subject should signification to him/her immediately in write and he/she should recognize it in maximal time in 24 hours preliminary dossier is sending to judgmental resources and trial arrangements, is providing immediately.
Offender from this principal according law has punishment».35 principal of constitutions declares «in every courts, parties have right to choose attorneys» and according 38 principal forbidden torment and confession to force were considered on one of innocence principal in pursuit procedure.
According 129 clause constitution of Islamic republic of Iran, judge should investigate identity of accused man and important specification and also home address exactly so that signification and summons and other documents is to be easily and it was mentioned that accused man should careful about saying and then he/she realizes accused charges and its reasons in exactly to accused man.Whenever investigations are begins.Questions should obvious and clear.Induce questions, cheating, reluctance or compulsion to accused man is forbidden.So that accused man refuses from answering, his/her refuse was written in process verbal.Objection right for accused man to detain arrangement was inserted at (33) clause and investigation of review court will be out of alteration.Anyway accused situation should be clear at one month.
Respect rule to legal freedoms and maintain citizenship rights has been paid to some of dominant basics rules on investigation and pursuit procedure.According one clause this clause, pursuit crimes and investigations and issue security arrangements and temporary detain should to be obvious and clear according to observing rules with judgmental sentence and order and it is avoid from execution personal styles and abusive from power and/or execution violence and/or additional detains and without necessity.In arresting and investigation or wanting information and investigation, it is avoid from tease individuals such as closing eye and other organs, letdown and lowering to them.According (7) clause all of investigators and investigation agents should avoid to cover face accused man or sit behind him/her or taking them to invisible places and in general illegal actions.According 9 clause unity article there is forbidden every kind torment for taking confession from accused man and/or compulsion him/her to another actions and takin confessions have not validity legal.Also all of investigations and examinations, should to be according legal practical procedures and previous trainings and under necessary supervision and when individuals ignore these rules and arrangements and resort to illegal methods for execution their duties, we have to pursuit them according considered law (10 clause)
In International and Regional Documents
The 10 clause of human international act, was considered audit right to individuals claims by independent and unaligned court in fairly and aboveboard and declares these courts could take decision about retribution charge for individuals.Also according 11 clause of this act, audit to public claim and charge of individuals should execution with observation whole of necessary securities for defense.Also nobody for doing or not doing that during commitment that action according national or international rights is not crime, has not trial .
Also European convention supporting of human right and substantial freedoms in (6) clause has referred to these securities and also it is 8 clause of human rights American convention has confirmed fairly judge as one of important results of innocence principle.also human rights African charter some of judge principles including defense right, trial right during rational period time at independent court, lawsuit right in authority national resources has included in (7) clause.
International retribution bureau statue in (67) clause, at especial dominant principles on fairly judge, accused man has right for every charge that has trial in aboveboard, fairly and unaligned and with observation following these considerations in equal: a-In language that accused man understanding it and speaks in that language, he/she should to be informed from nature, cause and content of charge immediately and exactly.b-Court should give time and necessary possibilities for defending him/her and could communicate freely and in secret with attorney that chosen.
c-Accused man should not have trial without unjustifiable delay.
d-With observation 2 clause of 63 article and for defending, attendants in court and/or via chosen judgmental counselor and informed that if not enjoying of judgmental counselor, he/she has right for enjoying from judgmental counselor that statue provides for him/her in necessity and in straightforward justice execution, even have not financial power for paying related expenses for that.
e-He/she could ask questions from witnesses himself/herself or by another one that whom were introduced against him/her and according same conditions that those witnesses had against himself/herself ask related witnesses to himself/herself have attendance in court session and ask them questions.
f-If investigations of court and/or the documents are offering in trial is in another language that accused man speaks it, he/she could enjoying from skilled translator freely and also the translated documents that it is necessary offering at court sessions.
g-Accused man not has to testimony and/or confession to his/her crime and could to be silent without his/her silence was considered as confession or deny.
h-He/she could offer written bill or verbal statements without he/she oath.
i-There is not force to mutual task for proving reasons and/or denies those on accused man.
Discussion
According justify principle, maintain social security and observe plaintiff rights and accused man are considered important principles in retribution judgement and attempt in realization social justice and guarding it is necessary on judgmental department.Creating equilibrium between necessity pursuit criminal and observe rights and protection human munificence of accused man is necessary related to judgmental department.
1) Observing judgmental rights of accused man in discovery crime, preliminary investigations from judgmental department.
2) Being enough executive securities due to breach citizenship right of individuals in judgmental procedures.
3) Forbidden torment and compulsion and threat, tease and non-human punishment execution for taking confession and protection validity for individuals and prevent from issuing unjust trials.
4) Careful and continuous supervision on judgmental procedures with appointment unaligned and informed supervisors on trials affairs.
5) Enjoying defense securities, necessity credit and human prestige related to accused men and necessity recovery comer losses on innocent detainees with emphasis on citizenship rights rules.
6) Removing legal gaps, meanwhile supervision on wellness execution being rules and basic revise with emphasis on religious judgment on Islamic judgmental department.
|
2018-12-30T05:14:07.375Z
|
2016-04-28T00:00:00.000
|
{
"year": 2016,
"sha1": "97da33562bd70b8687a6f462462b7efa50007725",
"oa_license": "CCBY",
"oa_url": "https://www.ccsenet.org/journal/index.php/jpl/article/download/59378/31819",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "97da33562bd70b8687a6f462462b7efa50007725",
"s2fieldsofstudy": [
"Law"
],
"extfieldsofstudy": [
"Sociology"
]
}
|
259970618
|
pes2o/s2orc
|
v3-fos-license
|
Case report: A case report of Alport syndrome caused by a novel mutation of COL4A5
Alport syndrome (#308940) is an X-linked genetic disease with clinical manifestations, such as hematuria, proteinuria, renal insufficiency, and end-stage renal disease. The disease is characterized by the thinning of the glomerular basement membrane in the early stages and the thickening of the glomerular basement membrane in the late stages and may be associated with ocular lesions and varying degrees of sensorineural deafness. Herein, we report a case of Alport syndrome caused by a de novo mutation in COL4A5. The patient was a young male with clinical manifestations of hematuria and massive proteinuria who was diagnosed with Alport syndrome based on renal pathology and genetic testing.
Introduction
The disease was first described by Arthur C. Alport in 1927(Alport, 1927 and was named "Alport syndrome" (AS) in 1961. AS is a genetically and phenotypically heterogeneous disease affecting the glomeruli, cochlea, and ocular basement membrane caused by mutations in collagen IV genes, COL4A3, COL4A4, and COL4A5 (Kashtan, 2021), which can lead to hematuria, proteinuria, and chronic progressive renal dysfunction. Some patients present with sensorineural hearing loss, retinopathy, or other extrarenal manifestations (Savige et al., 2015).
We report a case of Alport syndrome caused by a de novo mutation in COL4A5. This patient was encountered with a hemizygous mutation of COL4A5 c.3604G>A, which is rare and helpful in improving clinicians' understanding of Alport syndrome. revealed mildly impaired filtration function and significantly delayed excretion, with a bilateral renal GFR of 75.01 mL/min. He experienced back pain and yellowing of urine after ECT. After admission, the patient's laboratory test results showed the following values: urine protein +++, urine occult blood +++, albumin 1,480 mg/L, urine microalbumin/creatinine 117.7 mg/ mmol, IgG 4.59 g/L, light chain κ 4.09 g/L, light chain λ 2.44 g/L, 24-h urinary protein 4,679.1 mg/24 h, total protein 49.9 g/L, albumin 31.1 g/L, globulin 18.8 g/L, uric acid 543 μmol/L, urea 8.86 mmol/L, and creatinine 96.5 μmol/L. No significant abnormalities were observed in the hearing test, and only myopia was observed upon eye examination.
The initial diagnosis was nephrotic syndrome, and the patient underwent a renal biopsy to clarify the cause. As shown in Figure 1, light microscopy revealed two strips of cortical renal tissue, 12 glomeruli, one spherical sclerosis, one segmental sclerosis, one tubular glomerulus without glomeruli, mild widening of the remaining glomerular mesangial area, proliferation of mesangial cells, an increased mesangial matrix, well-developed capillary collaterals, and the segmental thickening of the Bowman's capsule wall. PASM-Masson's trichrome staining revealed equivocal complexophilic erythrocyte deposition and moderate chronic renal tubular interstitial lesions (approximately 25%) with multifocal tubular atrophy and basement membrane thickening. The multifocal tubular epithelial cells appeared turbid and swollen with fine granular degeneration; interstitial multifocal foam cell aggregates and infiltration of multifocal mononuclear, lymphatic, and plasma cells were observed. Immunofluorescence results showed negative for IgM, IgA, IgG, C3, C4, C1q, fibrin, HBs, HBc, IgG1, IgG2, IgG3, IgG4, κ, λ, and collagen IV α5 and positive for collagen IV α3 and collagen IV α1. Immunohistochemistry results showed negative for PLA2R. One glomerulus was detected through electron microscopy-light microscopy. Capillary endothelial cells were vacuolated and degenerated with erythrocytes visible in a single lumen; no obvious endothelial cell proliferation and open capillary loops were detected. No obvious thickening or stratification of the renal capsule wall layer or vacuolated degeneration of cells in the wall layer without notable hyperplasia was observed. The thickness of the basement membrane was variable, approximately 200-400 nm, and the dense layer of the basement membrane was thickened and partially torn and arachnoid-like. The epithelial cells in the visceral layer were swollen and vacuolated. Foot processes were mostly fused. The mesangial cells and stroma proliferated. No electron-dense deposits were observed. Vacuolar degeneration of renal tubular epithelial cells was observed. A few renal tubules were atrophic. The interstitium was infiltrated by a few inflammatory cells. Erythrocyte aggregates were observed in the lumen of individual capillaries. The walls of the small arteries were thickened.
The pathological presentation was first considered as Alport syndrome, and the patient was sent for genetic testing.
Genetic test results: the COL4A5 c.3604G>A hemizygous mutation was detected in the peripheral blood DNA of the patient. According to the classification criteria of the American College of Medical Genetics and Genomics (ACMG) (Richards et al., 2015) for the clinical significance of genetic variants, c.3604G>A is a "potentially pathogenic variant." The patient was treated with diltiazem 30 mg twice daily, valsartan 80 mg once daily, and febuxostat 40 mg once daily. The patient was lost to follow-up.
Discussion and conclusion
AS is a hereditary renal disease with both monogenic and biallelic inheritances. More than 2,000 mutants of COL4A5, which encodes α3, α4, and α5 chains of collagen IV, have been identified, while approximately 500 mutants have also been described in COL4A3 and COL4A4. Three classical Mendelian patterns of inheritance in AS exist: hemizygous X-linked due to mutations in COL4A5 (XLAS; MIM#301050; 85% of patients), autosomal recessive due to mutations in the COL4A3 or COL4A4 gene (ARAS; MIM#203780; 10%-15% of patients), and autosomal dominant (ADAS; MIM#104200) (Kruegel et al., 2013;Tryggvason, 2021). Currently, a digenic inheritance (DI) of AS has been reported, which shows the occurrence of two mutations in the α3-4-5 collagen IV gene. Several studies (Fallerini et al., 2014;Fallerini et al., 2017) have identified double-gene inheritance in new AS families using next-generation sequencing, RNA studies, and clinical reassessments, revealing that digenic inheritance explains the highly variable clinical phenotype in AS better than single-gene inheritance. Although many of these are predicted to cause specific changes in the genes, the genotype-phenotype relationship is complex. Autosomal recessive inheritance in female patients with AS leads to renal failure, hearing loss, keratoconus, and central retinopathy earlier than X-linked inheritance in female patients; however, affected relatives and the next generation of affected relatives are less likely to develop renal failure (Wang et al., 2014). Compared with X-linked and autosomal recessive inheritances in male patients, those with autosomal dominant inheritance have mild and slow renal manifestations, and extrarenal manifestations are relatively rare (Kamiyoshi et al., 2016;Furlano et al., 2021). Patients with double heterozygous mutations are believed to develop renal failure later than those with X-linked or autosomal recessive mutations. Despite similar genetic backgrounds, individual performance remains highly variable, and various modifier genes play important roles (Takemon et al., 2021). Mutants can be classified as in-frame, frameshift, missense, non-sense, and splicing mutations (Furlano et al., 2021). Severe mutations (large rearrangements, nonsense mutations, frameshift mutations, and splicing mutations) are associated with the earlier development of end-stage renal disease compared with missense mutations (Bekheirnia et al., 2010;Horinouchi et al., 2018;Mastrangelo et al., 2020).
AS is often diagnosed based on renal biopsy results. Genetic testing has become a non-invasive and definitive diagnostic technique (Adam et al., 2014;Braun et al., 2016;Mallett et al., 2017). In a 27-year-old man presented with hematuria and massive proteinuria, the histopathology of the kidney biopsy showed that the glomerular basement membrane was found with different thicknesses, and the dense layer of the basement membrane was thickened, some of which was torn and cobweb-like. Genetic testing revealed possible pathogenic missense mutations in COL4A5, which confirmed the diagnosis of AS.
Whole-exome capture and sequencing of the genomic DNA of the patient and his parents revealed suspected pathogenic variants that could explain the patient's phenotype. A mutation of base 3604 on the DNA of the coding region of COL4A5 from G to A resulted in a mutation of amino acid 1202 in its coding region from glycine to serine (NM_000495.4:c.3604G>A, p.Gly1202Ser), as shown in Table 1. This variant has not been reported in the population database and is rare, with moderate evidence of pathogenicity (PM2). The variant is a novel missense mutation resulting in an amino acid change that has not been reported previously; however, a variant causing change in another amino acid at the same locus has been confirmed to be moderately pathogenic (PM5). This variant was not detected in the parental sequencing data and caused by a de novo mutation, providing strong pathogenic evidence (PS2). The results of the functional prediction software were biased toward pathogenic variants with a REVEL prediction of 0.872 and a CADD prediction of 25.2. This variant is a probable pathogenic variant according to the ACMG guidelines: 1PS + 2PM, which are shown in Figures 2, 3. The mutation was de novo, and neither parent harbored a mutation at this locus. The mode of inheritance was de novo X-linked.
Glycine substitutions are among the most common pathogenic variants in AS, disrupting Gly-X-Y triplets in the type IV collagen α5 chain (Yeo et al., 2020). Owing to the complex structure of the type IV collagen network, AS-associated glycine missense mutations can lead to the loss of function of a variety of proteins in the glomerular basement membrane (GBM) (Cosgrove and Liu, 2017). Glycine is the only amino acid that has no side chain substituents and can be bent to fill the triple helix structure, which is involved in the formation of disulfide bonds between a-chains essential for the formation of the triple helix structure. Alterations in GBM and abnormal interactions between stroma and podocytes can induce a pathological diagnosis of AS and lead to associated secondary pathological changes, which then cause proteinuria and secondary FSGS (focal segmental glomerulosclerosis) lesions (Savige and Harraka, 2021). This also explains the large amount of proteinuria and secondary FSGS lesions observed by light microscopy.
Single-base substitutions at the last nucleotide position in each exon affect the splicing pattern and may result in splice variants (Teraoka et al., 1999;Sahashi et al., 2007). However, in XLAS, these variants are generally considered missense if transcriptional analysis is not performed, which could potentially underestimate the phenotype of some patients. The patient in this case had a missense mutation in the end-exon nucleotide, which is likely to have affected the mRNA splicing process. Aoto et al. (2021) selected 20 mutations from the Human Gene Mutation Database (professional release 2021.1) and the study cohort, all of which were pathogenic variants caused by single-base substitutions in the last nucleotide position of exons (most of which were glutamate substitutions), and were subjected to splicing analysis. The results showed that 17 variants (85%) exhibited abnormal splicing, suggesting that a single-base substitution at the last nucleotide position in the COL4A5 exon is likely to cause abnormal splicing. This study indicates that patients with aberrant splice variants exhibit a more severe renal prognosis than those with missense variants.
There is currently no specific treatment for AS. ACEI and ARB are often used to treat AS and delay renal failure (Yamamura et al., 2020b;Zhang et al., 2021;Boeckhaus et al., 2022;Zeng et al., 2023).
In conclusion, patients with unexplained hematuria and proteinuria who do not respond to glucocorticoid therapy or who have extrarenal manifestations, such as chronic and progressive renal failure, hearing loss, and ocular abnormalities, should undergo renal biopsy and/or genetic testing as soon as possible to determine the etiology. The identification of COL4A5 mutations in patients with suspected Alport syndrome confirms not only the diagnosis of X-linked disease but also the location and type of mutation, which help predict the clinical course and prognosis of the individual. Early diagnosis of AS is helpful in selecting a more appropriate treatment plan and assessing the patient prognosis. It also plays a guiding role for the patients' family members in reproduction.
FIGURE 2
Results of COL4A5 variant c.3604G>A sequencing and schematic representation of IGV.
FIGURE 3
Schematic of the validation results by Sanger sequencing of COL4A5 variants in the proband and his parents. Above: proband. Middle: proband's mother. Below: proband's father.
Frontiers in Genetics frontiersin.org
Data availability statement
The original contributions presented in the study are included in the article/Supplementary Materials, further inquiries can be directed to the corresponding author.
Ethics statement
The studies involving human participants were reviewed and approved by the Ethics Committee of Zhejiang Provincial People's Hospital. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article. Written informed consent has been obtained from the participant/patient for the publication of this case report.
Author contributions
SP and RY: drafting and refining the manuscript. SL: critical reading of the manuscript. All authors contributed to the article and approved the submitted version.
|
2023-07-19T15:10:12.492Z
|
2023-07-17T00:00:00.000
|
{
"year": 2023,
"sha1": "56961e77e12f5ddd4312f763a1475a517ffa0980",
"oa_license": "CCBY",
"oa_url": "https://www.frontiersin.org/articles/10.3389/fgene.2023.1216809/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "b3f8285b9bc0ebe682488a8cc304a4aa1d311640",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
10356682
|
pes2o/s2orc
|
v3-fos-license
|
Catechin prevents the calcium oxalate monohydrate induced renal calcium crystallization in NRK-52E cells and the ethylene glycol induced renal stone formation in rat
Background Reactive oxygen species play important roles in renal calcium crystallization. In this study, we examined the effects of catechin, which have been shown to have antioxidant properties on the renal calcium crystallization. Methods In the vitro experiment, the changes of the mitochondrial membrane potential, expression of superoxide dismutase (SOD), 4-hydroxynonenal (4-HNE), cytochrome c, and cleaved caspase 3 were measured to show the effects of catechin treatment on the NRK-52E cells induced by calcium oxalate monohydrate (COM). In the vivo study, Sprague–Dawley rats were administered 1% ethylene glycol (EG) to generate a rat kidney stone model and then treated with catechin (2.5 and 10 mg/kg/day) for 14 days. The urine and serum variables were dected on 7 and 14 days after EG administration. The expression of cytochrome c, cleaved caspase 3, SOD, osteopontin (OPN), malondialdehyde (MDA), 8-hydroxy-2′-deoxyguanosine (8-OHdG) in kidney were measured. Furthermore, the mitochondrial microstructure in the kidney was also examined by transmission electron microscopy. Results Catechin treatment could prevent the changes in mitochondrial membrane potential and expression of SOD, 4-HNE, cytochrome c, and cleaved caspase 3 in NRK-52E cells induced by the COM. For the in vivo experiments, the EG administration induced renal calcium crystallization was also prevented by the catechin. The expression of SOD, OPN, MDA, OPN and 8-OHdG, were increased after EG administration and this increase was diminished by catechin. Moreover, catechin also prevented EG induced mitochondrial collapse in rat. Conclusions Catechin has preventive effects on renal calcium crystallization both in vivo and in vitro, and provide a potential therapeutic treatment for this disease.
Background
Nephrolithiasis is the condition marked by the development of renal stones. Renal stones are aggregates of crystals that are formed in supersaturated urine [1]. Calcium oxalate may occur as multiple stones or may recur, can induce pain with both passage and obstruction, and is commonly caused by treatable metabolic disorders of hypercalciuria [2]. It appears as a complex disease that may develop change-able clinical manifestations [3].
Despite the increasing prevalence of the disease in China, little research is available for indicating the mechanism and prophylaxis for renal calculus in detail.
It has already known that exposure to high levels of oxalate and calcium oxalate crystals can induce oxidative stress such as an increase in free radical generation, increased lipid peroxidation, a decrease in cellular antioxidant status and an increase in phospholipase-A2 (PLA2)-induced release of arachidonic acid [4][5][6]. Sustained exposure to high levels of oxalate or calcium oxalate crystals injures the cells [7]. Mitochondria have been demonstrated to show excessive uptake of calcium when the cytoplasm level of free calcium markedly increases, causing abnormalities in the respiratory chain and increasing the mitochondrial production of ROS (Reactive Oxygen Species) [8,9]. Calcium-induced mitochondrial injury can be prevented by antioxidants suggesting that oxidative stress may be an important event in its development [8].
Renal tubular cell injury is induced by the oxidative stress, which is produced during the attachment of crystals to renal tubular cells [10]. Exposure to high concentrations of oxalate can give rise to the generation of ROS, mitochondrial collapse and increased lipid peroxidation, which induces the cell death in cultured renal epithelial cells [11,12]. Renal toxicity is assumed to be caused by the elevation of serum-free iron concentration, following its reduction at the luminal side of the proximal tubule, which generates ROS decreasing antioxidant systems and also leads to the enhancement in lipid peroxidation [13,14]. It has been demonstrate that toxic action of acephate on kidney cells is partly through an ROS-mediated mechanism [15]. And ROS are known to mediate many toxin induced renal tubular injuries [16][17][18].
In renal tubular cell injury, mitochondrial damage has been recognized as a crucial cause for tubular cell death which involves disruption of respiration complexes and loss of mitochondrial membrane potential [19,20]. And cell apoptosis is precipitated by mitochondrial outer membrane permeabilization and consequent release of apoptogenic factors such as cytochrome c [21]. Caspases are crucial mediators of programmed cell death (apoptosis). Among them, caspase-3 is a frequently activated death protease, catalyzing the specific cleavage of many key cellular proteins [22]. Caspase-3 is activated in the apoptotic cell both by extrinsic (death ligand) and intrinsic (mitochondrial) pathways [23,24]. In intrinsic activation, cytochrome c from the mitochondria works in combination with caspase-9, apoptosis-activating factor 1 (Apaf-1), and ATP to process procaspase-3 [22,25,26].
Catechin, regarded as one of the main components of green tea, is considered to exert antioxidant effects acting directly as radical scavengers or metal-chelators and also indirectly through modulation of transcription factors or enzymes [27,28]. As ROS plays an important role in renal calcium crystallization, catechin may have the therapeutic effect.
Therefore, we administered catechins to renal proximal tubular cell and rats with stone formation to investigate its inhibitory effects on urolithiasis. In this study, we examined the changes in oxidative stress in an in vitro study and in vivo study. We further showed that renal calcium crystallization was significantly decreased by catechin treatment. Together, these results provide compelling evidence for a role of oxidative stress, which activates the initial process of renal calcium crystallization and can be prevented by catechin.
Preparation of calcium oxalate monohydrate (COM) crystal suspensions
Oxalic acid (200 mM, 0.5 ml) and 200 mM calcium chloride we remixed at room temperature to a final concentration of 10 mM, and the COM crystals that immediately formed in suspension were equilibrated for 3 days. The COM crystals were then washed three times with sodium and chloride-free distilled water saturated with calcium oxalate, resuspended to a final concentration of 2.92 mg/ml, and adjusted to pH 6.8.
Cell culture
The renal proximal tubular cell line NRK-52E (the Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China.) was cultured in Dulbecco's modified Eagle's medium (GIBCO, USA) and was supplemented with 10% newborn calf serum,100 IU/ml penicillin, and 100 μg/ml streptomycin (Life Technologies, Burlingont, ONT, Canada). The cells were routinely seeded at a density of 1 × 10 6 /60-mm culture dish (Corning Incorporated, Glendale, Arizona, United States) at 37°C in a humidified atmosphere of 5% CO 2 in air. The medium was changed every three day and the cells were subcultured before forming confluent monolayers.
NRK-52E cells were seeded at a density of 1 × 10 6 / 90-mm dish and cultured to 90% confluence. The cells were then treated with or with-out catechin hydrate (Beyotime Biotechnology, Haimen, Jiangsu, China) (0.4 μl/ml) for 10 min and then with COM crystals (80 μg/cm 2 ). In our study, cell suspension was prepared and counted in the blood counting chamber.
Measurement of mitochondrial membrane potential (Δψm)
Cells were loaded with the membrane potential-sensitive dye tetramethylrhodamine ethyl ester perchlorate (TMRE; 20 nM in Hepes-buffered salt solution; Invitrogen, Carlsbad, CA, USA). Cells loaded with TMRE were then analyzed using a confocal micro-scope (LSM5 PASCAL; Carl Zeiss Co. Ltd., Oberkochen, Germany) equipped with × 20 and × 100 oil-immersion, quartz objective lenses. The cells were then treated with or without CsA (Cyclosporin A, 2 μM) for 10 min and then with COM crystals (100 μg/cm2) for 0, 5, 10, 15, and 30 min. As a negative control, untreated NRK-52E cells were ob-served, and as a positive control, we used carbonyl cyanide m-chlorophenyl hydrazone (CCCP; 10 μM), an uncoupler that causes mitochondrial depolarization. Mitochondrial fidelity in cells stained with TMRE was quantified by flow cytometry. After three washes with phosphate-buffered saline (PBS) to remove COM crystals, stained cells in each group were detached using 0.05% trypsin -EDTA (Ethylene Diamine Tetraacetic Acid), washed with PBS, and diluted to 1 ml. A total of 30,000 events were collected from each sample and the data were displayed on a logarithmic scale of increasing red fluorescence intensity using a FACS Calibur HG (Becton-Dickinson, Franklin Lakes, NJ, USA).
Experimental animals
All experimental procedures were performed with the approval of the Animal Care Committee of the Faculty of Medicine, Tongji University. Male Sprague-Dawley (SD) rats (Charles River Japan, Yokohama, Japan), age 7 weeks weighing approximately 280 g were used. Animals were provided with a standard mEq diet (containing calcium, 1.12 g; phosphorus, 0.9 g; magnesium, 0.26 g; and sodium, 0.21 g/100 g; Oriental Yeast Co., Tokyo, Japan) and free access to water. To induce calcium oxalate deposition in the kidneys rats were given 2 doses of 0.12 ml 5% ethylene glycol (EG) (Wako, Tokyo, Japan) and then treated twice per day with catechin through a stomach tube. The rats were assigned to one of the following groups (n = 10 per group) and weighed weekly: one group received EG only (EG group) and two EG and catechin groups also received 2.5 and 10 mg/kg/day catechin (EG + catechin 2.5, EG + catechin 10.0 groups, respectively). At 7 and 14 days after the start of drug therapy, blood was sampled from the inferior vena cave of 5 rats per group. These rats were sacrificed under ether anesthesia and both kidneys were immediately excised and dissected. One kidney was histologically examined and RNA was extracted from the other. Two days before euthanasia, the rats were placed individually in metabolic cages for 24 h and urine samples were collected into cups containing HCl for oxalate measurements. The kidneys and urine samples were obtained from control rats without EG or catechin at day 0 (n = 5).
Western blotting
NRK-52E cells stored at − 20°C were immersed in 1 × lysis buffer and lysed by sonication on ice. The total protein concentration in the supernatant was spectrophotometrically quantified using an Ultrospec 3100 Pro (GE Healthcare, Wallingford, CT, USA). Samples containing 30 μg total protein were mixed with loading buffer (Laemmli sample buffer; Bio-Rad Laboratories, Hercules, CA, USA).
The same membrane was used for catechin (−) COM (+) and catechin (+) COM (+) group to ensure uniformity. The protein expression levels in the bands corresponding to SOD, 4-HNE, cytochrome c, and cleaved caspase 3 (n = 5 each) were quantified using Image Quant LAS 4000 (GE Healthcare Japan, Tokyo, Japan), which is a multipurpose CCD (Charge Coupled Device) camera system for quantitative imaging of blots developed by Amersham for enhanced chemiluminescence, with standard UV (ultraviolet) transillumination for ethidium bromide gel visualization.
Measurement of urinary and serum variables
Urinary volume and urinary pH were measured manually. Urinary calcium and serum creatinine, serum calciumwere determined using an automated analyzer (Model 705, Hitachi, Tokyo, Japan). Urinary oxalate was determined using oxalate decarboxylase and citrate was determined by citrate lyase conversions to oxaloacetate.
Transmission electron microscopy (TEM)
The microstructure of mitochondria in the kidney was examined using TEM. Kidneys were perfusion-fixed using 20 ml of 0.1 M phosphoric acid buffer and 20 ml of 2.5% glutaraldehyde, extracted and washed with phosphoric acid buffer, and then fixed with 2% osmium tetraoxide for 2 h. Tissues were dehydrated using a graded series of ethanol (50-100%), embedded in epoxy resin, and polymerized at 60°C for 48 h. The ultrathin sections (99 nm) were double stained with uranium and lead for observation using a JEM-1011 TEM microscope (JEOL Ltd., Tokyo, Japan).
Statistical analysis
All data are expressed as means ± standard deviation. The statistical significance of differences among groups was examined using the Mann-Whitney U test. A P value of < 0.05 denotes a statistically significant difference.
Results
Changes in mitochondrial membrane potential in NRK-52E cells exposed to COM or COM&catechin NRK-52E cells stained with TMRE, which are aggregated in mitochondria with red fluorescent bodies, while blue fluorescent ones show mitochondria around nuclei ( Figure 1). The negative control cells indicated no change in TMRE intensity from 0 min to 30 min. The TMRE intensity of catechin (−) COM (+) cells exposed to COM crystals gradually decreased during 0 min to 30 min. On the other hand, red fluorescent bodies in catechin (+) COM (+) cells exposed to both catechin and COM crystals had no significant change from 0 min to 30 min. In the positive control, TMRE intensity completely disappeared within 15 min and 30 min.
Form the results of flow cytometry, the peak TMRE intensity of catechin (−) COM (+) cells was shifted to the left at 15 min and shifted slightly more to the left at 30 min, compared to that of the catechin (−) COM (+) cells at 0 min ( Figure 2A). However, in catechin (+) COM (+) cells, the shift in peak TMRE intensity was negligible during 0 min to 30 min ( Figure 2B).
For the expression of cytochrome c, in catechin (+) COM (+) cells, the expression increased at 1 h, 3 h, and 6 h after exposure to COM crystals, but the expression was lower than that in catechin (−) COM (+) cells ( Figure 3D). The change of cytochrome c expression between catechin (−) COM (+) cells and catechin (+) COM (+) cells was significantly observed at 1 h, 3 h, and 6 h after exposure to COM crystals (P < 0.05, Figure 3E).
The expression of cleaved caspase 3 was increased from 0 h to 6 h in catechin (+) COM (+) cells. The expression of cleaved caspase 3 increased from 0 h to 3 h whereas reduced from 3 h to 6 h in catechin (−) COM (+) cells ( Figure 3D). There were significant differences in the expression of cleaved caspase 3 between catechin (−) COM (+) cells and catechin (+) COM (+) cells at 1 h, 3 h and 6 h after exposure to COM crystals (P < 0.05, Figure 3F).
Urinary and serum variables in the control and crystal-model rat
There were no changes in urine volume or urinary pH on 7 and 14 days after EG administration. In the EG group, urine volume was significantly lower than in the EG + catechin10.0 group on 14 days. In the EG group, 24-hour urine oxalate excretion was significantly higher than in the control group after administration (P < 0.05). The 24-hour urine oxalate of EG + catechin2.5 group was more than 5.6-fold and 8.7-fold higher than that of the control group on 7 and 14 days, respectively. The 24-hour urine oxalate of EG + catechin10.0 group was more than 6.4-fold and 8.2-fold higher than that of the control group on 7 and 14 days, respectively (P < 0.05, Table 1).
On the other hand, there were no significant differences in 24-hour urine calcium among the control group, the EG group and the EG + catechin groups. Meanwhile, there were no marked differences in serum creatinine and calcium levels among the control, EG, and EG + catechin groups on 7 and 14 days after administration (P > 0.05, Table 1).
Immunohistochemical staining for SOD, osteopontin, MDA and 8-OHdG in the control and crystal-model rat kidneys
Strong expression of SOD could be observed in the control group, but SOD was not detected in EG group. The expression of SOD was slightly lower in EG + catechin 2.5 group than that in EG + catechin10.0 group (Figure 4).
Osteopontin (OPN) was also tested by immunohistochemical staining. OPN was barely detectable in the (See figure on previous page.) Figure 3 The expression of superoxide dismutase (SOD), 4-hydroxynonenal (4-HNE), cytochromec, and cleaved caspase 3. Proteins were subjected to Western blotting using antibodies against SOD, 4-HNE (A) cytochrome c, and cleaved caspase 3 (D) and quantified by densitometric scanning, results being expressed in arbitrary units. Figure The statistical significance of differences among groups was examined using the Mann-Whitney U test. *, P < 0.05.
control group, but it was densely distributed throughout the renal tubular cells of whole kidneys in the EG group. OPN expression was slightly higher in EG + catechin2.5 group than in EG + catechin10.0 group (Figure 4). MDA was undetectable in the renal tubular cells of whole kidneys in the control group, but it was detectable in EG group kidneys. In the EG + catechin groups, MDA production was undetectable (Figure 4).
Similarly, the expression of 8-OHdG was undetectable in the renal tubular cells of whole kidneys in the control group, but it was detectable in the nuclei of renal tubular cells of whole kidneys in the EG group. In the EG + catechin groups, the expression of 8-OHdG was nearly undetectable (Figure 4).
Finally, the ratios of the OPN, MDA and 8-OHdG expression areas were significantly higher in the EG group than those in the other groups (P < 0.05, Figure 5). Meanwhile, there was no significant difference in the ratios of SOD, OPN, MDA and 8-OhdG expression areas among the control group, EG + catechin 2.5 group and EG + catechin 10.0 group (P > 0.05, Figure 5).
OPN, cytochrome c and cleaved caspase 3 expression changes evaluated by western bolting in the control and crystal-model rat OPN expression increased gradually after EG treatment in rat. In EG + catechin 2.5 group and EG + catechin10.0 group, the expression of OPN was higher than that in EG group (P < 0.05). Additionally, administration of catechin after EG treatment enhanced the expression of OPN. There was a significant change in OPN expression between the control and EG group. A significant differences between EG group and EG + catechin groups were also observed (P < 0.05, Figure 6).
The expression of cytochrome c had no changes among the four groups. There was a significant change in cleaved caspase 3 between the control and EG + catechin 2.5 group (P < 0.05). There were no significant differences among the other three groups for the expression of cleaved caspase 3 in crystal-model rat kidneys (P > 0.05, Figure 6). All the images represent serial sections. EG, EG + catechin2.5, EG + catechin10.0 represent rats administered with EG or EG and catechin at 0, 2.5, and 10.0 mg/kg/day, respectively. Strong expression of SOD was could be observed in the control group, but SOD was not detected in EG group. The expression of SOD was slightly lower in EG + catechin 2.5 group than that in EG + catechin10.0 group. OPN was barely detectable in the control group, but it was strongly expressed in the EG group. The OPN expression in EG + catechin2.5 group was slightly higher than that in EG + catechin10.0 group. MDA was undetectable in the control group and EG + catechin groups, but it was detectable in EG group kidneys. The expression of 8-OHdG was undetectable in the control group, but it was detectable in the EG group. And the expression of 8-OHdG was nearly undetectable in the EG + catechin groups. Figure 5 The results of the semi-quantification of immunohistochemical staining of SOD, OPN, MDA and 8-OHdG in the kidney of the rat on day 14. EG, EG + catechin2.5, EG + catechin10.0 represent rats administered with EG or EG and catechin at 0, 2.5, and 10.0 mg/kg/day, respectively. *, P < 0.05.
Ultrastructural findings of kidneys exposed to EG in the control and crystal-model rat The renal tubules were circular, microvilli were evident in the lumen, and mitochondria were located around the nuclei in the control group (Figure 7).
In contrast, renal tubules of the EG group were thin with flattened tubular cells, the lumen of the renal tubule was expanded, microvilli were barely recognizable, and crystals were present in the lumen (Figure 7). Swollen mitochondria resembling fat droplets around the nuclei had an indistinct, discontinuous, and partly collapsed double membrane. However, the renal tubules of the EG + catechin (EG + catechin 2.5, and EG + catechin 10.0) groups were circular and microvilli were detected in the lumen; they were longer than those in the EG group, and the layer was thicker, but slightly shorter than that in the control group. The mitochondria had a regular internal structure with a continuous double membrane, similar to the mitochondria in the control group.
Discussion
Tea catechins, a subclass of compounds in the flavonoid family, have been found to have several biologically beneficial properties including antioxidative effects [29,30]. Flavonoids leading to cytoprotective effects against oxidative stress. Our findings, taken together with other observations [27,[30][31][32], indicate that catechin demonstrates renoprotective abilities in several models of renal disease, especially nephrolithic nephropathy. The experiments indicate that catechins can attenuate functional and immunohistochemical changes in the renal proximal tubular cell line NRK-52E treated with COM and kidneys of EG induced nephrolithic rats.
In our in vitro study, catechin attenuated the changes of mitochondrial membrane potential, and normalized expression of SOD, 4-HNE, cytochrome c, and cleaved caspase 3 in the renal proximal tubular cell line NRK-52E treated with COM.
The disappearance of TMRE in NRK-52E cells indicated mitochondrial collapse through depolarization of the mitochondrial membrane. Disappearance of TMRE was gradual after exposure to COM crystals. COM crystals stimulate renal tubular cells to generate superoxide (O2•−) via NADPH oxidase [33]. Furthermore, O2• − leads to mPTP opening, which is considered to induce mitochondrial collapse [34]. The opening of the mPTP changes the Δψm, which can be monitored using TMRE, a fluorescent probe that accumulates in polarized mitochondria and is released upon their depolarization. As catechin is a scavenger of free radicals or reactive oxygen species [35], it may attenuate the superoxide released by NADPH oxidase, and leads to normalize potential of the mitochondrial membrane.
Decreased superoxide dismutase and increased 4hydroxynonenal are often used as indexes of oxidative stress and cell injury. The generation of reactive oxygen species have been found to increase after the COM treatment in other studies [1,36]. The increase of antioxidant enzymes (SOD) and lipid peroxidation products (4-HNE) is likely a result of higher level of oxidative stress [37]. Binding of the released cytochrome c to cytosolic apoptosis protease activating factor 1 activates caspase 9 and caspase 3, which results in the induction of apoptosis. During this process, ROS are released from the intramembrane compartment into the cytosol, which further injures renal tubular cells. In our study, the increased expression of cytosolic cytochrome c and cleaved caspase 3 indicated that mitochondrial collapse and cell apoptosis could be induced by COM crystals.
In the in vivo study, oxidative stress was evaluated by SOD, MDA, and 8-OHdG expression in rats with EG induced renal calcium crystallization, which prevented by catechin. Oxalate and calcium oxalate (CaOx) crystals are injurious to renal epithelial cells [38]. For monitoring DNA damage, 8-OHdG are the general marker [39], whereas MDA is examined for lipid peroxidation [40]. Our results showed that, mitochondria in the EG + catechin groups showed a regular internal structure with a continuous double membrane similar to that of mitochondria in the control group, while mitochondria in EG group showed an indistinct, discontinuous, and partly collapsed double membrane. This indicated that catechin prevented mitochondrial collapse, which induced by EG.
In the in rats with EG induced renal calcium crystallization, similar to the observation in NRK-52E cells, we also found the increase of cytochrome c release and the activation of caspase 3, which leads to the apoptosis or necrosis. These lipids disrupt mitochondrial function by increasing ROS, decreasing mitochondrial membrane potential, and increasing mitochondrial permeability [41]. Moreover, calcium oxalate (CaOx) kidney stones are formed attached to Randall's plaques, which is the subepithelial deposits on renal papillary surfaces and triggered by reactive oxygen species [12].
As ROS plays an important role in the process of the development of the nephrolithiasis, antioxidant therapy has been described. In the previous studies, they found that vitamin E administration has been shown to prevent calcium oxalate precipitation in the rat kidney and decreased urinary oxalate excretion in patients with kidney Figure 7 The ultrastructure of rat kidney sections examined by TEM on 14 days. The renal tubules were circular, microvilli were evident in the lumen, and mitochondria were located around the nuclei in the control group. In the EG group, the renal tubules were thin with flattened tubular cells; the lumen of the renal tubule was expanded; microvilli were barely recognizable; crystals were present in the lumen; swollen mitochondria resembling fat droplets around the nuclei had an indistinct, discontinuous, and partly collapsed double membrane. In the EG + catechin (EG + catechin 2.5, and EG + catechin 10.0) groups, the renal tubules were circular and microvilli were detected in the lumen; the renal tubules were longer than those in the EG group, and the layer was thicker, but slightly shorter than that in the control group; the mitochondria had a regular internal structure with a continuous double membrane which was similar to the mitochondria in the control group. Scale bars, 1 μm. stones [36]. It also restored antioxidant levels in the blood and decreased the urinary excretion of oxalate and calcium in patients who underwent surgical stone removal [42]. A previous study of antioxidant enzyme levels in rats with stone formation showed that almost all antioxidant enzyme activities were attenuated except that of catalase [43]. In agreement with previous studies, our results confirmed and extended the observation that the oxidative stress was greatly reduced by catechin treatment, which is an antioxidant component of green tea.
The expression of OPN was evenly and densely distributed throughout renal tubular cells of whole kidneys in the EG group. While in the control group and in the EG + catechin groups, the expression of OPN was barely detectable or localized to limited renal tubular cells. OPN, which identified as a major component of stone matrix protein, its expression was remarkably increased in the renal tubular cells of stone-forming rats [44,45]. Some studies suggested that OPN is an inhibitor of abnormal calcification in rat kidneys [46,47]. However, other studies suggested that OPN plays a role in stimulating the deposition and adhesion of crystals to cells in the early stages of crystallization. It is showed that calcium oxalate monohydrate crystal coating with OPN is correlated with increased adhesion tendency [48,49]. It is also reported that the kidney crystal deposition is decreased in OPN-deficient mice compared to crystal deposition in wild-type mice [50].
Conclusions
In conclusion, the results of our study have showed that catechin have preventive effects on renal calcium crystallization both in vivo and in vitro. However, the specific molecular mechanisms of catechin in renal calcium crystallization need to be further studied.
|
2017-06-26T04:15:35.410Z
|
2013-09-17T00:00:00.000
|
{
"year": 2013,
"sha1": "01a6140792a4c8d35e2a38d83fa9888d1bdcacbb",
"oa_license": "CCBY",
"oa_url": "https://bmccomplementalternmed.biomedcentral.com/track/pdf/10.1186/1472-6882-13-228",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "49bf22cf57adc207f160ceb8cb3d97624bc9a2df",
"s2fieldsofstudy": [
"Biology",
"Chemistry"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
231622360
|
pes2o/s2orc
|
v3-fos-license
|
Telemedicine for Pre-Employment Medical Examinations and Follow-Up Visits on Board Ships: A Narrative Review on the Feasibility
Background: Telemedicine has already been applied to various medical specialties for diagnosis, treatment, and follow-up visits for the general population. Telemedicine has also proven effective by providing advice, diagnosis, and treatment to seafarers during emergency medical events onboard ships. However, it has not yet been applied for pre-employment medical examinations and follow-up visits on board ships. Objective: This review aimed to assess the possibility of using telemedicine during periodic visits between one pre-employment medical examination and others on board ships, and to recommend necessary medical examination tests with screening intervals for seafarers. Methods: Various databases including PubMed, EMBASE, Scopus, CINAHL, and Cochrane Library were explored using different keywords, titles, and abstracts. Studies published between 1999 and 2019, in English, in peer-reviewed journal articles, and that are conference proceedings were considered. Finally, the studies included in this review were chosen on the basis of the eligibility criteria. Results: Out of a total of 168 studies, 85 studies were kept for further analysis after removing the duplicates. A further independent screening based on the inclusion and exclusion criteria resulted in the withdrawal of 51 studies that were not further considered for our analysis. Finally, 32 studies were left, which were critically reviewed. Out of 32 accepted studies, 10 studies demonstrated the effectiveness of the electrocardiogram (ECG) in monitoring and managing remote patients with heart failure, early diagnosis, and postoperative screening. In 15 studies, telespirometry was found to be effective in diagnosing and ruling out diseases, detecting lung abnormalities, and managing patients with chronic obstructive pulmonary disease (COPD) and asthma. Seven studies reported that telenephrology was effective, precise, accurate, and usable by non-medical personnel and that it reduced sample analysis times and procedures in laboratories. Conclusion: using new technologies such as high-speed internet, video conferencing, and digital examination, personnel are able to make the necessary tests and perform virtual medical examination on board ships with necessary training.
Introduction
Telemedicine is the delivery of medical services over distance through information and communication technologies [1]. Telemedicine applications include disease prevention and control, healthcare, health promotion, education, training, and research for health [1]. Telemedicine can be traced back to the end of the 19th century to the mid-20th century with the use of radiotelegraphy to provide medical advice to patients onboard ships [2]. The evolution versus modern telemedicine began in the 1960s, mainly spurred on by the military and space technologies, as well as by some people using readily available commercial equipment [3,4]. More recently, the applications of telemedicine are manifold are addressed to the SASN (Servizi Assistenza Sanitaria Naviganti)-Health Assistance Services for Navigation, Maritime and Civil Aviation Personnel, reference USMAF-SASN for the territory or to an authorized medical officer [30]. After approval, depending on the appointment, periodic medical check-ups are to be carried out, with most medical examinations being performed onboard ships by a doctor or other health professionals. Hence, telemedicine for periodic medical check-ups can reduce unnecessary travel, skip/shorten some process steps, reduce transport costs for regular check-ups, avoid waiting lists, and increase the quality of services by offering specialty doctors the opportunity to participate in virtual examinations.
Telemedicine has proven effective by providing advice, diagnosis, and treatment to seafarers during emergency on board ships [31]. It has already been applied to various medical specialties for diagnosis, treatment, and follow-up visits for the general population and has made it possible to overcome various constraints such as geography and the resources to provide healthcare to remote populations [32]. However, it has not yet been used for pre-employment medical examinations and follow-up visits onboard ships. The purpose of the present review was to assess the possibility of using telemedicine during periodic visits between the pre-employment medical examination and others and to propose necessary medical tests with screening intervals to be included in PEME protocols.
Materials and Methods
The review was conducted by searching the different published scientific literature, indexed in various databases, including PubMed, EMBASE, Scopus, CINAHL, and Cochrane Library. We included studies in areas of 4 medical specialties reporting medical exams/tests frequently used in telemedicine, namely, electrocardiogram (ECG), spirometry, blood glucose monitoring, and urinalysis. Different key terms for search, including "telecardiology", "tele diabetology", "telespirometry", "telenephrology", "teleconsultation", "telemedicine", and "telehealth" were used. Inclusion criteria for selected studies included studies published between 1999 and 2019, studies published in English, peer-reviewed journal articles, conference proceedings, and full-text paper. Studies published with only abstracts and not in English were excluded.
In total, 168 potentially relevant studies were selected from the above databases. Out of a total of 168 relevant studies, 85 studies remained after the duplicates were removed. A further 51 studies were excluded upon reviewing the abstract, title, and assessment of full-text as not consistent with our research questions. Finally, 32 studies were selected after a thorough review and included in this review ( Figure 1). The review analysis involved 3 independent reviewers and an expert in the event of disagreement.
The summary of results for the application of telemedicine in various medical examinations/tests of occupational medicine interest is summarized in The summary of results for the application of telemedicine in various medical examinations/tests of occupational medicine interest is summarized in Table 1 (Table 1). The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1). The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1). The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1). The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1).
Respiratory function test
Healthcare 2021, 9, x FOR PEER REVIEW The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1). The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1). Remote glycemic self-monitoring. The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1). The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1). The summary of results for the application of telemedicine in various medical inations/tests of occupational medicine interest is summarized in Table 1 (Table 1).
Electrocardiogram (ECG)
Telemedicine is not a separate medical modality but includes a growing variety of applications and services that use telephone lines, videos, e-mails, smartphones, wireless tools, and other forms of telecommunication technology. Among the wide range of medical specialties in which telemedicine has been successfully applied, cardiology has been found to be one of the most common fields of application. Through the transmission of clinical data and the electrocardiogram (ECG), telecardiology allows access to a realtime assessment (teleconsultation) without the need to travel for both the patient and the cardiologist. Telecardiology has proven to be useful in the clinical management of remote patients with real or suspected heart disease in different clinical settings. Over 20 years, several attempts have been made to try to introduce and expand telecardiology in the hospital setting, especially for the diagnosis and treatment of patients in remote locations [33]. Positive impacts of telecardiology have been demonstrated in the literature on patients with heart failure. In these patients, remote monitoring compared to traditional monitoring procedures decreased the risk of recurrence of the event [34].
A study conducted in 2010 demonstrated that the accuracy of an ECG performed remotely via wireless procedures in detecting an episode of atrial fibrillation was 93%, and the accuracy of this examination was 94% [35]. A study by Ong et al. (2016) focused on the effectiveness of remote patient monitoring (RPM) for the transition of care in adult patients with heart failure, finding no significant difference in patient readmissions for 180 days. Those who received the remote monitoring intervention had readmission rates of 50.8%, while those who did not receive the intervention had readmission rates of 49.2% [36]. Another study compared hospital readmission rates and death rates for two groups of heart failure patients, followed by telemonitoring and conventional monitoring. The types of post-treatment screening were comparable in that no statistically significant difference was found. Readmission rates were 49.3% for remote monitoring and 47.4% for routine care, while the mortality rate for remote monitoring was 11.1%, and in routine care it was 11.4% [37].
The use of portable, non-invasive tele-cardiologic screening equipment was found to be less costly for hospitals and more comfortable for patients, allowing them to remain in domestic environments [38]. According to Sohn, costs would be around 25% lower in patients with mild symptoms [39]. In addition, to diagnosis and postoperative screening, telecardiology can also be used in prevention. In 2010, an Italian project was launched aimed at using a telecardiology device to perform early diagnosis of 13,016 students aged 16 to 19 [40]. In 2016, the same author collected the results showing that 24% of the suspects had altered signals in the electrocardiogram. The conclusion was that the use of telecardiology procedures in mass screening has many advantages such as lower costs, the possibility of use in environments far from hospitals, and not requring qualified personnel [41]. Another study evaluated the effectiveness of remote monitoring through the use of various devices, including the electrocardiogram (ECG) in healthy clients, in patients at risk for cardiovascular diseases such as those with diabetes and hypertension, and in patients with a previous cardiovascular event. The healthy clients were monitored by a 12-lead ECG installed in pharmacies and connected to a telemedicine platform manned by a cardiologist 24 h a day. Between 1 January 2015 and 31 December 2017, the study involved a total of 79,898 women (mean age 52 ± 12 years) and 68,458 men (mean age 49 ± 11 years). Of all ECGs performed, approximately 8% showed electrocardiographic abnormalities inconsistent with the patient's medical history. The authors confirm that tele-cardiological screening systems can be used successfully in the prevention of cardiovascular diseases with a consequent positive impact on public health [42].
Spirometry
Spirometry, also defined as spirometric test or simply respiratory function test, is a diagnostic test that is performed using a spirometer, a computerized instrument, connected to a mouthpiece. It is a very simple, painless, and non-invasive examination. Spirometry plays an important role in the diagnosis and monitoring of chronic obstructive respiratory disease. It should be noted that relevant clinical guidelines indicate the need for widespread use of spirometry in primary care for the early diagnosis and appropriate management of chronic asthma and chronic obstructive bronchopulmonary disease (COPD). Hence, this test is of relevant importance in the health surveillance in preventive medicine for seafarers who, for working reasons, often need to face long journeys [43,44]. The possibility of remote use of spirometry through telemedicine equipment has been investigated by several authors both as a prevention and monitoring tool in patients suffering from respiratory syndrome [45]. Telespirometry is used in clinical practice to monitor asthma patients and patients with COPD living at a distance from the hospital [46,47]. These integrated systems, in patients with respiratory diseases and frequent exacerbations, can reduce both emergency room visits and the number of hospitalizations [48].
In 2009, Bonavia published an article reporting an Italian project investigating the possibility of using telespirometry in general medicine. As a part of the project, 937 family physicians exchanged data via telespirometry equipment with 56 specialist centers, visiting their patients with risk factors, persistent respiratory symptoms, or a previous diagnosis of asthma or chronic obstructive pulmonary disease for two years. About 90% of the spirometry tests (20,757 spirometry tests performed in total) met the criteria and made it possible to make a diagnosis or rule out pathologies. 40% of the spirometries made it possible to detect pulmonary abnormalities. The authors consider telespirometry to be a reliable and useful alternative in the management by general practitioners of chronic respiratory diseases. The quality of the spirometric examination is highly dependent on the skills of the technician administering the test. The pulmonology company's guidelines indicate the skills and training to be acquired to manage this exam [49]. An interesting perspective, which involves several studies, is given by the possibility of remote monitoring at home. This technology enables the self-measurement of clinical parameters/symptoms of patients at home and allows the communication between healthcare professionals and remote patients [50,51].
A study investigated the possibility of carrying out the spirometric test directly at home in total autonomy. The study involved four patients with previous chronic obstructive bronchopathy who were provided with telespirometry equipment equipped with tablets capable of supporting the patient in the examination. This system was used for 12 weeks in which patients performed several daily spirometries without any medical assistance. As a result, the large part of the spirometry (94.5%) was considered acceptable and usable by qualified personnel [52]. Another study investigated the possibility of self-administration of the spirometric examination in asthmatic patients [53]. The author equipped 86 asthmatic patients with a portable instrument for spirometric measurements at home without medical supervision, evaluating their acceptability according to the American Thoracic Society and European Respiratory Society (ATS/ERS) criteria. The author set the primary endpoint with the following criteria: correct use of the device three or more times within 7 days (±1 day) in one of the 3 weeks of the study. Of 78 patients, 67 (86%) reached the primary endpoint. Seventy-five (96%) participants used the device correctly one or more times, and 10 (13%) patients managed to use the tool every day during the 3 weeks. The authors showed that remote self-assessment using spirometry equipment is a feasible practice [53].
Adequate training in performing the spirometric examination of technicians or operators onboard the vessel capable of ensuring high-quality standards in spirometry could be fundamental to generate reliable results that the occupational physician could then assess from another clinic suitable for the examination, obviously using portable technological spirometers. The quality of spirometry tests strongly depends on adherence to international recommendations [53,54]. There are different tools today, such as those that use Android micro-control technology to measure lung function. For example, this latest technology has reported excellent results on patients who have been analyzed both with an examination performed with the traditional spirometer and with the new micro-control technology providing very similar final values, with an accuracy of high measurement and for forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC) indicating that this device, for example, could be usefully used in telemedicine and health surveillance. Furthermore, there is a remote assistance technology that uses a simple spirometer with a Bluetooth module, an ES application based on MATLAB, and a mobile app based on Android. In this case, the portable spirometer used in this study can be connected to a mobile phone via Bluetooth. This technology has been used to evaluate the chronic course of diseases such as COPD and asthma. During 6 months, 780 patients were assessed and diagnosed with an accuracy of 97.32% [55].
Blood Glucose Monitoring
Telemedicine interventions for diabetes can vary from simple reminder systems via text messages over complex web interfaces. Patients can upload their glucose levels measured with a home meter and other relevant data such as medications, eating habits, level of activity, and anamnesis. The measurement of blood glucose concentration represents an essential step in the management of diabetic disease-glycemic self-monitoring, in fact, is widely used by patients with type 1 and type 2 diabetes to verify metabolic compensation, to identify and treat episodes of hyper-and hypoglycemia and to adapt the hypoglycemic therapy to the conditions of life (nutrition, physical activity, stress, intercurrent diseases) [56]. In the field of diabetes, telemedicine is used in its various forms: remote glycemic monitoring, teleconsultation, personal medical records, telenursing, and call centers. Several studies have shown that real-time transmission of blood glucose data is achievable with evidence demonstrating improvements in terms of glycemic control [57].
Urine Analysis
Urinalysis is an important diagnostic screening test useful for diagnosing and monitoring nephrological and urological conditions. This test is also often used in general preventive screening [58]. Information technology has significantly reduced the analysis times following the collection of urine and often performed in specialized laboratories. Nowadays, many portable electronic readers are available that can analyze in real-time the urine collected on particular test strips [59]. To be effective, these devices must be economical, portable, precise, reliable, robust, powered by batteries, and usable by non-medical personnel as well [59].
A recent development consists of the use of smartphones to read and interpret the results of the test strips where urine is collected [60]. These smartphones are often combined with electronic pocket readers and reagent strips [61]. Dae-Sik Lee proposed in 2011 a mobile health platform by combining a pocket colorimetric reader with a smartphone and paper strips for urinalysis capable of analyzing glucose, proteins, bilirubin, ketones, nitrites, pH, specific gravity, erythrocytes, and leukocytes. Various tests carried out show that the device is efficient, inexpensive, and accurate [62]. The urine test strip is a device made up of strips on different distinct reactive zones allowing for the determination of specific gravity, pH, proteins, glucose, ketones, bilirubin, blood, nitrites, urobilinogen, and leukocytes in the urine. Often, the results have to be interpreted, and, in order to avoid human error, some digital scanners are able to interpret the results independently via the smartphone camera [63,64].
Legal Implications of Telemedicine
Telemedicine has significant repercussions in the delicate ethical sphere, as this different way of managing the interaction and communication between the patient and the doctor (or in general, the health workers involved) impacts a particular situation for who is in need of health care. On the other hand, on establishing the patient-doctor relationship, the safeguard of the dignity of the patient should be always safeguarded.
The legal problems related to telemedicine have not yet found standard solutions at the international level. Any request for telemedicine is considered a medical act. Hence, although many teleconsultation procedures are unique, the traditional principles of traditional doctor-patient relationships are also valid in telemedicine. Three concerns can be involved in legal performance issues [65,66]. As a result, in terms of the person who transmits the data, "informed consent must govern the relationship between the patient and other interested parties." This should include the patient's awareness of the technical aspects; the potential risks; the precautions required; and, at the same time, and the guarantee of the confidentiality of information [66]. The person who receives the data is the medical service user. Regarding the service provider, confidentiality, as well as the quality of the transmitted and received data, must be guaranteed by the service provider [66].
Recommendations
Telemedicine can be used successfully for pre-employment medical examinations and follow-up visits on board ships. This study used a narrative review to evaluate the possibility of telemedicine for periodically pre-employment medical exams and follow-up visits for workers needing to be followed aboard ships. We considered four major medical exams/tests for telemedicine applications in this review, namely, electrocardiogram (ECG), spirometry test, blood glucose monitoring, and urinalysis. As a result of telemedicine, we proposed the necessary medical examinations/tests with screening intervals to be included in the pre-employment medical examination (PEME) protocols (see Table 2). Table 2 summarizes features, medical tests, suggested periodicity, telemedicine modality, and equipment to be used for telemedicine practice in occupational medicine. Possible applications of telemedicine in the PEME, follow-up, and future visits are detailed below: 1. PEME: ECG, spirometry, measurement of vital signs (blood pressure, heart rate, and temperature), oxygen saturation assessed by a pulse oximeter, blood glucose measurement by glucometer, and urinalysis are tests that can be monitored remotely in some cases even without medical assistance. These tests are currently used for the general population in different medical specialties. We thus suggest using them both for PEME visits and/or for periodic checks of seafarers aboard ships, provided that on board there is the necessary technological resource training.
2.
Follow-up visits: Telemedical Maritime Assistance Services (TMAS) doctors can make multiple medical visits a single patient, guaranteeing appropriate evaluation standards. They can also follow the same patient for the necessary time by monitoring over time without the need for large displacements and without the need to visit inside the ships in person. The technological equipment should provide a secure and high-speed internet connection; a clinical telemedicine software that acts as a hub capable of sending the patient's vital parameters to the doctor; devices capable of monitoring the patient's body parameters; and, finally, an ad hoc training program with periodic simulations. If fitted onboard, these systems could also help a TMAS doctor make a correct diagnosis and plan adequate treatment.
3.
Future activities: the possibility of carrying out preventive medicine tests using telemedicine technologies could be considered in the future as a fundamental element for remote maritime preventive medicine practice. Specific experimental studies of devices integrated into platforms, protocols, and patient satisfaction should be conducted, possibly using targeted comparisons with traditional systems.
Conclusions
This narrative review has highlighted how telemedicine and devices for the detection of body parameters in the medical field are extensively practiced in many specialties of modern medicine. This review evaluated telemedicine's feasibility for pre-employment medical examinations (PEMEs) and follow-up visits aboard ships. It recommended the necessary medical examinations/tests such as ECG, spirometry test, blood glucose monitoring, and urinalysis with screening intervals for seafarers' onboard ships be considered in the PEME protocol. Moreover, recommendations were provided to responsible bodies, stakeholders, and researchers to implement telemedicine during the PEME and study its effectiveness. However, in many countries around the world, telemedicine is not an integral part of the health system, and thus it is legitimate to assume that the full potential of telemedicine has not yet been exploited. This review shows that with the advent of new technologies such as high-speed internet, video conferencing, and digital examinations, it is possible to report the necessary tests and perform virtual medical exams on board vessels. Telemedicine can be a fundamental element for the prevention and treatment required in health surveillance, particularly for the conditions in which reaching the patient is difficult and expensive, such as onboard ships.
|
2021-01-17T06:16:14.398Z
|
2021-01-01T00:00:00.000
|
{
"year": 2021,
"sha1": "2b9a17ce27a89bbcd1016e83473da25bb267a17d",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2227-9032/9/1/69/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "fdf6968abb65146e2919ab249ddd28f7ccaa442e",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
119155225
|
pes2o/s2orc
|
v3-fos-license
|
Suppressed dispersion for a randomly kicked quantum particle in a Dirac comb
I study a model for a massive one-dimensional particle in a singular periodic potential that is receiving kicks from a gas. The model is described by a Lindblad equation in which the Hamiltonian is a Schr\"odinger operator with a periodic $\delta$-potential and the noise has a frictionless form arising in a Brownian limit. I prove that an emergent Markov process in a semi-classical limit governs the momentum distribution in the extended-zone scheme. The main result is a central limit theorem for a time integral of the momentum process, which is closely related to the particle's position. When normalized by $t^{5/4}$, the integral process converges to a time-changed Brownian motion whose rate depends on the momentum process. The scaling $t^{5/4}$ contrasts with $t^{3/2}$, which would be expected for the case of a smooth periodic potential or for a comparable classical process. The difference is a wave effect driven by Bragg reflections that occur when the particle's momentum is kicked near the half-spaced reciprocal lattice.
Introduction
Mathematical models of a quantum particle in a periodic environment have been used to describe an electron in a metal and, more recently, an atom in an optical lattice. One topic of mathematical and physical interest is the transport behavior for the particle in the periodic environment. The periodic situation stands in contrast with the random or quasi-periodic situation, which may exhibit Anderson localization [4]. Another important topic is the study of the motion of the particle when acted upon by a static force. Zener predicted an electron in a metal would exhibit some periodic motion when a constant force was applied [41]. This behavior, called Bloch oscillations, is related to Bragg scattering and has been observed experimentally in conductor superlattices [18]. More recently, atoms in optical lattices have provided an analogous setting in which it is possible to measure Bloch oscillations with fewer noise effects [7,5].
The current article studies a suppressed dispersion effect that, like Bloch oscillations, is generated by a combination of outside forcing (in this case from a noise) and Bragg scattering in a periodic potential. My model concerns a one-dimensional massive quantum particle in a periodic singular potential that receives random momentum kicks (e.g. from a gas of light particles). The massive particle effectively does not "feel" the potential except for infrequent instances when its momentum is kicked near an element of the half-spaced reciprocal lattice of the potential. Near the lattice values, the particle's momentum has a chance of being reflected, and these reflections in momentum occur often enough to inhibit the motion of the particle. I imagine the model to describe an atom in a very singular one-dimensional optical potential. In the physics literature, the article [21] discusses Bragg reflections of atoms from a weak a one-dimensional optical potentials. The articles [8,35] report the experimental observation of Bragg scattering in atoms with lower kinetic energy through optical potentials.
My mathematical starting point for modeling the particle is a quantum Markovian dynamics generated by a Lindblad equation. In the following section, I introduce the dynamics, state the main theorems, discuss some background for the model, and make conjectures for a similar model. Section 3 contains an outline for the proof of the central limit theorem that is the main mathematical result of this article. Section 4 contains a proof that the probability density of the extended-zone scheme momentum behaves approximately as an autonomous Markov process when the Hamiltonian dynamics operates on a faster scale than the noise. Section 5 connects basic facts from the original quantum model to the limiting Markovian dynamics for the momentum process. Section 6 contains the details for the proof sketched in Sect. 3. I show that a time integral of the momentum process, when properly rescaled, converges in distribution to a variable diffusion process whose diffusion rate depends on the absolute value of the momentum.
2 Results and discussion
The model and statement of the main results
Let B 1 L 2 (R) be the space of trace class operators over the Hilbert space L 2 (R). I begin with a quantum Markovian dynamics in which the state of the particle, as expressed by a density matrix ρ λ,t ∈ B 1 L 2 (R) , evolves according to a Lindblad equation from the initial state ρ λ,0 = ρ. In this equation, P = −i d dx is the momentum operator, V is a periodic δ-potential (i.e. Dirac comb potential) with strength α > 0 and period 2π, and Ψ : B 1 L 2 (R) is a completely positive map describing the noise acting on the system and having the form where ρ ∈ B 1 L 2 (R) , X is the position operator, and j(v) ∈ L 1 (R) is the rate-density for momentum kicks of size v. I will assume the rates satisfy j(v) = j(−v) and σ = R dv j(v) v 2 < ∞. In (2.1), Ψ * (I) is the adjoint map Ψ * evaluated for the identity operator I on L 2 (R), and it happens that Ψ * (I) = R I for R = R dv j(v) in my case. Equation (2.1) describes a quantum particle in dimension one evolving in a potential V and receiving random momentum kicks v with rate-density j(v). The noise is effectively frictionless, since intuitively, the rate of momentum kicks does not depend on the current momentum of the particle. This excludes the possibility of energy relaxation in the model, and there is a linear rate of growth for the mean energy of the particle: Tr[ρ λ,t (P 2 + V )] = Tr[ρ(P 2 + V )] + σ t. (2.3) By Bloch theory, the Hamiltonian H = P 2 + V has continuous spectrum and decomposes through a fiber decomposition of the Hilbert space over the Brillouin zone φ ∈ [− 1 2 , 1 2 ) as where the Hilbert spaces H φ are canonically identified with L 2 [−π, π) , and the restriction of H to the φ-fiber is a self-adjoint operator H φ . The operators H φ have a complete set of eigenvectors ψ n,φ , n ∈ N with eigenvalues E n,φ satisfying Through the extended-zone scheme, the eigenvectors ψ n,φ can be associated with a collection of eigenkets |k Q parameterized by k ∈ R such that where the dispersion relation has the form E(k) = q 2 (k) for the anti-symmetric, increasing function q : R → R satisfying the Krönig-Penney relation cos(2πk) = cos 2πq(k) + α 2q(k) sin 2πq(k) (2.4) for k ∈ R − 1 2 Z and q( n 2 ) = n 2 for n ∈ Z (see (4.1) for the corresponding Bloch functions in the position representation). The Bloch structure and my conventions are discussed in Appendix B. The dispersion relation essentially has the form E(k) ≈ |k| 2 + α 2π for |k| ≫ 1 except for values of k in small neighborhoods around the lattice 1 2 Z, where E(k) makes jumps E( n 2 +) − E( n 2 −) = 0. The kets |k Q also have discontinuities at values k ∈ 1 2 Z: lim ǫ→0 |k − ǫ Q = |k + ǫ Q . The first result is concerned with the limiting behavior as λ → 0 of the diagonal distributions in the extended-zone scheme representation: Q k|ρ λ,t |k Q = D λ,t (k). I show, in a sense defined below, that D λ,t converges for small λ to the solution D t of a classical Markov process where the rates J(k, k ′ ) are determined by the rates j(v) and values κ v (k, n) (defined below) through the formula The values κ v (k, n) ∈ C arise as coefficients in the formula n∈Z κ v (k, n)|k + v + n Q = e ivX |k Q , k, v ∈ R. (2.7) The fact that e ivX |k Q is a combination of the |k + v + n Q , n ∈ Z is a consequence of the fiber decomposition. By the unitarity of e ivX , the coefficients satisfy n∈Z |κ v (k, n)| 2 = 1, and the process has a constant escape rate: R = R dk ′ J(k ′ , k).
Also, for t < λ, the norm of the difference is bounded by a multiple of t.
Let K r be the Markov process satisfying the master equation (2.5) and define the integral functional Y t = t 0 drK r . My main result concerns the limiting distributional behavior for the processes (σt) − 1 2 |K st | and t − 5 4 Y st , s ∈ [0, 1] for t ≫ 1. I will make the following technical assumptions on j(v): List of rate assumptions 2.2. There is a µ > 0 such that 1. R dv j(v) e a |v| < µ for some a > 0, The theorem below states that the processes |t − 1 2 K st | converge in law as t → ∞ to the absolute value of a Brownian motion and t − 5 4 Y st converges to a time-changed Brownian motion whose rate of diffusion emerges as the limit law of ν −1 |t − 1 2 K st | 3 for ν := Rα. It is clear from the above statement that the process t − 1 2 K st itself does not behave as a Brownian motion; otherwise the appropriate scaling for Y st would be t 3 2 , and the limiting process would be differentiable rather than diffusive.
Theorem 2.3 (Main result).
Assume j(v) satisfies List 2.2. Let K t be the Markov process whose probability densities q t satisfy (2.5), q 0 have finite second moment, and Y t be the time integral of K r up to time t. As t → ∞, there is convergence in law with respect to the Skorokhod metric where B s and B ′ s are independent standard Brownian motions.
Further discussion and background
2.2.1 The Lindblad dynamics, the noise, and the Hamiltonian.
Introductory material on Lindblad equations can be found in [3]. Some basic mathematical questions regarding existence and uniqueness of solutions to Lindblad equations with unbounded generators are not completely understood except for specific classes such as those for which the generator is translation covariant [27]. Because the Hamiltonian part of the Lindblad equation (2.1) is unbounded, the mathematical definition of a solution to the Lindblad equation is less direct than the bounded case. I discuss the rigorous definition for the dynamics and related technical issues in Appendix A. In the case discussed here, these issues are not interesting or challenging, since the noise term Ψ : B 1 L 2 (R) is bounded and Ψ * (I) = RI commutes with the Hamiltonian. Consider a Lévy process with density j(v), and let t 1 , . . . , t n and v 1 , . . . , v n be the Poisson times and increments of the Lévy process up to time t. The state for the particle ρ λ,t = Φ λ,t (ρ) at time t is equal to Φ λ,t (ρ) = E U λ,t (ξ) ρ U * λ,t (ξ) , (2.8) where the expectation is with respect to the law of the Lévy process, ξ = (v 1 , t 1 ; v 2 , t 2 ; . . . ) ∈ (R×R + ) ∞ is the full sequence of random events, and the unitary operator U λ,t (ξ) : L 2 (R) is defined by the product U λ,t (ξ) := e − i(t−tn) λ H e ivnX · · · e − i(t 2 −t 1 ) λ H e iv 1 X e − it 1 λ H . (2.9) The construction of the maps Φ λ,t thus only depends on the existence of the unitary groups e −irH , r ∈ R. The equation (2.8) implies the trace for the state is preserved such that Tr[ρ λ,t ] = Tr[ρ] = 1, since the expression on the right is a convex combination of unitary conjugations of ρ.
A noise of the type appearing in (2.1) was originally introduced as a phenomenological model for the study of wave collapse in quantum mechanics [24]. It was later derived in [22] starting with a heuristic scattering analysis that was meant to model an interaction of a test particle with a gas in the limit that the test particle has much greater mass than the gas particles. This scattering analysis was clarified in [29], which yielded a minor correction by a unitless multiplicative factor in the final expression for the jump rates j(v). Also, the article [26] contains a mathematical derivation for a noise of the form (2.1) through a singular coupling limit of a simple system-reservoir Hamiltonian dynamics. The noise model has been discussed in relation to experimental frameworks in matterwave optics [2,30] and appears in other discussions of decoherence [40]. See [39,Sec.7.1] for the connection of the noise with a quantum linear Boltzmann equation in a large mass limit. A similar frictionless noise with some spatial dependence recently appeared in [33] to model the dampening of Bloch oscillations for an atom in an optical lattice.
It is clear from the mean energy growth (2.3) that the model for the noise is transient in nature. The classical analog of the noise map appearing in the Lindblad equation (2.1) is given by the map for joint position-momentum densities Υ ∈ L 1 (R 2 ). The association of (2.10) with the quantum noise can be justified by looking at the Wigner representation W ρ (x, p) of ρ. Of course, most quantum noises are not so readily identifiable with classical analogs. Equation (2.10) describes the momentum as receiving random kicks v with rate-density j(v). Based on this evidence, the momentum undergoes an unbiased random walk, and the ensuing stochastic acceleration explains the mean energy growth in (2.3).
The Hamiltonian H = P 2 + α n∈Z δ(X − 2πn), α > 0 is defined as a particular self-adjoint extension of the symmetric operator − d 2 dx 2 with domain consisting of all functions f ∈ H 2,2 (R) ∩ {g | g(2πn) = 0, n ∈ Z} having two weak derivatives and taking the value 0 on the lattice 2πZ. The domain of the self-adjoint extension is the space of functions f ∈ H 2,1 (R) ∩ H 2,2 (R − 2πZ) that have one weak derivative in the domain R and two weak derivatives in R − 2πZ, and that satisfy The Dirac comb is a limiting case of the Krönig-Penney model [34], which is a periodic Schrödinger equation in which the potential has the form V (x) =V n∈Z 1 [2πn−a, 2πn+a] for a < 2π. The limit connecting them isV → ∞ with 2V a = α. The Krönig-Penney model has been used to model the transport of electrons through a crystal. One computational advantage of these models is that there are closed equations determining the spectral values and the form of the eigenkets. Both the periodic δ-potential and the Krönig-Penney model are discussed in [1]. Some general theory regarding the structure of periodic Schrödinger equations can be found in [16,38,6]. The articles [23,9,15] contain recent results on the dispersion of wave packets evolving according a periodic Schrödinger equation with a smooth potential.
Bragg reflections
For a high momentum quantum particle in a periodic potential, the dominant behavior is simply transmission over the potential. However, there is a lattice 1 2 Z of wave frequencies around which the particle is likely to be reflected by the potential. For my purpose, the difference between a smooth periodic potential and the Dirac comb is the proximity a high momentum particle must have to a lattice momentum to experience reflections. These zones are wider for the Dirac comb, and it is possible for the test particle to score accidental reflections in the process of colliding with the gas. A more immediate indication of a contrast between the Dirac comb and a smooth period potential is in the size of the jumps in the dispersion relation E(k) at momenta k ∈ 1 2 Z (as pictured in Figure 1). Unlike smooth potentials, where the energy gaps vanish as k → ∞, the gaps for the periodic δ-potential approach the constant value α π . Due to the spatial translation symmetry of the potential by 2π, the kets |k Q can be written as discrete combinations of the momentum kets |k + n for n ∈ Z: |k Q = n∈Z η(k, n)|k + n , (2.11) where n∈Z |η(k, n)| 2 = 1. For |k| ≫ 1, the n = 0 term dominates the sum except when k is close to a lattice point n 2 ∈ 1 2 Z, in which case the −n term becomes nonnegligible. When k ≈ 1 2 n, the value k − n ≈ −k is approximately a reflection of the momentum. A high-momentum plane wave tuned near a lattice frequency will be driven by the Hamiltonian evolution to oscillate though a cycle of quantum superpositions between the original wave and the reflected wave. These are called Pendellösung oscillations, and the corresponding wave velocities E ′ (k) are small for momenta k close enough to an element in 1 2 Z to exhibit the Pendellösung oscillations (see also in Figure-1). A plane wave with a high momentum sufficiently away from any lattice value will transmit nearly freely through the potential with velocity E ′ (k) ≈ 2k. I will refer to the regions near the lattice momenta in which nonnegligible reflection occurs as the reflection bands (see [21]). For the Dirac comb potential, the widths of the reflection bands scale as ∝ n −1 for n 2 ∈ 1 2 Z with |n| ≫ 1.
2.2.3
The small λ limit The regime of λ ≪ 1 in (2.1) should be considered as a semi-classical regime in which phase oscillations generated by the Hamiltonian term occur on a faster time scale than the mean time between collisions with the gas. As it appears in (2.1), the parameter λ takes the place of 2M , where M is the mass of the test particle. Since the noise arises in a limit in which M is large, the value −1 must be even "larger" to make λ ≪ 1. A more honest comparison of relevant scales requires that I include more physical parameters in the model such as the period and strength of the δ-potential, and I do this for a slightly richer noise model in Sect. 2.3. Without the noise, the dynamical evolution generated by the Hamiltonian in the extended-zone scheme representation is for ρ ∈ B 1 L 2 (R) . The noise map Ψ operates in the momentum representation as Since the ket |k Q is a combination (2.11) of the kets |k + n for n ∈ Z, the equations (2.12) and (2.13) imply that only the values Q k ′ |ρ|k Q with k ′ − k ∈ Z interact dynamically with the diagonal values Q k|ρ|k Q . The relevant phase velocities are λ −1 E(k ′ )−E(k) for k ′ −k ∈ Z. Due to the non-vanishing energy band gaps for the Dirac comb potential, there is a non-zero minimum λ −1 s min for the phase speeds, where (2.14) These phase oscillations thus have period proportional to λ, which is small compared to the mean time R −1 between momentum kicks. This provides the mechanism through which the diagonals Q k|ρ λ,t |k Q tend to an autonomous evolution as λ → 0. The λ → 0 limit connecting the diagonal of ρ λ,t in the basis of kets diagonalizing P 2 + V to a Markovian dynamics is analogous to a Freidlin-Wentzell limit [20,Ch.8] for a classical Hamiltonian flow perturbed by a weak noise. In the Freidlin-Wentzell limit, a Markovian dynamics emerges on the energy graph in a scaling limit combining a noise of strength λ ≪ 1 and long time intervals ∝ λ −1 . The "energy graph" is the collection of connected level curves for the Hamiltonian. My dynamics fits the Freidlin-Wentzell limit description with a simple change of time variable s := λ t: The connected level curves of the Hamiltonian and the kets |k Q have parallel roles in imprinting the Hamiltonian structure on the limiting Markovian dynamics. A similar study for a time integral of a momentum-related quantity after a Freidlin-Wentzell limit has previously been performed for a classical model in [25]. This model is a Markovian dynamics for a particle in a periodic potential with a white noise, and the authors have shown that the spatial diffusion and Freidlin-Wentzell limits commute. Theorem 2.3 is a statement only about the classical process K t and its time integral Y t = t 0 drK r . This result has no direct consequence for the original model, and in particular, I have not proven that the position distribution d λ,t (x) := x|ρ λ,t |x itself exhibits subdiffusion for λ ≪ 1. However, I believe the suppressed dispersion effect holds in the original model for any finite λ. In other words, d λ,t (x) has spread ∝ t 5 4 when the Dirac comb is present rather than the scaling ∝ t 3 2 that holds otherwise (see Appendix C). The effect depends on the occasional development of Hamiltonian-generated quantum superpositions between momenta in the reflection bands and their reflected counterparts. As the particle is kicked out of a reflection band, the quantum superposition collapses into a classical superposition.
Energy submartingales
I chose the Dirac comb for this study as opposed to some other singular periodic potential because there are convenient closed expressions for the eigenkets |k Q in the position and momentum representations. This is helpful for estimating the coefficients κ v (k, n) appearing in the definition (2.6) of the jump rates for the Markov process K t . Nevertheless, it is easier to perceive certain critical features of the classical process K t by taking a step back into the original quantum framework than through the closed formulas. For instance, E 1 2 (K t ) is a submartingale. It follows that E(K t ) is also a submartingale, although it can be additionally learned from the quantum setting that the increasing part of its Doob-Meyer decomposition is σt. That in not surprising given the mean energy grown (2.3). The process E 1 2 (K t ) is useful for showing Thm. 2.3, since |K t | ≈ E 1 2 (K t ) when |K t | ≫ 1 by the approximately parabolic shape of the dispersion relation, and because the martingale structure allows E 1 2 (K t ) to be treated (see Thm. 3.1). An equally important application of the submartingale structure is to show that the process |K r | spends most of the time interval r ∈ [0, t] with large values on the order t 1 2 for which the Bragg scattering is dominated by only transmitted and reflected waves.
The proof that E 1 2 (K t ) is a submartingale is based on the Heisenberg representation of the equation (2.8). The stochastic operator-valued process is a positive submartingale in the sense that for all f ∈ D(H 1 2 ) ⊂ L 2 (R), the process f |H 1 2 λ,t (ξ)f is a positive submartingale. Section 5 contains the analysis of the various relevant operator martingales.
Heuristics for the limit theorem
The Kolmogorov equation (2.5) determines a pseudo-Poisson process K r whose jump times are determined by a Poisson clock with rate R. The jumps are a sum of two contributions: one coming directly from a particle collision, and another being a lattice-valued Bragg scattering. More precisely, a jump w from the starting momentum k ∈ R is equal to w = v + n, where the component v ∈ R has density R −1 j(v) and the component n ∈ Z has conditional probabilities |κ v (k, n)| 2 when given v and k. I refer to these as the Lévy and lattice components of the jump. When |k| ≫ 1, most of the lattice jump probabilities |κ v (k, n)| 2 are negligible except for n = 0 and a few values corresponding to reflections in momentum. The high-energy behavior is dominant, since by earlier discussion, |K r | will typically spend most of a time interval r ∈ [0, t] with |K r | ∝ t 1 2 ≫ 1. An idealized picture emerges in which the process makes a Lévy jump v from k with an additional optional jump −n ∈ Z for which occurs with probability If this extra jump occurs, the resulting value k + v − n ≈ −k − v is approximately a reflection of the value it would have had otherwise. The probability (2.15) of a momentum reflection decays when |θ| ≫ |n| −1 . Hence, when |k+v| ≫ 1, the value k + v must land near a lattice value 1 2 Z in order to have a good chance of reflection. The integral serves as an effective reflection probability if the particle is dropped randomly in the cell centered around n 2 . The factor of 2 in (2.16) normalizes the integration. This suggests the number of reflection times N r is approximately a Poisson process with a rate depending on K r as ≈ Rα|K r | −1 = ν|K r | −1 (since the Lévy jumps occur with rate R). In other words, the Poisson rate of reflections is inversely proportional to the absolute value of the momentum. Over a time interval where |K r | ∝ t 1 2 , the average time between reflections will be ∝ t 1 2 . Hence, if |K r | behaves as the absolute value of a random walk, the number of reflections over a time interval [0, t] will be (at least) on the order of t 1 2 , and t − 1 2 K st itself can not have a limiting distribution. However, it is reasonable to expect a limit theorem for the process Y r , since the sign-flipping of K r is smoothed by the time integration. Roughly speaking, Y t can be written where τ n are the reflection times. By the considerations above, the integrand |t − 1 2 K r | and normalized interval t − 1 2 (τ n+1 − τ n ) will both be O (1). Since the number N t of terms in the sum is on the order O (t 1 2 ), a scaling factor of N 1 2 t ∝ t 1 4 is appropriate if I expect central limit theorem-type cancellation among the summands in (2.17). Hence, it is reasonable to expect t − 5 4 Y st , s ∈ [0, 1] to have a nontrivial diffusive limit. The diffusion rate will be proportional to t − 1 2 K st 3 with two powers of t − 1 2 K st coming from the integrand (2.17), and another factor coming from the less frequent reflections (and thus diminished cancellation) that occur when the momentum is large.
2.2.6
Related limit theorems The article [12] is a classical analog of the current work in which the periodic potential is continuous and the noise is frictionless. There, the rescaled momentum process t − 1 2 K st , s ∈ [0, 1] converges in distribution to a Brownian motion, and thus the position process simply converges to the integral of a Brownian motion. Also for a classical model, the articles [13,10] work to control and characterize a periodic potential as a perturbative contribution to a dissipative dynamics driven by a linear Boltzmann equation in the limit of large mass for the test particle. The noise there is analogous to the quantum noise discussed in Sect. 2.3.
The study of the limit law of t − 5 4 st 0 drK r for large t ∈ R + fits under the general category of central limit theory for integral functionals of Markov processes. It is not covered by previous results that I am aware of, since the process K t is null-recurrent and systematically makes large jumps in the form of sign-flips from arbitrarily high values in phase space (see [28] for martingale limit theory relevant to a broad class of null-recurrent situations). A simplified version of the problem is given by the following: let the Markov process K ′ t make jumps at times determined by a Poisson clock with rate R and transition density T (k ′ , k) from k to k ′ given by In words, the process first makes a jump from k to k + v with density R −1 j(v), and if |k + v − n 2 | ≤ α |n| for some n ∈ Z − {0}, then the sign either flips or remains the same with equal probability 1/2. Note that the process |K ′ t | has the same law as the absolute value of a Lévy process with rates j(v). The limit statement of Thm. 2.3 holds with K t replaced by K ′ t . If the reflection bands around the lattice values n 2 ∈ 1 2 Z in the simplified model above are replaced by bands with diameter α| n 2 | −2ϑ for 0 < ϑ < 1, then the limit law will be The Krönig-Penney model lies at the boundary ϑ = 1. I conjecture that if the periodic δ-potential is replaced by a continuous potential, the limiting behavior will agree with the classical case. The Krönig-Penney model will have an intermediary behavior in which t − 1 2 K st converges in law as t → ∞, but the limiting process is not a Brownian motion due to a number of random reflections over the interval s ∈ [0, 1]. The limit of the time integral process will be differentiable rather than diffusive.
Further mathematical questions
The first question is whether the original Lindblad model (2.1) actually exhibits suppressed spatial dispersion for a fixed value of λ. Within the current program of passing through a Freidlin-Wentzell limit, there is the mathematical challenge of beginning with a more sophisticated noise that generates energy relaxation for the test particle (see Sect. 2.3). Also, it would be interesting to compare the Dirac comb with other singular periodic potentials and to see how the situation changes in higher dimensions.
Analogous conjectures for a dissipative model
In this section, I introduce a related quantum Markovian dynamics complex enough to include energy relaxation. This material is presented with the intent to broaden the reader's perspective on the original model. In other words, this is a continuation of the discussion in the last section and does not concern the mathematical results of this paper. The model is a one-dimensional version of the quantum linear Boltzmann dynamics discussed in the review [39], which models a test particle interacting with a dilute gas of distinguishable particles. The one-dimensional case certainly can not be derived from first principles, since even the classical one-dimensional linear Boltzmann equation does not arise in a low density limit from a microscopic Hamiltonian model for a test particle interacting with a gas. Analogous mathematical objects in this section to those previously defined will be denoted with a tilde, and the meaning of symbols introduced here will reset in future sections.
Let the state of the particle at time t ∈ R + be given by a density matrixρ t ∈ B 1 L 2 (R) whose evolution is determined by the Lindblad equation whereρ 0 = ρ, and the Hamiltonian H and the completely positive mapΨ are defined below. The Hamiltonian H is a Schrödinger operator with Dirac comb potential where M is the mass of the test particle, a is the spatial period, and α > 0 is the strength of the Dirac comb. Let η be the spatial density of the gas, m be the mass of a single gas particle, and R(p rel ) be the reflection coefficient determined by the interaction potential between the test particle and a single gas particle. For the noise mapΨ, I takê where L q (P ) is the multiplication operator in the momentum representation with function The operator Ψ * (I) is also a function of the momentum operator given bŷ The function E(p) serves as an escape rate for getting kicked out of the momentum p. I will assume the interaction between the test particle and a reservoir particle is a "hard-point" interaction (i.e. an infinite strength δ-interaction). For the hard-point case, the reflection coefficient is R(p) 2 = 1. When the Dirac comb is set to zero, the distribution in momentum p|ρ λ,t |p converges exponentially fast in L 1 -norm to a Gaussian of width ( M β ) 1 2 . This is simple to prove, since it can be reduced to showing exponential ergodicity for a classical Kolmogorov equation. I have discussed exponential dissipation in the article [11] for a three-dimensional case (hard-sphere interaction) for the purpose of studying diffusion. The discussion of the HamiltonianĤ is the same as before except for the inclusion of physical constants. The Hamiltonian has a basis of kets |p Q with energies given byÊ(p) = 1 2M q 2 (p) for the anti-symmetric, increasing function q : R → R determined by the relation and q( nπ a ) = nπ a for n ∈ Z. The dispersion relationÊ(p) has jumps g n = 0 at the values p = nπ a for n ∈ Z, which approach 2α a as n goes to infinity. I will consider the model (2.18) in a limit in which the constants M, , α, a and a time variable s ∈ [0, t 0 ] scale with a single parameter 0 < λ ≪ 1 as follows for fixed constants m, h, α 0 , a 0 , t 0 , η, and an exponent ̺ > 0. I will add "λ" as a subscript to the solutionρ λ,t of the Lindblad equation (2.18) and the escape rate E λ (p) to indicate the parameter dependence. LetD λ,t (p) = Q p|ρ λ,t |p Q be the probability density for the momentum given by the diagonal of the density matrixρ λ,t in the extended-zone scheme representation. DefineD λ,t ∈ L 1 (R) to be the solution of the master equation withD λ,o (p) = Q p|ρ|p Q , where the jump ratesĴ λ (p, p ′ ) are defined below. Letκ λ,q (p, m) ∈ C be the coefficients in the equation n∈Zκ λ,q (p, n)|p + q + n Q = e i qX µ 2 −1 (1 + λ)q + λP The ratesĴ λ (p, p ′ ) have the form (2.23) I will denote the Markovian process whose densities obey the Kolmogorov equation (2.21) asK t , although the reader should note that the process has units of momentum rather than wave number. The following conjecture is analogous to Thm. 2.1. The exponent ̺ > 0 from (2.20) will only appear in the error of the following theorem.
Conjecture 2.5. LetK t be a Markov process whose probability densities obey the master equation (2.21) for a fixed λ > 0. Define the time integral processŶ t = 1 m t 0 drK r . In the limit λ → 0, there is convergence in law with respect to the Skorokhod metric where P t is the Ornstein-Uhlenbeck process (2.24), and B ′ t is standard Brownian motion independent of P t .
If the Dirac comb is set to zero, then there is the standard limit result forŶ t given by the convergence in law The Markov processK t without the Dirac comb has the same jump rates as the classical linear Boltzmann equation studied in [13]. The case α > 0 is subdiffusive, since the spread in position is on the order λ − 3 8 rather than λ − 1 2 for λ ≪ 1.
Physical characteristics of the scaling regime
The following lists the essential characteristics for the regime given by (2.20). The same mathematical mechanisms outlined in Sect. 2.2.5 should apply for this model, so I focus on a qualitative comparison of the relevant physical scales.
The Brownian limit
The scaling of the mass ratio as λ = m M while considering the dynamics over a time interval [0, t 0 λ ] growing proportionally to λ −1 is the standard regime for a Brownian limit. Since the temperature β −1 is fixed, the typical speed for a single particle from the reservoir and the test particle are ( 1 mβ ) 1 2 and λ 1 2 ( 1 mβ ) 1 2 , respectively. Hence, the reservoir particles are moving faster than the test particle by a factor λ − 1 2 ≫ 1. On the other hand, the momentum is ( m β ) 1 2 for a single gas particle and λ − 1 2 ( m β ) 1 2 for the test particle. Individual collisions with gas particles impart momenta that are much smaller than the typical momentum of the test particle.
Frequency of phase oscillations versus collisions
As I described in Section 2.2.3, the autonomous evolution arising for the densities in the extendedzone scheme representation depend on the noise operating on a comparatively slow scale to the Hamiltonian dynamics. More precisely, certain phase oscillations driven by the Hamiltonian occur on a smaller time scale than the mean time between collisions. I characterize the relevant Hamiltonian-driven phase cancellations with the frequency lim inf for λ ≪ 1. I associate the frequency of collisions with the escape rates E λ (p) for |p| on the order ( M β ) (2.25) Therefore, the phase oscillations occur on a shorter time scale than the collisions for small enough λ.
Kinetic energy outweighs potential energy
The kinetic energy of the test particle is typically larger than the momentum stored in the potential by a factor of λ −1 . For this comparison, I associate the typical potential energy with the strength α of the δ-potential divided by the period length: α a = α 0 a 0 λ. This is the potential energy, for instance, in a spatial wave φ(x) = (2L) − 1 2 1 [−L,L] (x) in the limit L → ∞. The mean kinetic energy is simply β −1 .
Reciprocal lattice momenta and the reflection bands
The half-spaced reciprocal lattice of momenta are multiples of π a = πh a 0 . This is on the same order in λ as the typical momentum transfers for collisions ≈ 2( m β ) 1 2 , and so the test particle's momentum has a chance of being kicked out of a given Bloch cell after several collisions. The reflection band around a lattice momentum π a n for n ∈ Z, where nonnegligible probabilities for Bragg reflections may be found, has a width of approximately 2M α |n| = 2mα 0 h|n| for high enough n so that 2πh a 0 ≫ mα 0 2h|n| . By a similar idea as in (2.16), the probability of a reflection when the particle's momentum is randomly dropped in the interval [ π (2n−1) The frequency of Bragg reflections is equal to the frequency of collisions multiplied by a local averaged probability for reflection after a collision, which depends on the current momentum p. Multiplying (2.25) with (2.26), the effective frequency of reflections when the test particle has momentum |p| ≫ ( M α a ) 1 2 is 32m βπ Since the momentum will typically be found on the scale |p| ≈ λ − 1 2 ( m β ) 1 2 , a number of reflections on the order of λ − 1 2 will occur.
3
Overview for proof of Theorem 2.3 In this section, I will state the results that enter directly into the proof of Thm. 2.3, and then proceed with a presentation of the proof assuming those results. Thus, this section concerns only the classical Markovian process K r whose densities obey the Kolmogorov equation (2.5).
Recall that E r is the square root of the energy at time r ∈ R + : E 1 2 (K r ). In Sect. 5, I show that E r is a submartingale. The following theorem states a central limit theorem for a rescaled version of E r . Since E 1 2 (k) − |k| is a bounded function, the convergence of t − 1 2 E st , s ∈ [0, 1] to the absolute value of a Brownian motion as t → ∞ is equivalent to the same statement for t − 1 2 |K st |. Thus, the first component for the convergence stated in Thm. 2.3 follows directly from Thm. 3.1.
Theorem 3.1. In the limit t → ∞, the processes t − 1 2 E st , s ∈ [0, 1] converge in law to the absolute value of a Brownian motion with diffusion constant σ. Moreover, the martingale and predictable components M r , A r in the Doob-Meyer decomposition for E r have convergence in law for large t given by where B is a standard Brownian motion. The convergences are with respect to the uniform metric.
Theorem 3.1 only characterizes the behavior for quantities that depend on the absolute value of the momentum process K r , and thus the sign-flipping does not play a role. The concept of "sign-flips" is only meaningful when the particle has a high momentum from which a Lévy jump to a momentum with the opposite sign would be unlikely. It is useful to define a series of stopping times that parse the time interval [0, t] into a series of excursions from the low momentum region. The following is a list of rough definitions for the notations related to sign-flipping and the time-integral process Y r = r 0 dvK v . More precise definitions are given below, although the details for the definitions are not strictly necessary to understand the structure of the argument in the proof of Thm. 2.3 at the end of this section. The technical definition for the sign-flip times will not be quite adapted to the original filtration F r . Martingale with respect to F (t) Martingale with respect to F (t) s approximating Y (t) s as t ≫ 1 Let S : R → {±1} be the sign function. A sign-flip is said to occur at a Poisson time t n if S(K tn ) = S(K t n+1 ) and there are an odd number m of sign changes leading up to t n : S(K t n−r ) = −S(K t n−r+1 ) for r ∈ [1, m] and S(K t n−m−1 ) = S(K t n−m ). This definition avoids counting double-flips, which occur frequently in the dynamics and would distort the counting. The sign-flips are not hitting times, since information from the following Poisson time is required to identify them. Related matters are discussed in the beginning of Sect. 6.2. To define the time intervals where sign-flips will be counted, let the stopping times ς j , ̟ j be given by ς 0 = ̟ 1 = 0 and for some 0 < ι ≪ 1. The intervals [̟ j , ς j ) are the incursions into low momentum for the process K r . Let Υ r ∈ N be the number of excursions begun by time r: max{j ∈ N | r ≥ ς j }. The times τ m and increments ∆τ m are defined inductively for m ∈ N such that • ∆τ m is the waiting time after τ m such that either a sign-flip occurs or |K r | jumps out of • τ m+1 = τ m + ∆τ m when τ m + ∆τ m occurs during an excursion, and • τ m+1 = ς j when τ m < ̟ j and τ m + ∆τ m ≥ ̟ j .
In most cases, τ m are sign-flips and ∆τ m = τ m+1 − τ m . Let N r be the number of τ m 's to have occurred up to time r. I denote the standard filtration generated by the process K r with F r = σ K s : 0 ≤ s ≤ r . Let F r be the σ-algebra given by When r ∈ [τ m , τ m + ∆τ m ) for some m, the σ-algebra F r includes knowledge of the time τ m + ∆τ m and all information about the process K r up to time τ m + ∆τ m . Since the sign-flip times are not necessarily adapted by the remark above, F r usually contains some information from the Poisson time following τ m + ∆τ m to verify that the momentum does not change sign again. For F (t) The process E r in the statement of Lem. 3.3 can be equivalently replaced by |K r |. The lemmas below all have the purpose of placing my convergence questions within the context of martingale theory. In particular, Part (2) of Lem. 3.3 is to establish independence for the copies of Brownian motion B and B ′ in the statement of Thm. 2.3. I will postulate the limiting law in a different way in the proof of Thm. 2.3. Lemma 3.2 (Martingale approximations). In the limit t → ∞, there are the following convergences in probability: Lemma 3.3 (Quadratic variation approximations). In the limit t → ∞, there are the following convergences in probability: I will make frequent use of the reference book [31] in the proof below.
Proof of Thm. 2.3. All convergence in law will be with respect to the Skorokhod metric. I will show that there is convergence in law as t → ∞ s is adapted to the filtration F Consider a sequence r n ∈ R + such that Q (rn) s converges in law as n → ∞ to a limit The relationships between the first, second, and fourth components are determined by the considerations above and Thm. 3.1. By definition of C-tightness for Q (t) , t ∈ R + , the third component m has continuous trajectories. The second and fourth components are explicitly determined by the copy of Brownian motion B, so I will focus on determining the joint law of (B, m).
is a sequence of martingales and the family of random variables h 4 The Freidlin-Wentzell/semi-classical limit In this section, I prove Thm. 2.1. First, I present Lem. 4.1, which collects some technical estimates on the probabilities |κ v (k, n)| 2 appearing in the jump rates (2.6) for the limiting momentum process. The estimates pertain to |k| large and include a technical restriction that the jump v is not too large (|v| ≤ 1 2 |k|). This is an important regime in future sections also, since the particle will stochastically accelerate to high momentum values, and the jumps v will be much smaller by the moment assumption (1) of List 2.2 on the jump rates j(v). There are only a few possible values of n for which |κ v (k, n)| 2 is non-vanishing when |k|, |k + v| ≫ 1.
As mentioned in the introduction and is discussed further in Appendix B, the Hilbert space L 2 (R) for the test particle has a canonical decomposition into a direct integral 2 ) of copies of L 2 [−π, π) that are invariant under the Hamiltonian dynamics. It is useful to relate the kets |k Q with their associated representations in L 2 [−π, π) . For k ∈ R − 1 2 Z, the kets |k Q are identified with Bloch functions ψ k ∈ L 2 [−π, π) given by where N k > 0 is a normalization constant, and q : R → R is determined by (2.4). The form of the Bloch functions (4.1) can be found in [1, Sec.III.2.3] (under the standard quasimomentum and energy band labeling). The Bloch functions formally satisfy x|k Q = ψ k (x), and ψ k is an eigenvector with eigenvalue E(k) for the fiber Hamiltonian H φ with k = φ mod 1. Analogously, the momentum kets |k The following notations are designed for a description of the reflection bands around the lattice momenta 1 2 Z. Let Θ : R → [− 1 4 , 1 4 ) and n : R → Z be defined through Note that the variables Θ(k) and n(k) are not the quasimomentum and energy band, respectively. Define the set I(k, v) ⊂ Z to include the integers 0, −n(k), −n(k + v) and −n(k + v) + n(k). Also , which is scaled to characterize the limiting local profiles of reflection probabilities for high momenta k near lattice values 1 2 Z. In the following lemma, S : R → {±1} is the sign function, and dist n, I(k, v) is the smallest distance between n ∈ Z and an element in the set I(k, v).
There exists a c > 0 such that for all |k| large enough and |v| ≤ 1 2 |k|, the following inequalities hold: 3. When n(k) = n(k + v), where I have used S(k + v − n(k)) = −S(k) and S(k) = S(k + v) by the restriction |v| ≤ 1 2 |k|. The minus sign on the fourth term on the right side appears because k − n(k) has the opposite sign of k. It follows that which will complete the proof once (4.3) is established. Now I prove (4.3) by examining ψ k in the Bloch representation (4.1). Note that (4.3) is a little more precise than is strictly required to prove Part (1), but it will be useful later. Computing the coefficients η(k, n) yields It follows from (2.4) that for large enough k, the inequality holds: |q(k)− k| ≤ α π|k| . The normalization N k can be written as All terms from the sum have been absorbed into the error O (|k| −2 ) except for n = 0, −n(k). The approximation above is due to the equality q(k) for |k| large enough so that |q(k) − k| ≤ 1 4 . When multiplied by e 2πi(q(k)−k) − 1 , and the special terms η(k, 0) and η k, −n(k) have approximations To complete the proof of (4.3), I require that the top and bottom expressions on the right side are , respectively. By the definitions of Θ(k), n(k), and β(k), I can write Define the variable γ 0 (k) = n 2 (q( n 2 + 2β(k) n ) − n 2 ) for n = n(k). The Krönig-Penney relation (2.4) to second-order for large |k| yields that γ 0 (k) satisfies This gives the following asymptotics for γ 0 (k): where for the case S(k)Θ(k) > 0, I have used the identity Therefore, combining (4.9) with (4.8), the coefficient η k, −n(k) is equal to S kΘ(k) r Part (2): By converting ψ k to the momentum basis, operating with e ivX , and translating back to the Bloch basis ( ψ k+v+n ) n∈Z , the coefficients κ v (k, n) can be written as From (4.6), I have the inequality where h := sup k∈R q 2 (p)−p 2 and the first inequality follows by using the lower bound 4q 2 (p)|e 2πi(q(p)−p) −1| 2 for N p (i.e. from the single term n = 0 from its sum). Recall that for large enough |p|, then |q(p) − p| ≤ α π|p| . Roughly speaking, I will find the inequality (4.11) useful when |p| is large enough so |q(p) − p| ≤ 1 4 and r is not 0 or −n(p). In that case, the terms in the denominator have the lower bounds q(p) + p + r , q(p) − p − r ≥ 1 4 . When (4.11) is not of use, I still have the trivial bound |η(p, r)| ≤ 1.
For notational ease, I will restrict to the case n(k) = n(k + v) in this proof. Let me first focus on the values n such that α I will use the inequality (4.11) to bound the sum of the terms m from (4.10) with the special values m = 0, −n(k), n, −n(k + v) − n removed: Let d > 0 be the largest radius for intervals centered at integer points such that the intervals never contain more than two of the a ′ j s: If no three elements of {a 1 , a 2 , a 3 , a 4 } are within a radius d > 0 from any single integer, then the sum (4.12) has the following bound The inequality is a Riemann upper bound using that the distance of the a j 's to an integer not equal to 0, −n(k), n, or −n(k+v)−n is ≥ 1 4 . I claim that the maximal radius d(k, v, n) increases proportionally to |k| ∨ |n|. Within the nearest integer, I have a 1 ≈ 0, a 2 ≈ −2k, a 2 ≈ n, and a 4 ≈ −2(k + v) − n. For n large enough (i.e. |n| ≥ 6|k|), then the claim is clearly true, since |v| ≤ 1 2 |k|. For |n| on the order of |k| or smaller, I already have that 0 and −2k are far apart. In order to have three of the a j 's within a small radius, then both n and −2(k + v) − n must be near 0 or −2k. However, n and −2(k + v) − n can only reach within a distance ≤ 1 4 |k| of each other if both are at least a distance ≥ 1 4 |k| from 0 and −2k (again by the constraint |v| ≤ 1 2 |k|). For the finitely many values of n such that α π |k+v+n| −1 > 1 4 , I can give almost the same treatment.
|k| . The remainder of the terms can be treated as in the case above. In both cases, the sums were bounded by a constant multiple of (|k| ∨ |n|) −2 . Now, I deal with the four exceptional terms m = 0, −n(k), n, −n − n(k + v), where exactly one the terms η(k + v + n, m − n) or η(k, m) is of the form η(p, r) for r = 0, −n(p). For that term, I use the inequality |η(p, r)| ≤ 1 rather than (4.11). For the other term, I apply (4.11) and approximate 2(k + v) ≈ n(k + v) and 2k ≈ n(k) (and I double the constant h to cover the error resulting from the approximation). The term |η(k + v + n, m − n)η(k, m)| is smaller than All of the four terms are bounded by 2h times For values of n near I(k, v), this will be larger than the expression (|k| ∨ |n|) −2 that bounds the remainder of the terms not in I(k, v), although for larger values of n, they have the same order. Thus, the values |κ v (k, n)| are bounded by some multiple h ′ of (4.13).
Since |v| ≤ 1 2 |k|, the above is bounded by the following where the inequality is by a Riemann upper bound.
Part (3): The analysis from Part (1) gives errors for |κ v k, m | 2 , m ∈ I(k, v) of order O (|k| −1 ), but there is additional decay when β(k) or β(k + v) are large. All the cases involve similar reasoning, so I focus on the case for for some C ′ > 0, then The second inequality is for some C ′′ > 0, since r − (k) is bounded by a constant multiple of (1 + |β(k)|) −2 . Thus, establishing (4.14) is sufficient for the proof. By (4.10) and the triangle inequality, .
and with Cauchy-Schwarz the last term above is O (|k| −2 ). Moreover, by the bounds in Part (2), the first two terms are each bounded by a multiple c ′ > 0 of |k| −1 . For large enough |k|, I have By the analysis in Part (1), there is a C > 0 such that Putting the above inequalities together, then I obtain (4.14) for C ′ = C + 3c ′ .
Part (4): By the bound in Part
where the order equality uses that |m| ≥ |k|.
Let T : L 1 (R) be the trace preserving map with integral kernel T (k 1 , k 2 ) = R −1 J(k 1 , k 2 ). The idealized momentum process (2.5) has a pseudo Poisson form in which jump times are determined by an outside Poisson clock, and the jump transition densities are given by the operator T . A similar structure holds for the original Lindblad dynamics by (2) of Lem. A.1.
Proof of Thm. 2.1. Define the map ρ → [ρ] D from B 1 L 2 (R) to L 1 (R), which sends a density matrix to its diagonal density in the extended-zone scheme representation [ρ] D (k) := Q k|ρ|k Q . The diagonal map is well-defined by the discussion in Appendix B.2. Let ξ = (t 1 , · · · ; t N ) be the sequence of Poisson times less than t. Define the map Φ The maps Φ where E is the expectation with respect to the Poisson process with rate R for the sequences ξ. Also r,ξ is contractive in trace norm, and conjugation by e − ir λ H is contractive in trace norm, I have the second inequality below where I identify t 0 ≡ 0 and t N +1 ≡ t for the boundary terms. The first inequality above is the triangle inequality. With the above and the summation formula The remainder of the proof is concerned with proving that the supremum in (4.15) is bounded by a multiple of λ for λ ≪ 1. By a direct calculation, Using the triangle inequality and making a change of variables where the values C M are defined as The second inequality in (4.17) can be found by splitting the integration R dk into the regions |2k + M | ≤ 1 and |2k + M | > 1. This splitting isolates some bad behavior (non-decay for large M ) occurring in regions of k where |2k + M | is small. The C M 's in (4.17) arise by applying Holder's inequality over the integration |2k + M | > 1 and by the inequality |ρ(k 1 , k 2 )| ≤ 1 2 ρ(k 1 , k 1 ) + 1 2 ρ(k 2 , k 2 ) for the integral kernel of ρ. The kernel inequality follows because ρ is a positive operator. The 2π α term in (4.17) comes from the |2k + M | ≤ 1 integration for which I apply mainly brute force: where g n > 0 are the gaps between the energy bands occurring at momenta k ∈ 1 2 Z − {0}. The infemum of the energy gaps g n is α π . The sum over n ∈ Z of |κ v (k, n)| |κ v (k + M, n − M )| is bounded through the Cauchy-Schwarz inequality and n |κ v (k, n)| 2 = 1. The key observation is k and k + M must lie on different energy bands, and it follows that E(k) and E(k + M ) differ by at least the length of the smallest energy band gap.
Next, I need to show that the sum of the C M 's is finite. A single C M can be bounded by (inf n g n ) −1 using some of the same reasoning as above. I will show C M decays on the order of |M | − 3 2 , and thus is a summable series. The difference |E(k) − E(k + M )| necessarily becomes large for |M | ≫ 1 except for cases when k + M is close to −k. However, by my restriction |2k + M | ≥ 1, the momenta k and k + M will not lie on neighboring energy bands, and thus their energies must differ by at least the length L |M | of the |M |th energy band. By [1, Thm.2.3.3], L n grows with linear order for n ≫ 1. Also, By the observation above, are therefore summable, and I turn to the only somewhat delicate part of the proof, which requires isolating the problematic terms in the sum of the For fixed k, v, the Cauchy-Schwarz inequality and n |κ v (k, n)| 2 = 1 yield Under the constraint |k| ∧ |k + M | ≥ 1 4 |M | and by Part (1) of Lem. 4.1, the terms on the bottom line of (4.18) decay on the order R of the first term on the right side in (4.18) will now require invoking the decay of j(v) at infinity and its boundedness through Lem. 4.1. The constraints n ∈ I(k, v), n ∈ I(k + M, v) + M , |2k + M | ≥ 1, and |k| ∧ |k + M | ≥ 1 4 |M | leave the following possibilities: • The inequality |k + v + 1 2 M | ≤ 1 4 holds and either n = 0 or n = M holds. • The inequality |v − 1 2 M | ≤ 1 4 holds and either |k + 1 2 n| ≤ 1 4 or |k + v + 1 2 n| ≤ 1 4 holds. The second case vanishes, since |v| ≈ 1 2 |M | ≫ 1 and by Jensen's inequality: The first case follows by Part (4) of Lem. 4.1.
Submartingales related to energy
In this section, I discuss certain key submartingales appearing in both the quantum and the limiting classical settings. First, let me define an operator-valued submartingale. Consider a probability space (Ω, F, P ) with a filtration F t , a Hilbert space H, and an operator-valued process Y t : Ω → B(H) adapted to F t and satisfying In the unbounded case, I refer to the process by the tuple (Y t , D). For the following discussion, I will consider a Schrödinger Hamiltonian H = P 2 + V (X) with positive potential V and domain D(H) ⊂ L 2 (R). The noise in my model is generated by an underlying Lévy process L t with jump rate density j(v). As before, let v n ∈ R, t n ∈ R + denote the jumps and jump-times for the Lévy process, and N t be the Poisson counter for the jump-times. Also let the unitaries U λ,t (ξ) be defined as in (2.9). Define the operator-valued process G t as for an observable G acting on L 2 (R). On bounded observables G, the trajectories are right weak*continuous with weak*-limits existing from the left, since the Hamiltonian evolution in the Heisenberg representation is weak*-continuous. Analogously to (2.8), the Heisenberg evolution for G ∈ B L 2 (R) can be written as In Prop. 5.1, the formula (5.3) may be interpreted as the definition for the dynamical maps Φ * λ,t : B L 2 (R) . I will suppress the λ and ξ dependence for the operator processes in the future.
In the lemma below, I study (5.2) for the special cases G = H and G = H ) , respectively, with the forms found below.
3. Moreover, I have the operator relations Proof.
Part (1): The energy process H t can be written as for n = N t . Through iteration of the above calculation, I obtain the relation For f ∈ D(H), I will now verify condition (5.1) for H t and M t . For H t , I have since the martingale part vanishes under the expectation. By two applications of Jensen's inequality for the first inequality below and using that f |P 2 t |f ≤ f |H t |f for the last inequality, The domain for A t is clear from its form.
Part (2): The equality H can be shown through a telescoping sum and the conservation of energy between momentum kicks in a similar way to the argument in Part (1). Also, that M ′ t is formally a martingale is clear through the symmetry of the rates j(v) = j(−v), however unlike for Part (1), it is not immediately clear that A ′ t is an increasing process. I have the equality by the second-order Taylor expansion Showing that i[X, i[X, H 1 2 ]] is a positive operator will complete the proof that A ′ t is increasing. Using the identity a Evaluating both sides by the double commutator with X, The following operator inequalities hold: where I have used that f (u) = 1 u+w and f (u) = u u+w are operator monotonically decreasing and increasing functions, respectively. Applying this inequality to (5.6), To see the equality to zero, I compute the two integrals through a change of variables w = h tan 2 (θ) and find ] is a positive operator, and A ′ t is increasing. Next, I show that M ′ t satisfies condition (5.1). Again by Taylor's formula, By similar reasoning as above, Using that |P | ≤ H 1 2 , I have the inequality where I have used the same change of integration as in the functional calculus above. Since e ibX has operator norm one, the above remarks imply Finally, for f ∈ L 2 (R), The same calculation shows that H 1 2 t satisfies condition (4.1). That condition also holds for A ′ t , since the process A ′ t is the difference between H 1 2 t and M ′ t .
Part (3): Since the martingale part has expectation zero, the equality follows trivially from the decomposition in Part (1). For E H 2 t ], the classical theory would have where M, M t is the predictable quadratic variation. In my case, dA t = σdt is a multiple of the identity operator and therefore commutes with everything. It follows that the equality above holds by the same argument as for the classical case. The processes Nt n=1 P t − n v n , Nt n=1 v 2 n − σ R , and σ R N t − σt are uncorrelated martingales, and thus Using P 2 r ≤ H r and (5. In the proof of Prop. 5.2, I apply Prop. 5.1 to gain information about certain martingales related to the Markov process K r . Define the energy process E r := E(K r ), where E(k) is the dispersion relation determined by (2.4). Define E : R → R + as the square root of the energy: E(k) = E 1 2 (k). I also use the symbol "E" to refer to the corresponding process E r = E(K r ). Recall that the kets |k Q are associated though a fiber decomposition of L 2 (R) with normalized Bloch functions ψ k ∈ L 2 [−π, π) given by (4.1). The mathematical connection between the results in Prop. 5.1 and the classical process K r is made through formulae such as in the equality where L r is the underlying Lévy process. The first equality above follows from the definition of the coefficients κ v k, i 2 . The value κ v k, i 2 ∈ [0, 1] is the probability for a lattice jump i ∈ Z conditioned on a Lévy jump v occurring from the momentum k. The rigorous meaning of expressions involving bra-ket notation can be traced back to the fiber decomposition such as in where H φ is the fiber Hamiltonian for φ ∈ [− 1 2 , 1 2 ) with k + v = φ mod 1. Define the process m r as By the rate symmetry j(v) = j(−v) of the Lévy jumps, m t is a martingale.
1. The process E r is a submartingale, and the predictable increasing part of its Doob-Meyer decomposition is σr. Moreover, the second moment satisfies Proof.
Part (1): Let u ∈ L 1 (R) be a probability density with R dk u(k)E 2 (k) < ∞. Construct the density matrix ρ = |f f | ∈ B 1 L 2 (R) for f (k) = u 1 2 (k). Let D λ,t , ρ λ,t , and D t be defined as in Thm. 2.1. For all λ > 0, The third equality is by Part (3) of Prop. 5.1. By Thm. 2.1, D λ,t converges to D t in the 1-norm as λ → 0 for every fixed t. To guarantee the convergence of R dkD λ,t (k)E(k) to R dkD t (k)E(k), it is sufficient to have a uniform bound in λ on the second moments R dk D λ,t (k)E 2 (k). Again using Part (3) of Prop. 5.1, there are constants Hence, a uniform bound for the second moment of the energy is given by is finite and equal to (5.12), where E u is the expectation beginning from an initial distribution u. The jump rate densities j k (k ′ ) = J(k ′ , k) are continuous in L 1 (R) as a function of k over intervals between lattice points in 1 2 Z. I can approximate a δ-distribution at k ∈ R− 1 2 Z with densities, and I have the result d dt E k [E t ]| t=0 = σ for k not on the half-spaced lattice. Since my initial distribution will always be a density and the jump rates are densities, the behavior assigned to the lattice values is irrelevant. It follows that E t is a submartingale with increasing part E 0 + σt.
The bound for E E 2 r follows by plugging in the explicit values for c 1 (r) and c 2 (r).
Part (2): Let u ∈ L 1 (R) be defined as in Part (1). By Part (2) of Prop. 5.1, for every t, λ > 0 the inequality below holds A similar argument as in Part (1) shows that H 1 2 t is a submartingale. By Part (1), the increasing part for the Doob-Meyer decomposition for E 2 t = E t increases with linear rate σ. I thus have the relation Since both terms on the right are positive, it follows that d dt M, M t ≤ σ.
Part (3): By (5.10), the terms of m r can be rewritten The process E t can be written as For a single term in the sum and K t − n = k, dL tn = v, then However, I can reorganize S t in a way reminiscent of Part (2) in Prop. 5.1: The first sum on the right is m t , and I denote the second sum on the right by a t . By the analysis in the proof of Part (2) of Prop. 5.1, the terms in the sum of a t are positive. Also, I note that when K t − n = k and |dL tn | = |v|, then The process S t is the conditional projection of E t on to the set of processes whose value at time t depends only on F t − and the Lévy process L t . Moreover, the process a t is the projection of S t that depends only on the jump times and the absolute value of the jumps |dL t |. Finally, A t is the predictable projection of a t . It follows that E t − S t , S t − a t = m t , and a t − A t are uncorrelated martingales with the following inequality for their predictable quadratic variations: where the last inequality is restricted to |v| ≤ |k|. In my analysis, I will first work to bound the error of substituting Q p|, |p Q with p|, |p , and secondly, I bound the error of substituting H φ − e −ivX on L 2 [−π, π) has operator norm ≤ 2|v|. By (4.3), there is c > 0 such that the distance between Bloch the vectors ψ k , ψ k ∈ L 2 [−π, π) for |k| ≫ 1 is bounded by where the second inequality holds for some c ′ > 0, since r 1 2 − (k) and 1−r I now bound the difference H 1 2 − |P | when evaluated by kets |k ′ with |k ′ | ≫ 1. By the formula for the square root of an operator in terms of its resolvents, However, the difference between the resolvent of a Laplacian and the resolvent of the Laplacian perturbed by a δ-potential has a simple form [1]. To use this, I will focus on a single fiber from the decomposition (B.1). For w ∈ R + and φ ∈ [− 1 2 , 1 2 ), the Green's function G φ,w : [−π, π) → C for the operator w − ∆ φ is given by the form The operator on L 2 [−π, π) determined by the integral kernel G φ,w (x − y) is equal to (w − ∆ φ ) −1 . The difference between the resolvents in the φ-fiber is where the expression on the left in square brackets denotes the operator on L 2 [−π, π) corresponding to the φ-fiber, and the operator A φ,w has integral kernel A φ,w (x, y) = G φ,w (x)G φ,w (−y). For k = φ mod 1, then Going back to (5.16), I have the relations The first equality follows from (5.18), and the second follows by (5.19) for k ′ = k + v and because 1+|β(k)| by the restriction |v| ≤ 1 2 |k|. The third inequality is by the definition of σ and Chebyshev's inequality through |v|≥ 1 Lemma 5.3. Let K t − = k for |k| ≫ 1 and |dL t | = |v| ≤ |k|. There exists an a > 0 such that Proof. Note that where the inequality follows from the same computation as in Part (1) of Prop. 5.1. However, the quantity Var E(K t ) F t − , dL t appears in the follow equation: The bottom line is positive. The second line of (5.20) can be written as 6 The limiting classical process I now shift my focus entirely to the classical stochastic processes K r whose probability density evolves according to the equation (2.5). In Sect. 6.1, I prove Thm. 3.1, which stated that t − 1 2 E st , s ∈ [0, 1] converges in distribution as t → ∞ to the absolute value of a Brownian motion. Section 6.2 contains various lemmas related to the random variables τ n+1 −τn |Kτ n | being approximately exponentially distributed with expectation ν −1 , where τ n , τ n+1 are successive reflection times. Finally, Sect. 6.3 contains proofs of the lemmas in Sect. 3 and completes the proof of Thm. 2.3.
A submartingale central limit theorem
The difference between the quantities |k| and E(k) = E 1 2 (k) is bounded by a constant, and the difference is even O (|k| −1 ) for |k| ≫ 1 as a consequence of the Krönig-Penney relation (2.4). Working with the process E r is advantageous, since it is a submartingale by Part (2) of Prop. 5.2, and the increasing part of the Doob-Meyer decomposition for the square E 2 r = E r increases linearly. Thus, the processes t − 1 2 |K st | and t − 1 2 E st , s ∈ [0, 1] are close, and I will work with the latter. As before, M r and A r will denote the martingale and increasing parts the Doob-Meyer decomposition for E r . The main result of this section is proof of Thm. 3.1. One of the key inputs for the proof is Lem. 6.1, which yields that E r typically spends most of the interval r ∈ [0, t] with values E r ≫ 1. This is important, since some estimates, such as in Part (4) of Prop. 5.2, only become effective when E r is large. Lemma 6.2 offers a lower bound for the predictable quadratic variation of the martingale m t from Prop. 5.2. This lower bound shows that m, m r essentially grows linearly with rate σ, and when combined with Part (2) and (3) of Prop. 5.2, this implies that t − 1 2 M st and t − 1 2 m st are nearly equal. The result of the lemma below holds for a general class of positive processes whose squares are submartingales, whose variances increase linearly, and that satisfy some Lindberg condition (to avoid the situation where the jumps are very large but very infrequent). The lemma is only stated for E r here. The lemma implies that E r typically spends most of a time interval r ∈ [0, t] with values on the order of t 1 2 −ρ , 0 < ρ < 1 2 for large t, where "most" is all except for a total duration on the order O (t ρ ). Lemma 3.4 in [12] is similar, although the proof here employs different techniques. In the more general situation in which the second momentum of E r grows at varying rates, similar statements can be made for a stochastic time change of E r .
Proof. I begin with an application of Chebyshev's inequality to obtain Set ς 0 = ̟ 1 = 0, and define the stopping times ς n , ̟ n ≤ t such that Let Ξ t be the number ̟ n 's less than t. I clearly have Moreover, notice that I hence have an upper bound in terms of the expectation for the number of upcrossings Ξ t and the expectation for the duration of a single upcrossing ς n − ̟ n conditioned on the event n ≤ Ξ t . By the submartingale uncrossing inequality [14], I have the first inequality below The last inequality is for t large enough. Next, I focus on the expectation of the incursions ς n − ̟ n . Whether or not the event n ≤ Ξ t occurred will be known at time ̟ n , so In examining the expression on the right side above, I will set ̟ n = 0 and ς = ς n − ̟ n . Since σt is the increasing part of the Doob-Meyer decomposition for E 2 t − E 2 0 , the optional sampling theorem gives If the process E t were constrained to make jumps bounded by J > 0, then I could immediately reason that since E s is either still in the interval [0, 2ǫt ̺ 1 ] at time s = ς ∧ T or has jumped out by a value of at most J. Plugging (6.2) and the bound σ −1 (2ǫt ̺ 1 + J) 2 for sup n∈N E ς n − ̟ n n ≤ Ξ t into (6.1) leads to the desired result. For the case in which there is not a cap on the jumps, it is necessary to be more careful. Consider a modified process in which jumps E s − E s− greater than 2ǫt ̺ 1 are removed. The modified process can be written as for Poisson times t n ≤ r. The modified process E ′ r is still positive, and for large enough t, (E ′ r ) 2 will still be a submartingale such that the increasing part of its Doob-Meyer decomposition grows nearly at the linear rate σ (although growing at a lesser rate, such as σ 2 , is sufficient). Let ς ′ be analogously defined to ς as the hitting time that E ′ r jumps out of [0, 2ǫt 2̺ 1 ]. By the definitions of these processes, I always have ς ≤ ς ′ , Combining this inequality with (6.2) in (6.1) finishes the proof.
I will make a few comments about the Markov chain θ n = Θ(K tn ), which is the sequence of momenta K tn contracted to the torus T = [− 1 4 , 1 4 ) at the Poisson times t n . Since the lattice components of the momentum jumps live on Z, they do not influence the jump rates for the contracted process θ n . Define the map · : If q t n−1 , q tn are the distributions for the momentum at times t n−1 ,t n , then q tn = T q t n−1 , where the operator T has integral kernel As a consequence of (2) of List 2.2, j ∞ = sup − 1 4 ≤θ≤ 1 4 i∈Z j(θ + i 2 ) is finite, and thus T maps L 1 functions to L ∞ : By (6.4), the process K r forgets its previous locations on the torus T exponentially fast. In particular, the process Θ(K r ) will not not be disproportionately concentrated around the value zero. This is important for the proof of the following lemma, which relies on Part (4) of Prop. 5.2, since the bound in Part (4) of Prop. 5.2 is only useful for |β(K r )| = 1 2 |Θ(K r )||n(K r )| ≫ 1, i.e., when |Θ(K r )| is not too close to zero. for some constant g > 0 and all t ∈ R + .
Proof. By Parts (2) and (3) where the second term on the right subtracts over-counting in the first term and will be of order Approximating m, m t by the process s t := Nt n=1 E V(K tn ) F t n−1 e n , yields a difference bounded by since sup k V(k) ≤ σ and the e n 's are independent of each other and everything else. The conditional expectation E V(K tn ) F t n−1 is a function Q : R → [0, σ] of K t n−1 . My expression of choice will be t 0 drQ(K r ), which is nearly equal to s t except for two extra boundary terms of order O (t −1 ).
The key input for the proof will be an upper bound for σ − Q(k) when |k| ≥ 2t 1 6 . The probability that a Poisson jump v has absolute value |v| > t 1 6 will decay superpolynomially for large t by (1) of List 2.2, and thus By Part (1) of Lem. 4.1, the sum of κ v (k, i) 2 for i / ∈ I(k, v) will be ≤ ct − 1 3 for some c > 0. Since |v| ≤ t 1 6 , the values |k + i + v| must be ≥ t 1 6 for i ∈ I(k, v). Applying Part (4) of Prop. 5.2, the integral in (6.6) is bounded by where I have used that the values k + i + v have absolute value ≥ t 1 6 and the shifts i ∈ Z do not change the values k + v modulo 1 2 . The supremum in the second inequality is smaller than µ by (2) of List 2.2. Thus, for |k| > 2t − 1 6 , the above gives Now, I am ready to bound the expectation of t 0 dr σ − Q(K r ) . Define T t = t −1 t 0 drχ(|K r | ≤ 2t 1 6 ). By considering the events T t > t 1 6 and T t ≤ t 1 6 , I get the bound For the event T t ≤ t 1 6 , the third term on the right corresponds to the part of the trajectory in which |K r | is ≥ 2t 1 6 , and the second inequality uses (6.7). By Lem. 6.1 with δ = 1, ǫ = 2, and ̺ 1 = ̺ 2 = ̺ 3 = 1 6 , the probability that T t ≥ t − 1 6 has order t − 1 6 (since |K r | ≈ E r + O (|K r | −1 )).
The proof that the processes t − 1 2 E st , s ∈ [0, 1] converge as t → ∞ to the absolute value of a Brownian motion relies on the following features of E t : 1. The process E r is a positive submartingale with martingale component M r .
The increasing part of the Doob-Meyer decomposition for E 2
r is E 2 0 + σr.
3. When E r ≫ 1, then σ − d dt M, M t goes to zero (at least on average over long periods of time).
4. The quadratic variation of M r satisfies a Lindberg condition.
Theorem 3.1 follows from a more general theorem assuming conditions of the type above. In principle, if (2) does not strictly hold but the expectation of E t tends to grow with t, then one can obtain this assumption by considering a stochastic time change τ t : E τt = E ′ t . Assumption (3) is required to guarantee that the process t − 1 2 E st behaves diffusively away from zero. It forbids, for instance, that the process E r has the deterministic trajectory E r = σ 1 2 r 1 2 . To make use of (3), I will need to verify that E t spends most of the time at "high" values, and this verification is made through Lem. 6.1. Assumption (4) is to guarantee asymptotic negligibility as is required in the martingale central limit theorem. Theorem 3.1 is similar to Thm. 4.1 in [12].
Proof of Thm. 3.1.
All "convergence in law" in this proof will refer to the uniform metric. Let M r and A r be the components of the Doob-Meyer decomposition for E r , and define the submartingale G r = M r + sup 0≤s≤r −M s . The convergence in law of t − 1 2 E st is implied by the following statements: (i). The processes (σt) − 1 2 G st , s ∈ [0, 1] converge in law as t → ∞ to the absolute value of a standard Brownian motion.
(ii). The random variables sup 0≤s≤1 t − 1 2 A st − sup 0≤r≤st −M r converge in probability to zero as t → ∞. Part (3) of Prop. 5.2 states that M r is a sum of the uncorrelated martingales m r and M r − m r . By Lem. 6.2, the predictable quadratic variation for t − 1 2 m st , 0 ≤ s ≤ 1 converges in distribution to σs. By [36, Thm.VIII.2.13], it will follow that t − 1 2 m st converges in law to a Brownian motion with diffusion constant σ if the following Lindberg condition holds as t → ∞: The quadratic variation of m r is dominated by the quadratic variation of the Lévy process by Part (3) of Prop. 5.2, and hence I have the first inequality below.
The second inequality is Jensen's, and the equality holds because the jumps v n occur with Lévy rate j(v). By the assumptions on the jump rates j(v), the fourth moment is finite, and the bottom line tends to zero for large t. That proves the convergence in law for t − 1 2 m st . Next, I argue that t − 1 2 sup 0≤s≤1 |M st − m st | converges weakly to zero as t → ∞. It will then follow that t − 1 2 M st converges to a Brownian motion. Since the martingales m r and M r −m r are uncorrelated, the formula holds: With Doob's inequality, the inequality (6.8), and Lem. 6.2, I have (ii). Note that G r is the smallest positive submartingale that has Doob-Meyer decomposition with M r as the martingale part, so E r ≥ G r . I begin with a technical point extending the discussion of (i). Let A s is a term related to the times when t − 1 2 G st is likely to jump to zero: Z (t) The dependence on K r enters through the statistics for M r − M r − conditioned on the event that r ∈ R + is a Poisson time (i.e. N r − N r − = 0). It follows that s is increasing in s and so sup By Part (1), the first term on the right converges in distribution to zero as t → ∞. Now to address Z (t) 1 . I always have the inequality F (G r , K r ) ≤ σ R , and for the regime |G r | ≫ 1, then To see the order equality, I first break M r − M r − into the two parts (M − m) r − (M − m) r − and m r − m r − and use the simple inequality to separate the terms in the expectation. By the reasoning in the proof of Part (3) of Prop. 5.2, |m r − m r − | 2 is smaller than the square of the Lévy jump |L r − L r − | 2 . Since the Lévy jumps have exponential tails, the expectation of |m r − m r − | 2 χ(|m r − m r− | ≥ 1 2 G r ) must decay exponentially in G r (and therefore be negligible). The expectation of (M − m) r − (M − m) r − 2 is smaller than a 1+|β(K r − )| by the proof of Part (4) of Prop. 5.2. Note that E r ≥ G r and E r ≈ |K r | to conclude that when G r is large, then |K r | is also large. Doubling the bound to cover the smaller term |m r − m r − | 2 above, I can write For some c > 0, the following inequality holds: This can be either proved using that t − 1 2 G st converges to the absolute value of a Brownian motion and a little estimate involving Gaussians or by a small modification of the proof of Lem. 6.1 with E r replaced by G r . Hence, for any δ > 0 I can pick an ǫ small enough to make the probabilities P[T ′ ǫ,t < δ] uniformly small for large t. By a similar argument as in the proof of Prop. 6.2, the right side of (6.10) converges to zero, which finishes the proof of (6.9).
With (6.9) as a tool, I go to the core of the proof. Since t −1 E 2 st and t −1 G 2 st are both submartingales, taking Doob-Meyer decompositions allows me to write their difference as a martingale M Since the left side is made up of positive terms, I can immediately conclude that Hence, showing that the terms on the right converge in distribution to zero for large t will be sufficient. I have already shown sup 0≤s≤1 σs − A (t) s converges to zero, and now I turn to the martingale component.
Since the left side of (6.11) is positive, it follows that −M (t) s for all s ∈ [0, 1], and thus τ has mean zero, the equality below holds: The first inequality is Chebyshev's and the last is (6.12). Since the random variables σ − A (t) 1 are bounded by σ and converge to zero in distribution as t → ∞, the expectation on the right tends to zero.
The waiting-times between momentum reflections
This section focuses on bounding and characterizing the length of the time intervals between "Bragg reflections" for the momentum process. The stochastic process K r exhibits a singular rate of signchanging in comparison to an ordinary random walk due to the lattice components of the momentum jumps. The lattice jumps yielding reflections are "macroscopic", since they transport the momentum to a value that would require many steps through the Lévy component to reach. This unusual signflipping characteristic was not apparent for the study of E r = E 1 2 (K r ) in Sect. 6.1, since E r depends only on the absolute value of the momentum. The results below are directed toward showing that the waiting-time τ m+1 − τ m between successive reflections τ m , τ m+1 is approximately an exponential distribution with mean ν −1 |K τm | and is independent of the subsequent fluctuations in momentum. This idealized behavior emerges for high beginning momentum |K τm | ≫ 1.
It will be useful to work with an idealization of the momentum process that makes the same Lévy jumps with rate j(v) except that the additional lattice jump m ∈ Z takes only the values 0 or −n(k + v) with the probabilities defined below. Define the functions Π − (k) = α 2 This process describes a particle whose momentum is reflected at the next Poisson time based on a weighted coin flip whose bias is determined by the sum of the current position k and next Lévy increment v. Of course, the simplified statistics is a useful approximation only for time intervals when the momentum is high. Note that the idealized process closely resembles the process proposed in Sect. 2.2.6. I will refer to probabilities and expectations in the law above by P and E, respectively. For a technical reason explained below, I define the particle to make an additional jump −n(K 0 ) with probability r − (K 0 ) at an infinitesimal time after zero . Consider a particle beginning with a momentum k with |k| ≫ 1 and making jumps according to the simplified law above, and let τ be the first reflection time, i.e., the non-trivial lattice jump. I will sketch why the random variable τ |k| is approximately an exponential with mean ν −1 . It is equivalent to consider a discrete-time random walk X n , n ∈ N beginning from k with jump increment density j(v) R and a hitting timeN marking the first tails outcome for coin tosses with tails probabilities R − (X n ). As long as X n does not stray too far from k, the probabilities R − (X n ) are approximately The problem is thus further reduced to the contracted Markov chain θ n := Θ(X n ) living on the torus T = [− 1 4 , 1 4 ) with the transition operator T : L 1 (T) (defined above Lem. 6.2). The probabilities Π − |k| θ n decay rapidly for | θ n | ≫ |k| −1 , so θ n will typically require many steps to score a reflection. Since the chain θ n is exponentially ergodic to the uniform distribution over T, the probability of a reflection on a given time step is approximately 2 T dθΠ − (|k|θ) ≈ α |k| . The random variableN |k| should thus be close in law to a mean-α −1 exponential for |k| ≫ 1. This conclusion is consistent with the ansatz (6.15), since the fluctuations in the random walk X n over [0,N ] will be on the order O (|k| 1 2 ), which is small compared to |k|. Since the Poisson times have rate R and ν := Rα, the distribution for τ |k| will be approximately a mean-ν −1 exponential. The connection between the momentum process and the simplified law (6.14) is not obvious, since the approximations for the probabilities |κ v (k, n)| 2 in Prop. 4.1 reduce for |k| ≫ 1 to expressions r ǫ 1 (k)r ǫ 2 (k + v), ǫ 1 , ǫ 2 ∈ {±} when n ∈ I(k, v) and zero for n / ∈ I(k, v). For a hint of the link, note the identity R − (k) = 2r − (k)r + (k). (6.16) An intermediary law between the original momentum process and (6.14) is given by the process that accompanies a Lévy jump v from the momentum k with four possible lattice jumps with values and probabilities given by: Naturally, when n(k) = n(k + v), then the probabilities are summed. I refer to these statistics by P ′ , E ′ . There is an underlying equivalence between the laws (6.14) and (6.17) such that they may be embedded on a single probability space in which the trajectories match for specific realizations except for time periods around reflection times where the values become staggered. To demonstrate this, I will consider only skeletal chains, since the waiting-times between jumps are independent of the states. As before, let v m , m ∈ N be a sequence of independent random variables with density j(v) R . Let the (non stationary) Markov chainX n have increments whose law depends on the parity of n ≥ 1 as where the second column on the right lists the probabilities for the lattice component. Thus, the jumps v m occur at even time steps and coins are flipped at each time step to determine whether there is an extra lattice jump. The chainsX 2m+1 andX 2m for m ≥ 0 are each Markovian and have the statistics of (6.14) and (6.17), respectively. This result depends on the symmetries j(v) = j(−v) and r − (k) = r − (k − n(k)). The two idealized laws are interchangeable for the purposes of this section. Let S : R → {±1} be the sign function. I will use the term "sign-flip" in a technical sense to refer to a Poisson time t n such that for some odd m ∈ N S(K t n−m−1 ) = S(K t n−m ) = −S(K tn ) = −S(K t n+1 ) (6.18) and S(K t n−r ) = −S(K t n−r+1 ) for r ∈ [1, m). In other words, there is an odd-numbered sequence of sign-changes ending at the time t n . Only sign-flips in which there is a single sign-change are likely to occur in practice. The complication in the definition results from the fact that there is a comparatively high probability of a sign-change at the Poisson time after a sign-change, and I would like to avoid counting these "fake" sign-flips. For example, suppose the beginning momentum is k with |k| ≫ 1 and the next two Lévy jumps are v 1 and v 2 . By the approximate statistics (6.17), the probability assigned for the lattice jump −n(k + v 1 ) at the first Poisson time is r + (k)r − (k + v 1 ), and the resulting landing can not be too small if this jump is likely to occur. The probability that the particle makes another lattice jump n(k + v 1 ) at the second Poisson time to land at k + v 1 − v 2 is approximately where the inequality is by the lower bound r + ≥ 1 2 and the symmetry r − (k − n(k)) = r − (k). Thus, the second sign-change has a nonnegligible probability ≥ 1 2 r − (k + v 1 ) regardless of the value v 2 . A sign-flip time is not a hitting time, since it requires the information from the following Poisson verifying that the sign of the momentum does not change again at next the momentum jump. However, this will not give much information. Consider K 0 = k with |k| ≫ 1, and let q ′ , q ∈ L 1 (R) be the distribution for the first Poisson time with and without conditioning to not change sign, respectively. Using the estimates (1) and (3) from Lem. 4.1, it can be proven that the probability of a sign-change can be at most ≈ 1 2 . Hence, I will have the bound In particular, there are no density peaks near lattice points 1 2 Z, where approximations tend to be weaker.
The following proposition is the main result of Sect. 6.2 and the proof is in Sect. 6.2.2. The inequalities in Prop. 6.3 have the following purposes: Parts (1) and (2) are characterizations for how close the random variables τ m+1 −τm |Kτ m | are to mean-ν −1 exponentials for |K τm | ≫ 1. Part (3) bounds the amount of time that the momentum process spends performing "fake sign-flips" before the actual sign-flip occurs in the technical sense (6.18). Part (4) states that the probability of a macroscopic fluctuation in the absolute value of the momentum before a sign-flip is small. Part (5) shows that the fluctuation in the absolute value of the momentum over the time interval between sign-flips is nearly uncorrelated with the length on the time interval. Notice that in Prop. 6.3 the momentum process is conditioned not make a sign-flip at the first Poisson time. This is exactly the situation that I have for the process K r , r ∈ [τ m , ∞) when given the information F τ − m and τ m is a sign-flip. This does not explicitly cover the case that τ m is not a sign-flip as when τ m = ς j for some j ∈ N or These events do not occur frequently enough to generate nonnegligible contributions to the quantities that I study (in fact, Part (4) of Prop. 6.3 bounds the probability of the event |K τm | / ∈ [ 1 2 |K τ m−1 |, 3 2 |K τ m−1 |]). Without conditioning the momentum to remain the same sign after the first jump, the estimates in Parts (1) and (2) may be distorted if K 0 is near a lattice point 1 2 Z, since τ will have a comparatively high probability of occurring at the first Poisson time. However, Prop. 6.3 still offers bounds for the moments of τ |k| in that case. There are other natural conditions on the beginning momentum that will yield the results of Prop. 6.3. The assumptions that K r begins with the value K 0 = k and is conditioned to avoid a sign change at the first Poisson time can be replaced by the assumptions that K r begins in a distribution given by a density q ∈ L 1 (R) concentrated around the value k and that the contracted density q ∈ L 1 (T) is bounded. Proposition 6.3. Let ζ > 0, K 0 = k for |k| ≫ 1, and K r be conditioned not to make a sign change at the first Poisson time (i.e. S(K 0 ) = S(K t 1 )). Define τ to be the first time that either K r has a sign-flip or |K r | jumps out of the set [ 1 2 |k|, 3 2 |k|] depending on what occurs first. For fixed ζ, m > 0, there exist C and γ 0 such that the following inequalities hold for all k and 0 < γ ≤ γ 0 : Recall by the remarks preceding Lem. 6.2 that the chain θ n = Θ(K tn ) on the torus T = [− 1 4 , 1 4 ) is Markovian with bounded transition operator sup θ 1 ,θ 2 ∈T T(θ 2 , θ 1 ) < µ R . Since the lattice jumps do not appear in T, the contracted process for the idealized momentum process is exactly the same. The torus chain has detailed balance, since the kernel is symmetric T(θ 2 , θ 1 ) = T(θ 1 , θ 2 ) as a consequence of the symmetry j(v) = j(−v). The process is thus time-reversible, and its stationary state is the uniform distribution. Due to the boundedness of the kernel, the process is exponentially ergodic, and even in the supremum norm I have the existence of an ε > 0 such that The laws P and P ′ determine two probability measures on the sequence of momenta K t j . I denote the total variation distance between the measures on sequences of length M by · Var,M . Let A M be the event |K t j | ∈ [ 1 2 |k|, 3 2 |k|] for 0 ≤ j ≤ M , and χ(A M )P be the positive measure that agrees with P for events in A M and assigns events in A c M zero weight. The key inputs for the proof of Lem. 6.4 are the estimates (1) where I have used the triangle inequality with a telescoping sum determined by inserting the measures χ(A) P (n) . Let dL n and dℓ n denote the Lévy and lattice jumps at the nth Poisson time. The laws P (n) and P (n−1) differ only at the nth Poisson time in the probabilities that they assign for the lattice jump dℓ n = m ∈ Z. I can bound a single term from the sum above through the inequality whereq n andq ′ n are the densitieŝ The first inequality (6.21) follows since A M ⊂ A n for M ≥ n. I will show that the right side of (6.21) is O ( log(|k|) |k| 2 ) for |k| ≫ 1. By (1) of List 2.2 and Chebyshev's inequality, I can replace the integration R dv in (6.21) by the restriction |v| ≤ 1 4 |k| with an error that decays exponentially for |k| ≫ 1 (since the integrand is bounded by one). Under P ′ , only the lattice jumps m ∈ I(p, v) have nonzero probability. For the law P, Part (1) when |k| ≫ 1. Since the densityq ′ n (p) is supported on the interval |p| ∈ [ 1 2 |k|, 3 2 |k|], I can restrict the summation m in (6.21) to m ∈ I(p, v) with an error O (|k| −2 ).
Lemma 6.5. Let t j for j ∈ N be the Poisson times and K 0 = k. For |k| large, there are c, C > 0 such that the following inequalities hold: The same statements hold with the statistics P replaced by P.
Proof.
Part (1): The laws P and P ′ determine measures on the sequences K t j . Let · Var,4 and the set A 4 be defined as in Lem 6.4. The probability of the event A c 4 is order O (|k| −2 ) for both P and P ′ , since the random variables |K t j+1 | − |K t j | have uniformly bounded second moments and by Chebyshev's inequality. Thus, where the second equality above is by Lem 6.4. However, the variational distance P − P ′ Var, 4 will bound the difference between the probabilities Therefore, I can substitute P with P ′ with an error o(|k| −1 ). One way in which the event S(K 0 ) = S(K t 1 ) = −S(K t 2 ) = −S(K t 3 ) can occur is if the Lévy jumps satisfy |v i | ≤ |k| 1 2 , i = 1, 2, 3 and the corresponding lattice jumps are respectively 0, −n(k +v 1 +v 2 ), 0. This possibility gives a lower bound I will use the identities r + (k ′ ) = r + k ′ − n(k ′ ) and Π − 2 −1 Θ(k ′ )n(k ′ ) = 2r + (k ′ )r − (k ′ ) for k ′ = k + v 1 + v 2 . Since r + is ≥ 1 2 and j(v 2 ) ≥ 1 µ for |v 2 | ≤ 1 by (3) of List 2.2, I have a bound from below given by where I have used that R dk Π − (k) = α.
Part (2): By the argument in Part (1), I can approximate P with P ′ . There are four possible pairs of lattice jumps that will give a sign change for the second jump and not the first: 0 for the first and −n(k + v 1 ) or −n(k + v 1 + v 2 ) for the second, or n(k) − n(k + v 1 ) for the first and −2n(k) + n(k + v 1 ) or n(k) − n(k + v 1 ) − n(k + v 1 + v 2 ) for the second. The cases require the same analysis (with more or less messy notation), so I will analyze the case in which the jumps are 0 and −n(k + v 1 ): In the first inequality, I have used that r + ≤ 1. The norm j ∞ is smaller than µ by assumption (2) of List 2.2.
6.2.1
Results for the idealized law Lemma 6.6 concerns the simplified momentum process (6.14), or alternatively (6.17), and it is applied in the proof of Lem. 6.7 to obtain the analogous result for the original momentum process. Part (1) of the lemma states that random variables Nτ n+1 −Nτ n |Kτ n | are approximately mean-α −1 exponentials for |K τn | ≫ 1. The proof is based on the heuristics in the paragraph (6.15). Lemma 6.6. Let β, ζ > 0 and K 0 = k with |k| ≫ 1 be conditioned not to change signs at the first Poisson time. Let τ be the first Poisson time that either the process makes a sign flip or its absolute value jumps out of the interval [ 1 2 |k|, 3 2 |k|]. For fixed ζ, there is a C > 0 such that for all k ∈ R, the following inequalities hold: where M := ⌊|k| β ⌋. Proof.
Part (1):
Since I am considering the probability of the event P N τ ≥ a|k| that is determined when the first reflection occurs (or, less likely, when |K Nτ | leaves the interval [ 1 2 |k|, 3 2 |k|]), the problem can be formulated in terms of a random walk X n = k + v 1 + . . . v n , where the increments v n are independent and have density j(v) R ; and the reflections are decided by flipping coins with heads weight Π + 2 −1 n(X n )Θ(X n ) . The problem is thus translated to characterizing the distribution for the first tails outcome. This scheme ignores the special rule for the first momentum jump for the law (6.14), but that will not generate a difference. I am also ignoring the technical definition (6.18) for a sign-flip, which is only relevant when working with the laws P and P ′ for which the double flips are likely. Let F ⊂ N be the set of times n with either a tail outcome or such that |X n | lands outside [ 1 2 |k|, 3 2 |k|]. Also, letN be the smallest element in F.
First, I will show that P N ≥ a|k| decays exponentially for a ≫ 1. By a simpler argument than in the proof for Part (1) of Lem. 6.5, there is a B > 0 such that for all |X n−1 | ∈ 1 2 |k|. With an iterated conditional expectation and an application of (6.25), then where B is the constant in the lower bound from Part (1) of Lem. 6.5. By induction, P N > n ≤ (1 − B |k| ) n . It follows thatN |k| has finite exponential moments E[e γ|k| −1N ] for 0 ≤ γ < B, which are uniformly bounded for all large |k|.
By the above P N ≥ m decays exponentially with rate at least (1 − B |k| ) m . Hence, the values P N ≥ a|k| and e −αa will be both small for a > |k| ζ , and I can focus my analysis on a ≤ |k| ζ . Let me setn = ⌊a|k|⌋ andm = ⌊|k| ζ ⌋. I can rewrite P N ≥n as where rn ,m (p) is the density for Xn −m conditioned on the eventN >n − 1. The third equality above holds by the strong Markov property. By definition, the density rn ,m (p) will have its support over |p| ∈ [ 1 2 |k|, 3 2 |k|]. I will argue that the integral in the last line of (6.26) has the |k| ≫ 1 asymptotics R dp rn ,m (p) P m / ∈ F N ≥m, The result (6.27) can be applied inductively in (6.26) for P[N >n − 1], . . . , P[N >m + 1] to obtain where A = B ∧ α. The second equality follows by a simple argument yielding that P N ≥m − 1 = O (|k| ζ−1 ). Since ζ > 0 is arbitrary, the result would follow. I will separate the analysis showing (6.27) into proofs of the statements (i)-(iii) below. For |p−k| ≤ |k| 1 2 +ζ and |k| ≫ 1, the following statements hold: (i). The difference between the left side of (6.27) and |p−k|≤|k| 1 2 +ζ dp rn ,m (p) P m / ∈ F N ≥m, X 0 = p (6.28) decays superpolynomially in |k|.
The value |p−k|>|k| 1 2 +ζ dp rn ,m (p) is smaller than the probability of the event sup 0≤n≤n |X n − k| > |k| 1 2 +ζ . The probability P sup 0≤n≤n X n − k > |k| 1 2 +ζ will decay superpolynomially fast for |k| ≫ 1, since where the first inequality is Chebyshev's, the second is Doob's maximal inequality, and σ ′ is the second moment of j(v) R . The last inequality uses the capn ≤ |k| 1+ζ . The same argument can be applied to bound P sup 0≤n≤n k − X n > |k| (ii). I first show that, when beginning from X 0 = p for |p| ∈ [ 3 4 |k|, 5 4 |k|], the probability of the event m / ∈ F is nearly the same with or without conditioning on the informationN ≥m. I will bound the probability P m / ∈ F X 0 = p in (ii ′ ) below. First, I need to show that P N ≥m X 0 = p is nearly 1. This requires another inductive argument concluding that The first inequality follows from the analysis below. I can assume that Xm = Xm −1 + vm has absolute value |Xm −1 + vm| ≥ 1 2 |k| by (6.29), so Now, I use that P N ≥m X 0 = p is close to 1. Let h, h ′ be the distributions for Xm −1 when beginning from X 0 = p and conditioned or not conditioned on the eventN ≥m, respectively. By similar reasoning as above where the last inequality follows from (ii ′ ). Approximating the unconditioned expression P m / ∈ F X 0 = p is more straightforward. I will denote the density P Xm = w X 0 = p = Tm δ(· − p) by qm ,p . I have the following approximations By the same argument as (6.29), the probability that |K r | leaves the interval [p −m, p +m] afterm steps after beginning from p will be superpolynomially small. Hence, the approximation in the second line above with the restriction of the integration to [p −m, p +m] will have a superpolynomially small error. The error in the last equality comes from the inequality for x = 2 −1 n(p) Θ(w) and y = 2 −1 n(w) Θ(w). Finally, Π + 2 −1 n(p) Θ(w) only depends on w modulo 1 2 , so I have that where q ′ m,p (w) := q m,p (w)χ(|w − p| ≤m). The second approximation above follows by (6.20), since qm ,p = Tm δ(· − p) must be superpolynomially close for |k| ≫ 1 to the uniform distribution 2. The order equality on the bottom line uses that |p| ≥ 1 2 |k|.
(iii). Plugging in the result (i) to (6.28), and expanding |p| −1 around k results in |p−k|≤|k| 1 2 +ζ dp rn ,m (p) 1 − α |p| It remains to show that |y|≤|k| 1 2 +ζ dy y rn ,m (y + k) is O (|k| 2ζ ). This will require an analysis of the density rn ,m and related densities r ′n ,m , r ′′ n,m , which can be written I claim that the following order equalities hold: I will return to a discussion of (6.36) below in (iii ′ ). It follows from (6.36) that In that case, it is sufficient to bound the expression |y|≤|k| 1 2 +ζ dy y r ′′ n,m (y + k). The density r ′′ n,m can be written as r ′′ n,m (p) where qm is density for the random variable Xm. The density r ′′ n,m corresponds to the distribution for Xn −m conditioned on the event F ∩ [m + 1,n − 1] = ∅. Moreover, the integral of qm(w)χ(|w − k| > |k| ζ ) will be superpolynomially small, since the jumps in the random walk X i , i ∈ [1,m] have mean zero and exponential tails. Hence, I can split Xn −m −k in to a sum of components Xm −k and Xn −m −Xm, where |Xm − k| is typically small and Xn −m − Xm has density Um ,n,w ∈ L 1 (R) given by when Xm = w and conditioned on the event F ∩ [m + 1,n − 1] = ∅. Since the arguments of the expectations defining Um ,n,w depend only on the contracted random walk Θ(X m ), the density satisfies the shift invariance Um ,n,w (p) = Um ,n,w+ 1 2 m (p) for m ∈ Z. This shift symmetry gives the inequality in the second line below: The equality above follows by approximating Xn −m −k using Xn −m −Xm and restricting the integration over w ∈ R to |w − k| ≤ |k| 1 2 +ζ . The density qm ∈ L 1 (T) will be superpolynomially close to the uniform distribution over T = [− 1 4 , 1 4 ) by (6.20). However, if qm is replaced by 2, then the resulting expression is zero, since the symmetry Um ,n,θ (y) = Um ,n,−θ (−y) holds as a consequence of the symmetry j(v) = j(−v) for the jump rates. Thus, the last term on the right side of (6.37) is also O (|k| ζ ).
(iii ′ ). I will focus on bounding the difference rn ,m (y+k) r ′ n,m (y+k) − 1 over the domain |y| ≤ |k| 1 2 +ζ , since the order equality on the right side of (6.36) follows by simpler analysis. Let G be the event sup 0≤j≤n X j − k| ≤ |k| 1 2 +ζ . By (6.29), 1 − P[G] decreases superpolynomially with large |k|. The difference between the expressions in the denominators of (6.33) and (6.34) is smaller than The above inequality follows from inserting a telescoping sum and using the inequality (6.31).
Since I have conditioned on the event G, then Removing the conditioning on G will make another superpolynomially small error, and a single term from the sum above is bounded by The conditional density for Θ(X j ) given Θ(X j−1 ) is bounded by µ as a result of assumption (2) from List 2.2 (and the same for Θ(X j+1 ) given Θ(X j )). I can make use of this to obtain bounds in which Θ(X j ) is decoupled from the other variables, and where Θ(X j ) and Θ(X j+1 ) are integrated with respect to the constant density µ over T. Integrating out X j yields a factor 2 T dθΠ + 2 −1 n(k)θ ≈ α |k| < 2α |k| . Moreover, I can put the variable back in the expression by inserting 2Π + 2 −1 n(k)Θ(X j )), since 2Π + ≥ 1. The conditional density for Θ(X j ) given Θ(X j−1 ) (and Θ(X j+1 ) given Θ(X j )) is bounded from below by µ −1 by (2) of List 2.2. I can use this to recouple X j with the other variables, so that the expression bounding the difference is a constant multiple of the denominator in the definition for r ′n ,m . The same argument as above applies in order to bound the difference between the numerators of (6.33) and (6.34).
Part (2):
I will sketch the proof. Let ζ < β, M = ⌊|k| β ⌋, andm = ⌊|k| ζ ⌋. As in Part (1), I can phrase the problem in terms of the random walk X n . Since there is no actual sign-flipping for X n (only coin tossing), I can remove the absolute values from X M , X 0 in the expression: where the approximation uses that there is a superpolynomially small probability that the event N ≤ M occurs because |X n − k| ≥ |k| 2 for some n ≤ M . By the same reasoning as in Part (1), I can replace the factors n(X r ) by n(k) in the arguments of the functions Π + with an error of order O (|k| ζ−1 ), and I can also remove the factors Π + 2 −1 n(X j )Θ(X j ) from the expectations for j ≤m. By the triangle inequality, I can bound the right side above by The first term will be superpolynomially close to zero for large |k|, since the distribution qm for Θ(Xm) will be superpolynomially close to the uniform distribution and by the symmetry argument used in the proof of Part (1). The probability of the eventN ∈ [n + 2, M ] when X n+1 = p for n <N and |p − k| ≤ 1 4 |k| will be O |k| β−1 , so the nth term on the lower line of (6.38) is equal to Finally, I have the equality I can use the exponential decay of j(v) to cap the jumps v by |k| ζ with a small error, so the expression above is approximately where I have followed the usual method for bounding expressions with integrals including j(v) and Π − .
6.2.2
Proof of Proposition 6.3 Part (1) of the lemma below makes another step toward showing that the duration between between successive sign-flip times τ n , τ n+1 is approximately an exponential distribution with mean ν −1 |K τn |.
Lemma 6.7. Let τ, ζ, β, k be as in the statement of Lemma 6.7. There exists a C > 0 such that all k where M := ⌊|k| β ⌋.
Proof.
Part (1): By Part (1) of Lemma 6.6 and the equivalence between P and P ′ , it is sufficient to show that for all ζ > 0. First, I will show that the probability P[N τ > n] decays exponentially for large n. Let F ⊂ N be the set of times n such that t n is a sign-flip or |K tn | jumps out of [ 1 2 |k|, 3 2 |k|]. I have the following relations where the second inequality uses the Markov property and Part (1) of Lemma 6.5 to bound P 3n − 1 / ∈ F K t 3n−3 . My purpose of removing the terms j = 0, 1 mod 3 in the product above was to accommodate the technical definition for a sign-flip, which requires information from the next Poisson time. This argument can be applied inductively to obtain a bound for P[N τ > n] that decays exponentially Nτ |k| ] must be uniformly bounded for 0 ≤ γ < B and k. The same arguments hold for the law P ′ . Since the probabilities P Nτ |k| > a and P ′ Nτ |k| > a decay exponentially in a ∈ R + with a rate that is uniform in k, I can take a cut-off |a| ≤ |k| ζ 2 as in the proof of Part (1) of Lemma 6.6. Let the set A M and the variational norm · Var,M be defined as in Lem 6.4. Notice that the event Nτ |k| > a is contained in the set A M for M = ⌊|k| 1+ ζ 2 ⌋. The variational norm for the difference χ(A M )P − χ(A M ) P ′ bounds the difference in probabilities for any event contained in A M , and thus the first inequality below holds: The second inequality is by Lem 6.4.
Part (2): If |K t M | − |K 0 | were a bounded random variable, then this result would follow immediately from the proof of Lem. 6.4. By Part (2) of Lemma 6.6, it is sufficient to prove that As in the proof of Lem 6.4, I will break down the difference into a sum of differences between the expectations E (n−1) , (6.41) The difference K tn − K t n−1 is equal to v n + ℓ n , where v n , ℓ n are the nth Lévy and lattice jumps respectively. For m ∈ Z, let Γ(m; p, v) be the closest element to m in I(p, v) (where ties can be assigned arbitrarily). It is convenient to split ℓ n into two parts Γ(ℓ n ; K t n−1 , v n ) and ℓ n − Γ(ℓ n ; K t n−1 , v n ) that are treated separately. Let K ′ t M ,n be the momentum at time t M if the nth lattice jump is replaced by Γ(ℓ n ; K t n−1 , v n ): K ′ t M ,n := K t M + K t n−1 − K tn + Γ(ℓ n ; K tn , v n ) + v n . I have the following bound for the nth term of the sum (6.41) when K t M is replaced by K ′ t M ,n in the expectation E (n−1) : where the densityq ′ n is defined as in the proof of Lem where the restriction of the integration in v to the set |v| ≤ |k| 1 2 will have a superpolynomially small error by the decay of j(v). I can apply the same estimates as in the proof of Lem 6.4 to show that (6.42) To complete the proof, I must bound the error of having substituted |K t M | by |K ′ t M ,n | in the expectation E (n−1) above: Again employing a cutoff |v| ≤ |k| 1 2 and using that the support ofq ′ n (p) is over |p| ∈ [ 1 2 |k|, 3 2 |k|], I can apply Part (2) of Lem. 4.1, since m − Γ(m; p, v) = dist m, I(k, v) . This yields a c > 0 such that for all k, Therefore, a single term from the sum (6.41) is O log(|k|) |k| ). Since there are M = ⌊|k| β ⌋ terms, the result follows.
Part (1): Similar to Part (2).
Part (2): The Poisson times t n can be thought of as the sum of independent mean-R −1 exponentials e m , m ∈ N: t n = n m=1 e m . It is sufficient to prove the statement of (2) with τ replaced by N τ and ν replaced by ν R = α. This can be seen with the short calculation: By the remark in the proof of Part (1) of Lem. 6.7, the expectation of e γ Nτ |k| is uniformly bounded for all k ∈ R and γ in 0 ≤ γ 0 ≤ γ 0 for small enough γ 0 > 0. Using the following expectation formula for a random variable X: da γ e aγ P X ≥ a , I can bound the difference in the case γ < γ 0 : The second inequality uses Part (1) of Lem. 6.7 for the first term. For the second term of the second line, the probability P Nτ |k| ≥ a is smaller than e −aγ 0 by the proof of Part (1) of Lem. 6.7.
Part (3): The quantity τ 0 drχ S(K r ) = S(K 0 ) sums the amount of time that K r spends up to time τ with the opposite sign of what it began with. By the definition of τ , K r cannot spend two consecutive Poisson times t n−1 , t n ≤ τ with S(K t n−1 ) = S(K tn ) = S(K 0 ), and thus S(K tn ) = S(K 0 ) implies S(K t n−1 ) = S(K tn ).
I can rewrite τ 0 drχ S(K r ) = S(K 0 ) in terms of the exponential waiting times e n as follows: where I do not count n = 1, since the sign of the momentum has been conditioned not to change at the first Poisson time. The probabilities in the summand on the third line are smaller than a constant multiple c > 0 of |k| −1 by the same argument as in the Part (2) of Lem. 6.5. To apply the argument from Lem. 6.5, I must use that |K r | ≥ 1 2 |k| for r ≤ τ . The last inequality is for |k| large enough, since E τ |k| approaches ν −1 at k → ∞ by Part (1).
Part (4): In the notation from Part (1), e Nτ +1 is the waiting time from τ to the next Poisson time.
The random variable τ is not a hitting time, since determining τ requires information from the time τ ′ := τ + e Nτ +1 , but τ ′ is a hitting time. The results for Parts (1) and (2) will also hold for τ ′ , since it includes only an extra contribution from an additional independent exponential random variable with mean R −1 . I will apply a submartingale argument using the fact that E s ≈ |K s | + O (|K s | −1 ) when |K s | ≫ 1. As before, let M r , A r be the Doob-Meyer components of the submartingale E r − E 0 . Recall that M r is a sum of two uncorrelated submartingales m r and M r − m r by Part (3) where the inequality follows because the event N τ ≥ n is contained by the intersection of N τ ≥ n − 1 and |K tn | ∈ 1 2 |k|, 3 2 |k| . The exponential random variables e n are independent of everything else, so they can be left outside the conditional expectation. By the same argument as in the proof of Lem. 6.2, Using the above and the result of Part (1), Now applying standard martingale arguments, I can bound M r − m r and A r .
The second inequality is Doob's for the first term, and for the second term I use that A τ ≤ στ |k| , which is a consequence of the identity σ = d dr M, M r +2E r d dr A r along with the fact that E r ≈ |K r | ≥ 1 2 |k| for r ∈ [0, τ ). By Chebyshev's inequality, it follows that the probability sup 0≤r≤τ M r − m r + A r 2 ≥ |k| 2 4 will be O log(|k|) |k| 2 . The final part of the story is m τ . The m r martingale is well-behaved, since the absolute values of its jumps are less than those of the Lévy process by Part (3) of Prop. 5.2 and the Lévy jumps have exponential tails. On the other hand, the moments of τ are finite by Part (1). The probability |m τ | > t 1 2 +δ , δ > 0 will decay superpolynomially, since τ has order O (|k|) for |k| ≫ 1.
Part (5): This follows by an extension of the analysis in Part (2) of Lemma 6.7, which I will not include.
Proof of Lemmas 3.2-3.4
For r, t ∈ R + , s ∈ [0, 1], and n ∈ N, let the processes M r , A r , Y (t) s , m (t) s , and N r ; the filtration F r ; and the times τ n be defined as in Sect. 3. In the proofs of this section, I will treat all the times τ m as if they occur through sign-flips rather than elaborating on the exceptional occurrences in which τ m = ς j for some j ∈ N or |K τm | / ∈ 1 2 |K τ m−1 |, 3 2 |K τ m−1 | . Although it is slightly incorrect to neglect those cases, the omission avoids some messy and unenlightening case considerations, and the estimates in the proofs below imply that those cases have negligible contribution. Whereas the number of sign-flips will be on the order O (t 1 2 ), the expected number of ς j over the time interval [0, t] has the bound σ 1 2 t 1 8 +ι for 0 < ι ≪ 1 by (6.2), and the event that |K τm | / ∈ 1 2 |K τ m−1 |, 3 2 |K τ m−1 | for some τ m ≤ t is unlikely to occur (see (6.63)).
Proof of Lem. 3.4. The proof of the Lindberg condition for M r follows from the analysis contained in (i) of the proof of Thm. 3.1. Also, the random variables t − 1 2 M st for t ∈ R + and s ∈ [0, 1] are uniformly integrable, since sup where the second inequality is by Part (2) The random variables t − 1 8 sup 0≤r≤t t − 1 2 |K r | 2 converge to zero as t → ∞, since t − 1 2 |K st | converges to a Brownian motion with respect to the uniform metric by Thm. 3.1. Moreover, by Part (2) of Prop. 6.3, there is a C > 0 such that for small enough γ > 0, the expectations E e γ ∆τn |Kτ n | F τ − n are smaller than C for all n and t ≫ 1. With Chebyshev's inequality, P ∆τn |Kτ n | > t . Applying this with an inductive argument using conditional expectations, then The second inequality follows because the number of sign-flips will be less than the number of Poisson times, and the number of Poisson times is typically ∝ t. The random variables m (t) s for s ∈ [0, 1], t ∈ R + are uniformly integrable, since the variances are uniformly bounded: The second inequality uses that Nt−1 n=1 ∆τ n ≤ t. For the first inequality, I use nested conditional expectations twice to replace the factor |∆τ n | 2 by 2 ν ∆τ n |K τn | in the sum. In doing so, I invoke Part (1) of Prop. 6.3 to get the approximations and |K τn | 2 ≈ ν|K τn |E |∆τ n | F τ − n . For each approximation, I multiply the upper bound by a factor of 2 to cover the error of the approximation. Uniform bounds of E sup 0≤r≤t t − m 2 |K r | m , m = 3, 4 for large t are obtained by using that |K r | ≤ E r and applying Doob's maximal inequality to the submartingale E r . Bounds on the fourth moments of E r are contained in Part (1) of Prop. 5.2. Lemma 6.8. Fix some 0 < ι < 3 16 in the definition for the times τ n , n ∈ N. As t → ∞, there is convergence in probability Nst−1 n=1 τn+∆τn τn drK r =⇒ 0.
Proof. Since my estimates will depend on |K r | ≫ 1, I first show that the contribution to Y (t) s that accumulates when |K r | ≪ |k| 3 8 will be negligible. Let δ > 0, then where the last inequality is for t large enough by Lem. 6.1 with ς 1 = 3 8 − ι, ς 2 = 2ι, ς 3 = 1 8 − ι, and ǫ = 2. Hence, for any δ, I can pick t large enough to make (6.45) arbitrarily small. It follows that Nst n=1 (τn+∆τn)∧st τn drK r =⇒ 0 as t → ∞, since the contribution during incursions is negligible. The upper bound (τ n + ∆τ n ) ∧ st for the integrals can be replaced by τ n + ∆τ n , since The right side goes to zero for large t by the discussion following (6.43). For the same reason, I can replace the upper bound of the sum by N st − 1.
For the remainder of the section, I will set ι = 0 in the definition for the excursion intervals [ς n−1 , ̟ n ), and thus indirectly for the definition of the times τ n . The rate of decay computed for various expressions in the proofs would be a little slower if I kept ι > 0, but ι can be chosen arbitrarily small anyway. The following proof will rely heavily on applications of Lem. 6.7 and Prop. 6.3. Moreover, I define Γ m,n as the Poisson time preceding Γ m,n when n ≥ 1 and Γ m,0 = τ m . Also, I will abuse notation by identifying F Γ m,0 with F τ − m . I will show the following convergences to zero in probability:
Proof of
is obtained by the right term in (ii) minus the expression in (iii) and differs from the expression for m (i). Over an interval r ∈ [τ m , τ m + ∆τ m ), the process K r has the same sign except for isolated Poisson times at which it jumps to the opposite sign and back again at the next Poisson time. My first step will be to bound the net effect of these rogue sign changes. The first inequality below follows because |K r | ∈ 1 2 |K τn |, 3 2 |K τn | : The second inequality follows by Part (3) of Prop. 6.3. For the third inequality, I have used that E ∆τ m F τ − m ≈ ν −1 |K τm | for large t by Part (1) of Prop. 6.3, and I doubled the bound to cover the error. The last inequality holds since the ∆τ m 's sum up to less than t. Hence, I can take the sign of K r to be constant over the time intervals [τ m , τ m + ∆τ m ).
It is useful to approximate the expressions in (i) by replacing K r with S(K r )E r , since E r is a submartingale with convenient analytic properties. By (6.46) and because |K r | = E r + O (|K r | −1 ), I have that Next, I will show that the second term on the second line of (6.47) goes to zero. The intervals between successive Poisson times have mean R −1 and are independent of everything else, which gives the equality below: The first inequality follows because ω(m) ≤ |K τm | β , and |K τm | ≥ t 3 8 . By Part (1) of Prop. 6.3, I have that E ∆τ m | K τm ≥ 1 2ν |K τm | for large enough t. The last inequality follows by removing the nested conditional expectations and Nt−1 m=1 ∆τ m < t. Recall that β ∈ (0, 1 3 ), so I have 5 8 + 3 8 β < 3 4 . For a δ ∈ (0, 1 8 − 3 8 β), then (6.48) implies that t − 3 4 +δ Nt−1 m=1 ∆ m converges to zero in probability for large t, and Thm. 3.1 implies that t − 1 2 −δ sup 0≤r≤t E r converges to zero in probability. Hence, the product goes to zero.
To bound the first term on the second line of (6.47), I use the Doob-Meyer decomposition E s = M s + A s and the triangle inequality to get . The last line in (6.49) is therefore less than σt − 5 8 . For a single pair m, n from the sum on the second line of (6.49), the following inequalities hold: Γm,n dr M r − M Γm,n 2 F Γm,n The first inequality above is Jensen's inequality. The last inequality holds since the predictable quadratic variation M, M r grows at a rate ≤ σ.
The second line of (6.49) is bounded by 2( σ 3 ) 1 2 multiplied by where Γ : R + → R + is the gamma function, and it should not to be confused with the times Γ n,m . The first inequality and equality above follow because ∆ m,n is a sum of ω(m) independent mean-R −1 exponential random variables. For the second inequality in (6.50), I have used that Nt−1 m=1 Lm−1 m=0 ∆ m,n ≤ t and ω m ≤ sup 0≤r≤t |K r | β . Finally, by Jensen's inequality where the first inequality is Jensen's and the second is Doob's. The equality is by Part (1) of Prop. 5.2.
(ii). The difference is bounded by The second term decays to zero by the argument in (i). By partial summation, the left sum in (6.51) is equal to where H m is defined as For technical reasons involving conditioning, it will be convenient to replace K Γm,n by K Γm,n as in the first approximation of (6.52). The sum t − 5 4 Nt−1 m=1 Lm−1 n=0 ∆ m,n |K Γm,n | − |K Γm,n | is easy to bound using |K r | ≈ E r and techniques used in (i). The second approximation in (6.52) replaces Γ m,Lm by τ m + ∆τ m in the expression, and the resulting error decays by the argument in (6.48) again.
Since the probability that sup 0≤r≤t |K r | ≥ ǫ −1 t ) with probability close to one. Introducing cutoff's will be useful to avoid problems with higher moments of K r . So far I have made only minor adjustments to the expression. The strategy to show that t − 5 4 sup 0≤s≤1 Nst−1 m=1 H m converges in probability to zero will be to prove: constant multiple of |K τm | 3 , where the leading term comes from the diagonal sum n = n ′ . By the same argument in (6.57), the sum of terms |K τm | 3 is bounded by which goes to zero for large t.
(iii). I will first show that there is a vanishing error in replacing the terms E ∆τ m F τ − m in the expression by ν −1 |K τm |. To bound the difference, I apply Part (1) of Prop. 6.3 to get the first and second inequalities below for C > 0 depending on my choice of 0 < ζ < 1 2 : The third inequality follows by removing the nested conditional expectations and Nt−1 m=1 ∆τ m < t. The fourth inequality is Jensen's, and the fifth employs |K r | ≤ E r and Doob's maximal inequality to the positive submartingale E r . The equality is Part (1) of Prop. 5.2.
I am left to bound the sum Nst−1 m=1 K τm |K τm |. Let G ⊂ R + be the set of all times τ m such that τ m ∈ [ς j , ̟ j+1 ) and τ m is an even-numbered flip time following ς j (and this includes τ m = ς j ). Denote the number of sign-flips in the interval (ς j , ̟ j+1 ) by n j . The sum of terms K τm |K τm | can be written as where I have neglected a single extra boundary term K τm |K τm | that may occur in the last incomplete excursion. I can immediately treat the boundary sum on the right side of (6.60). Since τ m + ∆τ m occurs during an incursion, I have that |K τm+∆τm | ≤ t 3 8 , and thus the first inequality below holds: The second inequality is from (6.2) in the proof of Lem. 6.1 and relies on the submartingale upcrossing inequality.
For the first sum on the right side of (6.60), it is convenient to write K τm |K τm | + K τ m+1 |K τ m+1 | as a sum of the two terms (S(K τm ) + S(K τm+∆τm )) K τm+∆τm 2 and − S(K τm ) |K τm+∆τm | 2 − |K τm | 2 . For any δ > 0, the factor t −1−δ sup 0≤r≤t |K r | 2 will go to zero for large t. By definition, S(K τm ) = S(K τm+∆τm ) occurs when |K τm+∆τm | jumps out of the interval 1 2 |K τm |, 3 2 |K τm | . By inserting nested conditional expectations, I have the equality below The first inequality is from |K τm | > t 3 8 . For the second and third inequalities, I have applied Part (1) of Prop. 6.3 and Nt−1 m=1 ∆τ m ≤ t, respectively. Hence, the right side of (6.63) will go to zero for any choice of δ < 3 8 . The sum of terms Nst−1 m=1 τm∈G −S(K τm ) |K τm+∆τm | 2 − |K τm | 2 on the right side of (6.61) is easier, since I can approximate |K r | by the submartingale E r = M r + A r as before. It is convenient to write |K τm+∆τm | 2 − |K τm | 2 as a sum of |K τm+∆τm | − |K τm | 2 and 2 |K τm+∆τm | − |K τm | |K τm | and to treat the two terms separately. For technical convenience, I will work with the expression on the right with the upper summand N st replaced by N st − 1. The difference is negligible by Lem. 3.4. I will show the following convergences in probability: (i). sup 0≤s≤1 t − 5 2 Nst−1 m=1 (i). For small ǫ, P sup 0≤r≤t |t − 1 2 K r | > ǫ −1 is small. Hence, forK r = K r χ(|K r | ≤ ǫ −1 ), is typically equal for all s ∈ [0, 1] to the same expression with K r replaced byK r . The advantage of working withK r is that it has arbitrarily many moments. The following is a martingale with respect to the filtration F (6.65) Using Part (1) of Prop. 6.3 for the first two inequalities below, the second moment of the above martingale is bounded through the inequalities I have multiplied the bounds in the first two inequalities by 2 to cover the error terms for the approximations of the moments E (∆τ m ) m F τ − m from Prop. 6.3. The last inequality follows by |K τm | ≤ ǫ −1 t 1 2 and removing the nested conditional expectations and Nt−1 m=1 ∆τ m < t. By Doob's maximal inequality, E sup 0≤s≤1 W (t),ǫ s 2 tends to zero for any fixed ǫ. By similar applications of Part (1) of Prop. 6.3 as above, the difference tends to zero. The difference between s -martingale and tends to zero by a similar (but simpler) argument as for (6.65). Finally, the process ν −1 t − 5 (ii). First, I show the sum over |K τm | 3 ∆τ m can be replaced by a sum over τm+∆τm τm dr|K r | 3 . This approximation reduces to some martingale analysis using that |K r | ≈ E r : Nst−1 m=1 τm+∆τm τm dr |K r | − |K τm | . (6.66) For any δ > 0, the random variables t −1−δ sup 0≤r≤t |K r | 2 tends to zero, since t − 1 2 |K st | converges to the absolute value of a Brownian motion in the uniform metric. The quantity |K r | in the sum on the right side of (6.66) can be replaced by E r with an error O (|K r | −1 ). Moreover, I have the inequality ν E sup where the first and second inequalities hold for large enough t. Thus, the second line of (6.67) vanishes as t → ∞. The martingale term is handled by similar arguments. By the above, I can work with the sum of the integrals τm+∆τm τm dr|K r | 3 . By the definition of the low energy incursions, the momentum has the upper bound |K r | ≤ 2t In most unbounded cases, there is a technical issue in checking that A λ : − i λ H − 2 −1 Ψ * (I) defines an m-accretive operator over an appropriate domain, and therefore generates a contractive semigroup on L 2 (R). However, since Ψ * (I) = RI, this is trivial given that H is self-adjoint, and I can factor e tA λ = e −t R 2 e − it λ H . Since Ψ and e where Φ (λ) t,ξ and ξ are defined as in statement (2) of the lemma. The semigroup Φ λ,t is strongly continuous, since Ψ is bounded and the group e − it λ H is strongly continuous. To check conservativity, it can be computed that Φ * λ,t (I) = I using Ψ * (I) = RI and that e − it λ H is unitary. The summation on the right side of (A.2) is equal to E Φ (λ) t,ξ (ρ) , where the expectation is with respect to the a Poisson process with rate R. The other stochastic representation for Φ λ,t is obtained by expanding Ψ in its integral form. The resulting integrals can be commuted, because the integrands are completely positive maps.
B The fiber decomposition for one-dimensional periodic Schrödinger equations
The reader is directed to [38] for a more detailed discussion of the structure of periodic Schrödinger equations. A Schrödinger Hamiltonian P 2 + V (X) with a period-2π potential satisfies e i2πP (P 2 + V (X))e −i2πP = P 2 + V (X + 2π) = P 2 + V (X).
Commuting with e i2πP implies that P 2 + V (X) must have invariant spaces corresponding to the spectral values for e i2πP (i.e. the unit circle in C). The Hilbert space H = L 2 (R) admits a fiber decomposition with fiber-maps sending f ∈ L 2 (R) to [f ] φ ∈ L 2 [−π, π) and formally defined according to the partial Fourier transform e −i2πnφ f (x + 2πn), x ∈ [−π, π).
For each φ ∈ [− 1 2 , 1 2 ), the self-adjoint operator H φ has compact resolvent. The eigenvalues are non-degenerate for φ = − 1 2 , 0 and are labeled progressively as E n,φ by a parameter n ∈ N called the band index. When φ = − 1 2 , 0, the pair (n, φ) is related in the extended-zone scheme to a parameter k ∈ R − 1 2 Z through the relations where S : R → {±1} is the sign function. The assignment of (n, φ) for φ ∈ {− 1 2 , 0} is a matter of convention that I am not concerned with, since the set {− 1 2 , 0} has measure zero in the direct integral (B.1). The dispersion relation E : R → R + is defined as E(k) = E n,φ for φ = − 1 2 , 0 and k related to (n, φ) as above, and I hold the convention that E(k) is symmetric and left-continuous for k ≥ 0.
For φ = − 1 2 , 0, let ψ n,φ be a normalized eigenvector for H φ with eigenvalue E n,φ . I can pick the eigenvectors ψ n,φ to vary continuously (and, in fact, smoothly [38, Thm.XIII.90]) as elements in L 2 [−π, π) for φ ∈ (−π, 0) and φ ∈ (0, π). Given f ∈ H, I can assign an extend-zone scheme representation f ∈ L 2 (R) through where n,φ are determined by k as above. Again, the assignment of f (k) for k ∈ 1 2 Z is arbitrary. In analogy with the position and momentum operators, I can define a self-adjoint operator P Q that acts on the domain f ∈ H | R dk| f (k)| 2 < ∞ as multiplication in the extended-zone scheme representation (P Q f )(k) = k f (k).
By standard operator calculus, I can define functions of P Q , and the Hamiltonian is given by H = E(P Q ). For k ∈ R − 1 2 Z, I define a tempered distribution |k Q such that for an element f ∈ S(R) in Schwartz space, where (n, φ) is determined by k. In the usual senses, I have the formal relations Q k ′ |k Q = δ(k ′ − k) and P Q |k Q = k|k Q .
B.2 The diagonal in the extended-zone scheme representation
A density matrix ρ ∈ B 1 H determines a probability density [ρ] D ∈ L 1 (R) corresponding to the distribution in the extended-zone scheme variable. Intuitively, this is given by the diagonal of the integral kernel ρ(k 1 , k 2 ) := Q k 1 |ρ|k 2 Q , although kernels are only defined a.e. R × R for Hilbert-Schmidt operators, so this does not offer a rigorous definition without some additional condition on the kernel such as continuity. For a rigorous definition, notice that there is a unique probability measure µ ρ on R such that for all g ∈ L ∞ (R) Tr g(P Q )ρ = R dµ ρ (k) g(k).
This follows by the Riesz representation theorem, since the left side is positive for g ≥ 0, bounded in absolute value by g ∞ , and equal to one for g = 1 R . By the continuity of the spectrum of P Q , the measure must be continuous with respect to Lebesgue measure, and I denote the Radon-Nikodym derivative of µ ρ by [ρ] D ∈ L 1 (R).
B.3 Invariant fibers for the Lindblad dynamical semigroup
The rate of kicks from the gas is invariant of the spatial location of the particle. The total dynamics is thus invariant under spatial shifts by 2π. For the dynamical maps Φ λ,t : B 1 L 2 (R) , this feature is expressed as the covariance Φ λ,t e i2πP ρe −i2πP = e i2πP Φ λ,t (ρ)e −i2πP . (B.3) Not surprisingly, this implies that the Banach space B 1 L 2 (R) decomposes into invariant fibers indexed by the Brillouin zone. This can be understood formally by the statement that the kernel values Q k|ρ|k + n + φ Q for k ∈ R, n ∈ Z do not interact dynamically for different φ ∈ [− 1 2 , 1 2 ). This holds also with the kets |k Q replaced by the standard momentum kets |k . Mathematically, it is easiest to discuss the invariant spaces for the adjoint semigroup Φ * λ,t . Let A T ⊂ B L 2 (R) be the algebra of all bounded operators commuting with e i2πP . By the analogous covariance property (B.3) for Φ * λ,t , it follows that Φ * t (A T ) ⊂ A T . More generally, the Banach spaces e iφX A T for φ ∈ [− 1 2 , 1 2 ) will also be invariant by the Weyl commutation relation e i2πP e iφX e −i2πP = e i2πφ e iφX .
C Dispersion without the Dirac comb
The elementary lemma below implies that the spatial dispersion for the particle scales as t 3 2 for times t ≫ 1 when the Dirac comb is not present. This scaling agrees with the classical case.
Proof. I will show pointwise convergence as t → ∞ for the characteristic functions ϕ λ,t of t 3 2 Q λ,t (t 3 2 x). The function ϕ λ,t can be written in terms of a trace formula involving ρ λ,t by the following: where X is the position operator. The right side is a special case of the quantum characteristic function for the matrix ρ λ,t . The quantum characteristic function has a closed factored form given by Tr e ivX+iqP ρ λ,t = Tr Φ * λ,t e ivX+iqP ρ = e t 0 dr φ(q+ 2 where v, q ∈ R and φ(q) := R dvj(v)e ivq . The second equality above can be seen through the expressions for Φ λ,t in Lem. A.1, Weyl's intertwining relations, and the relation e it λ P 2 f (X)e − it λ P 2 = f (X + 2t λ P ) for bounded functions f : R → C. With the above formula,
|
2012-03-05T08:10:03.000Z
|
2010-08-26T00:00:00.000
|
{
"year": 2010,
"sha1": "a41b70924ffd8b7f3595d0c3c422041e3cff32f6",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/1008.4502",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "a41b70924ffd8b7f3595d0c3c422041e3cff32f6",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics",
"Mathematics"
]
}
|
1591584
|
pes2o/s2orc
|
v3-fos-license
|
Membrane-Proximal Epitope Facilitates Efficient T Cell Synapse Formation by Anti-FcRH5/CD3 and Is a Requirement for Myeloma Cell Killing
Summary The anti-FcRH5/CD3 T cell-dependent bispecific antibody (TDB) targets the B cell lineage marker FcRH5 expressed in multiple myeloma (MM) tumor cells. We demonstrate that TDBs trigger T cell receptor activation by inducing target clustering and exclusion of CD45 phosphatase from the synapse. The dimensions of the target molecule play a key role in the efficiency of the synapse formation. The anti-FcRH5/CD3 TDB kills human plasma cells and patient-derived myeloma cells at picomolar concentrations and results in complete depletion of B cells and bone marrow plasma cells in cynomolgus monkeys. These data demonstrate the potential for the anti-FcRH5/CD3 TDB, alone or in combination with inhibition of PD-1/PD-L1 signaling, in the treatment of MM and other B cell malignancies.
Correspondence junttila.teemu@gene.com In Brief Li et al. report that the size and epitope location of the target play a key role in the efficiency of T cell activation induced by T cell-dependent bispecific antibodies (TDBs). They develop a TDB targeting FcRH5 expressed in all multiple myeloma tumor cells and show its potential in treating this disease.
INTRODUCTION
Multiple myeloma (MM) is an incurable malignancy of plasma cells characterized by dysregulated growth of abnormal plasma cells in the bone marrow and overproduction of intact monoclonal immunoglobulins that ultimately lead to clinical manifestations including skeletal lesions, renal failure, anemia, and hypercalcemia. Currently the backbone of MM treatment involves combinations of proteasome inhibitors (PIs), immunomodulators, and corticosteroids, with bone marrow transplantation as an additional option for eligible patients. Newer agents are being developed for the treatment of MM, including the monoclonal antibodies targeting CD38 (daratumomab) and SLAMF7 (elotuzumab). Nevertheless, despite progressive improvements in myeloma treatment, the mortality rate remains high and median survival remains less than 5 years (http://seer.cancer.gov/).
Heterogeneous disease biology and genetics, limited availability of predictive preclinical models, and a paucity of known myeloma-specific surface targets remain key challenges in myeloma drug development. FcRH5 (also known as FcRL5, IRTA2, or CD307) has been identified as an attractive B cell lineage-specific surface marker in myeloma (Elkins et al., 2012;Hatzivassiliou et al., 2001;Polson et al., 2006). High-affinity ligands and biological significance of FcRH5 are largely unknown. FcRH5 is expressed exclusively in the B cell lineage. Expression is detected as early as pre-B cells (Polson et al., 2006); however, unlike other B cell-specific surface proteins (e.g., CD20, CD19, and CD22), FcRH5 expression is retained in plasma cells. Analogous to its expression in normal plasma cells, FcRH5 is expressed by myeloma tumor cells. Finally, FcRH5 expression has been reported in several B cell malignancies (Ise et al., 2007;Li et al., 2008;Polson et al., 2006), suggesting broader applicability of this target in hematological malignancies.
Therapies that direct T cells to tumors, including adoptive transfer of genetically engineered T cells and T cell-dependent bispecific antibodies (TDBs) that selectively recruit T cells to tumor cells have been clinically validated in the treatment of B cell leukemias and lymphomas (Bargou et al., 2008;Sadelain, 2015) Significance Our study describes how CD3-bispecific antibody ''triggers'' intracellular T cell signaling and shows that the dimensions of the target molecule and epitope location play a key role in the efficiency of the synapse formation and subsequent T cell activation. These findings are important for future design of T cell-recruiting therapies. Using this information we developed and preclinically validated an anti-FcRH5/CD3 TDB as an immunotherapy for multiple myeloma. The anti-FcRH5/CD3 TDB is highly efficacious in the killing of myeloma cells and depletes bone marrow plasma cells in primates.
and have demonstrated promising activity in myeloma Rapoport et al., 2015). Our previous preclinical studies have validated full-length bispecific antibodies as an optimal TDB format with favorable drug-like properties including long serum half-life and low risk for anti-drug antibodies (Junttila et al., 2014;Sun et al., 2015).
TDBs activate T cells upon ligation with target-expressing cells resulting in potent target cell killing. However, the molecular mechanism that induces T cell activation has not been described in detail. The close proximity of cell membranes forms the basis of the kinetic segregation model for T cell receptor (TCR) triggering (Davis and van der Merwe, 2006). The model proposes that the exclusion of inhibitory molecules, such as CD45 phosphatase, from regions of close cell-cell apposition causes increased kinase activity and leads to phosphorylation of peptide-major histocompatibility complex (pMHC)-bound TCRs within this region. This then initiates receptor triggering and subsequent downstream T cell activation. Exclusion of CD45 from the synapse has been shown to be a passive process driven by the large size of the extracellular domain (James and Vale, 2012). If correct, the model predicts that a tumor target with a large extracellular domain may be suboptimal for synapse formation by CD3-bispecific antibodies. The size of the target protein has previously been linked to the killing activity of bispecific T cell engagers (BiTE) (Bluemel et al., 2010).
Given the potential of FcRH5 as a target for antibody-based therapeutics, the goal of the current study was to develop a TDB targeting FcRH5 (anti-FcRH5/CD3 TDB) for the treatment of MM. Moreover, we characterized the molecular events in the immunological synapse that lead to triggering of the TCR upon stimulation by CD3-bispecific antibodies.
RESULTS
Anti-FcRH5/CD3 TDB Induces Target Clustering and Exclusion of CD45 from the Synapse Resulting in TCR Triggering We characterized the molecular events in the immunological synapse that lead to triggering of the TCR upon stimulation by CD3-bispecific antibodies. To do this, we utilized a recently described reconstituted system that allows investigation of the initial events that lead to receptor activation in a controlled manner. HEK-T cells are non-immune cells that express the minimal components of the TCR signaling network required to drive receptor activation (James and Vale, 2012). Previous studies using this model have demonstrated that exclusion of CD45 phosphatase from the cell-cell interphase is both necessary and sufficient for TCR-pMHC-driven TCR triggering, supporting the kinetic segregation model (James and Vale, 2012). The HEK-T cells were conjugated with FcRH5-expressing cells in the presence of the CD3 bispecific antibody, and the relative intensities of CD45, FcRH5, and the fluorescently labeled TDB at the conjugate interface were imaged by confocal microscopy. ZAP70 is normally cytosolic but binds to phosphorylated ITAMs on the TCR when the receptor is bound to pMHC. The translocation of ZAP70 provides a convenient microscopy-based assay to evaluate TCR triggering (James and Vale, 2012). TDB binding to the membrane-proximal immunoglobulin (Ig) domain of FcRH5 (1G7; Figure 1A) led to efficient synapse formation, a robust clustering of FcRH5 in the cell-cell interaction site, and exclusion of the CD45 from the synapse ( Figure 1B). The combined result of FcRH5 clustering and exclusion of inhibitory molecules was TCR triggering illustrated by ZAP70 translocation to the cell interface ( Figure 1B).
We then compared the sequence of events in synapse formation at the cell-cell interface when driven by CD3-bispecific antibody binding (Figures 1A and 1B) with those at the interface caused by the pMHC-TCR interaction itself (Figures 1C and 1D). Overall, the mechanisms leading to TCR triggering, including target clustering, CD45 exclusion, and ZAP70-translocation, showed remarkable similarity between the CD3-bispecific antibody and pMHC-driven ligation ( Figures 1B and 1D).
The interface between the two cells can be visualized by taking a three-dimensional volume of the cell-cell conjugate formed by the TDB ( Figure 1E). The analysis demonstrated that the segregation of CD45 and the concomitant clustering of FcRH5 caused by TDB binding extended across the entire interface. The analysis also demonstrated a spatial mutual exclusion of the FcRH5 and the CD45 phosphatase. To confirm that this was the case, we analyzed a line profile across an equivalent interface and quantified the relative intensities of CD45, FcRH5, and the TDB ( Figure 1F). The quantitation of the fluorescent signal confirmed the strong co-localization between FcRH5 and the TDB, and the complete inverse correlation with CD45 fluorescence intensity. In summary, our results demonstrate that the CD3-bispecific antibody replicates the mechanism of the TCR/pMHC interaction-mediated TCR triggering inducing clustering of the target molecule and exclusion of CD45 from the T cell synapse resulting in activation of TCR signaling. CD45 exclusion has been described in the synapse induced by an Ep-CAM/CD3 BiTE (Offner et al., 2006), demonstrating that the conjugate interface and the molecular mechanism leading to T cell triggering share common features despite distinct structures of these molecules.
A Membrane-Proximal Epitope Is Required for Efficient Synapse Formation and Killing Activity of the Anti-FcRH5/CD3 TDB Similar to CD45, the extracellular domain of FcRH5 is large (550 and 835 amino acids, respectively) allowing us to test the effect of structural parameters of the tumor target on synapse formation and killing activity. We generated three proof-of-concept TDBs that bind to different regions of FcRH5 ( Figure 2A). TDB binding to the membrane-proximal Ig domain of FcRH5 (1G7) led to synapse formation significantly more efficiently compared with TDBs targeting central (10A8) or distal domains (gD) of the target ( Figure 2B), driving efficient CD45 exclusion (Figures 2B and 2C) and target clustering (Figures 2B and 2D) at the cellcell interface. The efficiency of forming a synapse was reflected in the ability of the TDB to induce TCR signaling and mediate killing by human T cells. Using healthy donor CD8 + T cells, 1G7-TDB resulted in very robust SLP76 phosphorylation (Figure 2E), which is indicative of potent TCR signaling, and mediated efficient killing of target cells ( Figure 2F; median effective concentration (EC 50 = 0.5 nM). In contrast, gD-TDB targeting of a membrane-distal epitope resulted in undetectable TCR signaling and was unable to mediate T cell killing. Monovalent binding affinities of antibodies used (K D = 12 versus 3 nM by Scatchard analysis for 1G7 and 10A8, respectively) do not explain the differences in the synapse formation or signaling.
We next confirmed that the TDB activity is driven primarily by the location of the epitope and the size of the extracellular domain (ECD) by targeting cells that express a heavily truncated FcRH5 that retains the 1G7 and gD epitopes ( Figure 2G). The activity of the proximal 1G7-TDB increased by 25-fold (Figure 2H; EC 50 = 20 pM) and the gD-TDB was able to effectively mediate killing of cells (EC 50 = 0.19 nM) when the obstruction caused by the ECD was removed. The possibility of differential target expression level being the cause for the activity difference between cell lines was ruled out by flow cytometry (Figure S1A). To confirm that the differences in the killing activity are related to the epitope rather than being properties of the specific TDB clones, we tested a total of five unique antibody clones targeting the membrane-proximal domain for the FcRH5 in bispecific format and could demonstrate that the activity of each clone was $20-fold higher compared with 10A8 ( Figure 2I). These findings were confirmed using the MOLP-2 myeloma cell line, which endogenously expresses FcRH5 at a low level comparable with that of MM patients. When T cells were retargeted to kill MOLP-2 cells, only membrane-proximal TDBs induced killing of the MOLP-2 cells. Targeting the mid region of FcRH5 would not lead to sufficiently potent TCR triggering to kill MOLP-2 cells ( Figure S1B). Together these data demonstrate that formation of the immunologic synapse by a CD3-bispecific antibody is dependent on the dimensions of the target molecule, and that rational epitope selection based on membrane proximity can overcome the hindrance caused by a large target protein that would otherwise lead to suboptimal synapse formation.
The Anti-FcRH5/CD3 TDB Induces Target-Dependent Cell Killing and T Cell Proliferation The anti-FcRH5 clone 1G7 was humanized and paired with a humanized anti-CD3 arm that is cross-reactive to the cynomolgus monkey to generate an anti-FcRH5/CD3 TDB. Anti-FcRH5/CD3 TDB is specific to FcRH5 and binds to the membrane-proximal domain of the target ( Figures S2A-S2B). Anti-FcRH5/CD3 TDB binds to FcRH5-expressing myeloma cell lines (MOLP-2), healthy donor B cells, bone marrow plasma cells, and primary myeloma tumor cells ( Figure S2C). Preclinical characterization of anti-FcRH5/CD3 TDB activity to confirm its mechanism of action was performed. Treatment of FcRH5-positive MOLP-2 cells with the anti-FcRH5/CD3 TDB and CD8 + or CD4 + T cells from healthy donors resulted in dose-dependent T cell activation and killing of MOLP-2 cells ( Figures 3A and 3B). Anti-FcRH5/ CD3 TDB also had an effect on T reg cell activation ( Figure S2D). The cytotoxic activity of TDB required FcRH5 expression on target cells and the potency correlated with FcRH5 expression level ( Figure 3C). Stimulation of effector T cells with the anti-FcRH5/CD3 TDB in the presence of target cells led to a robust proliferation of T cells, with 95% of the CD8 + cells undergoing as many as six cell divisions in 5 days ( Figure 3D).
Expression of FcRH5 in Normal Tissues and MM
FcRH5 is expressed in the B cell lineage starting from pre-B cells, but unlike most B cell markers, its expression is retained in plasma cells (Polson et al., 2006). Analysis of FcRH5 RNA expression in the Genotype-Tissue Expression (GTEx) sample set (Ardlie et al., 2015), consisting of 8,555 samples from 544 donors over 53 tissues, demonstrated expression in Epstein-Barr virus-transformed lymphocytes, spleen, and the terminal ileum of the small intestine ( Figure 4A). The RNA signal detected in spleen and intestine is likely derived from infiltrating B cells. Further analysis demonstrated a strong correlation with the expression profile of known B cell and plasma cell markers (CD19, CD20, and BCMA; Figure 4B). Overall the selective expression in B lineage cells and tissues predicts a favorable safety profile for this target. Expression of FcRH5 in CD138 + CD38 + MM cells and normal bone marrow plasma cells was evaluated by flow cytometry using bivalent 1G7 antibody. In all samples tested, all patientderived tumor cells, and all normal plasma cells expressed FcRH5, suggesting 100% prevalence in myeloma ( Figure 4C). Considerable inter-patient variability in expression level was detected in MM samples. Generally, FcRH5 expression levels in tu-mor cells was not significantly elevated compared with normal plasma cells, suggesting that developing a tumor cell-selective, normal plasma cell-sparing anti-FcRH5/CD3 TDB may not be feasible. FcRH5 expression level in normal B cells was observed to be lower compared with normal plasma cells and MM tumor cells ( Figures 4C and S2C), a similar finding as a previous report (Elkins et al., 2012). To understand the prevalence of expression in a broader patient population we performed a bioinformatic analysis of FcRH5 mRNA expression in CD138-purified plasma cells from 170 non-treated newly diagnosed MM patients and 6 healthy donors (microarray dataset GSE39754 from the NCBI Gene Expression Omnibus). All myeloma samples were positive for FcRH5 RNA ( Figure 4D). At the mRNA level, FcRH5 expression was significantly higher in myeloma samples compared with healthy donor samples. Expression of FcRH5 was higher in 155 out of 170 (91%) malignant samples compared with the highest expression level detected in the normal samples. Only one clinical value of interest, response to the USP7 inhibitor P5091, is available for the samples, and there was no significant correlation between this treatment response and FcRH5 expression (p = 0.62; ANOVA). The FcRH5 gene is located in the chromosomal breakpoint in 1q21 (Hatzivassiliou et al., 2001). Analysis of $20 primary MM biopsies demonstrated a significant association between FcRH5 RNA expression and 1q21 gain , demonstrating that the 1q21 gain can lead to FcRH5 overexpression in high-risk myeloma patients.
The selective expression for B lineage cells and tissues predicts a favorable safety profile for this target. Overall, 100% prevalence in myeloma, the predicted favorable safety profile and overexpression in high-risk patients indicate FcRH5 as a promising target for MM.
The Anti-FcRH5/CD3 TDB Mediates Potent Killing of Normal Plasma Cells and Patient-Derived Primary Myeloma Cells
The ability of the anti-FcRH5/CD3 TDB to kill plasma cells was analyzed by targeting bone marrow mononuclear cells (BMMCs) isolated from bone marrow aspirates of healthy donors (Figure 5A). Anti-FcRH5/CD3 TDB treatment induced potent dosedependent killing of normal plasma cells (EC 50 = 85-180 pM). Similarly robust cytotoxic activity was detected when BMMCs from MM patients were exposed to the anti-FcRH5/CD3 TDB ( Figure 5B). The anti-FcRH5/CD3 TDB demonstrated near-com-plete and highly potent killing of primary myeloma tumor cells (EC 50 = 60-1,200 pM).
As FcRH5 expression is variable in myeloma ( Figures 4C-4D) and anti-FcRH5/CD3 activity correlated with expression level ( Figure 3C), we investigated whether patients whose tumor cells expressed low levels of FcRH5 would be predicted to respond to the anti-FcRH5/CD3 TDB. The MOLP-2 myeloma cell line was identified as a benchmark cell line that has similar expression levels of FcRH5 as plasma cells and primary MM cells (Figure 4C). We also identified additional cell lines that express extremely low levels of FcRH5 and determined the number of FcRH5 molecules per cell using Scatchard analysis. The 1G7 binding sites in these cell lines ranged from 160 to 2,200 FcRH5 molecules per cell. Even with very low target copy number, anti-FcRH5/CD3 induced killing of all tested cell lines. Despite the limited maximal killing seen in the case of one cell line, EC 50 values for all cells were in the pM range (EC 50 = 2-230 pM, Figure 5C). Occupancy calculations indicate that as few as $50 TDB molecules (2% occupancy at MOLP-2; EC 50 = 58 pM) are sufficient to induce T cell activation and target cell apoptosis.
Anti-FcRH5/CD3 TDB Suppresses Growth of Established MOLP-2 Tumors in Mice Reconstituted with Human Immune Cells
Modeling anti-myeloma activity of the anti-FcRH5/CD3 TDB in mice is challenging since anti-CD3 antibodies do not cross-react with mouse CD3 and there is no mouse FcRH5 ortholog. Therefore we established a mouse model with a reconstituted human immune system by transplanting CD34 + -selected human hematopoietic stem cells into irradiated mice (humanized NOD/ SCID gamma [huNSG] mice). Human CD8 + cells harvested from spleens of huNSG mice were shown to be able to kill MOLP-2 cells in vitro with comparable efficiency as human peripheral CD8 + T cells from healthy donors ( Figure 6A). Twenty weeks post transplantation, huNSG mice were inoculated with 20 million MOLP-2 cells subcutaneously. Mice with established tumors (100-200 mm 3 ) were treated with a weekly intravenous dose of vehicle or 0.5 mg/kg of anti-FcRH5/CD3 TDB. Anti-FcRH5/CD3 TDB treatment resulted in tumor regression in all animals ( Figure 6B), indicating that anti-FcRH5/CD3 suppresses tumor growth in vivo.
Cynomolgus Monkey Is an Appropriate Preclinical
Model for the Anti-FcRH5/CD3 TDB Flow cytometry analysis was used to confirm both the reactivity of anti-FcRH5 1G7 to FcRH5 and expression of the target in B cells and plasma cells in cynomolgus monkey (Figures S3A-S3C). FcRH5 expression was detected throughout the B cell line-(E) qRT-PCR analysis of FcRH5 mRNA level in bone marrow biopsies from patient myeloma samples with or without 1q21 gain. mRNA expression level was calculated by the delta Ct (dCt) method. Statistical analysis was performed using a Mann-Whitney U test. The box represents the 25th to 75th percentiles and median. Whiskers represent the range of minimum and maximum values. (F) Representative images of FISH analysis on primary multiple myeloma biopsies showing normal diploid of 1q21 (red; top panel) and a mixture of three to six copies of 1q21 (bottom). A tumor sample was identified as 1q21 gain when >20% of the tumor cells scored had three or more copies of the 1q21.3 locus. Scale bars, 10 mm in length. age in a similar manner to human (Polson et al., 2006), and the anti-FcRH5/CD3 TDB binds to FcRH5 and CD3 with comparable affinity. Treatment with anti-FcRH5/CD3 TDB of target cells expressing primate FcRH5 or MOLP-2 cells expressing human FcRH5 resulted in robust killing using peripheral T cells from either human or cynomolgus monkey with comparable efficiency ( Figures S3D and S3E). Adding anti-FcRH5/CD3 TDB to peripheral blood mononuclear cell (PBMC)/BMMC samples from cynomolgus monkey also resulted in a dose-dependent and robust killing of B cells ( Figure S3F) and bone marrow plasma cells ( Figure S3G). These results validate the cynomolgus monkey as an appropriate safety and efficacy model for the anti-FcRH5/CD3 TDB.
Anti-FcRH5/CD3 TDB Has a Long Serum Half-Life in Cynomolgus Monkeys
A single-dose study was conducted to evaluate efficacy, pharmacokinetic (PK) and pharmacodynamic (PD) properties of the anti-FcRH5/CD3 TDB in cynomolgus monkeys. Monkeys were treated with a single intravenous dose with slow infusion of vehicle or 1, 2, or 4 mg/kg of anti-FcRH5/CD3 TDB. Blood samples were collected for analysis of PK/PD response and cytokines for 7 days after the anti-FcRH5/CD3 TDB was administered. The study was terminated at day 8.
The anti-FcRH5/CD3 TDB demonstrated dose-proportional exposure (C max and area under the curve) between 1 and 4 mg/kg and with clearance ranging from 29 to 33 mL/day/kg in all cohorts ( Figure 7A). The C max at the 4 mg/kg dose level was 129 mg/mL, $2,000-fold higher than that required to reach the in vitro killing EC 50 for human plasma cells and MOLP-2. Receptor occupancy calculations suggested near-total saturation of FcRH5 engagement on peripheral blood B cells at C max at all dose levels ( Figure S4A). In summary, these results demonstrate that the anti-FcRH5/CD3 TDB has PK characteristics that support an intermittent weekly or less frequent dosing schedule.
Anti-FcRH5/CD3 TDB Depletes B Cells and Bone Marrow Plasma Cells in Cynomolgus Monkeys
Flow cytometry of peripheral blood demonstrated a robust pharmacologic effect at all dose levels. Anti-FcRH5/CD3 treatment resulted in T cell activation and a transient decrease in T cells likely reflecting a margination response within 24 hr ( Figures 7B, 7C, S4B, and S4C). CD4 + and CD8 + T cells recovered to baseline levels by the end of the study. In contrast, B cells remained to be undetectable in blood 7 days after anti-FcRH5/ CD3 TDB administration indicating that they were depleted by exposure to the molecule ( Figure 7D). All dose levels resulted in complete depletion of B cells in spleen and bone marrow (Figures 7E and 7F). Anti-FcRH5/CD3 TDB treatment also induced a robust, dose-dependent depletion of B cells from the lymph nodes ( Figures 7G and S4D).
Depletion of primate bone marrow plasma cells is a key efficacy endpoint in the preclinical development of the anti-FcRH5/CD3 TDB. Complete depletion of plasma cells following anti-FcRH5/CD3 TDB treatment was observed in the animals treated at 2 and 4 mg/kg ( Figure 7H). Anti-FcRH5/CD3 treatment also resulted in a dose-dependent reduction of IgG of 37% and 44% in the 2 and 4 mg/kg groups, respectively ( Figure 7I), an expected secondary outcome resulting from plasma cell depletion. These figures compare favorably with calculations based on the half-life of IgG suggesting that complete depletion of plasma cells would decrease IgG level $30%-40% by day 7. In summary, anti-FcRH5/CD3 TDB induced a robust PD response in cynomolgus monkeys consistent with its mechanism of action. Complete plasma cell depletion with subsequent decrease in serum IgG levels provides compelling evidence of TDB activity in the bone marrow microenvironment.
Anti-FcRH5/CD3 TDB Induces Transient Cytokine Release in Cynomolgus Monkey
Consistent with the mechanism of action, treatment with the anti-FcRH5/CD3 TDB at all dose levels tested induced a rapid, generally mild/moderate cytokine release ( Figure S5), including elevations in IL-6, IL-5, IFN-g, IL-2, IL-13, and MCP-1 as well as the anti-inflammatory response peaking at 2-6 hr. All cytokines returned to normal baseline levels within 24 hr. In summary, no signs of severe or prolonged cytokine release were observed at dose levels that are expected to saturate target and sufficient for complete depletion of B cells and plasma cells. A single dose at the %4 mg/kg dose level did not result in detectable neurotoxicity.
PD-L1 Blockade Enhances the Activity of Anti-FcRH5/ CD3 TDB A strong TCR stimulation signal normally leads to immunosuppressive feedback that limits T cell activity. Signaling through the PD-1/PD-L1 axis is a critical component of this feedback and a therapeutically validated immune escape mechanism in several tumor indications. PD-L1 is frequently expressed by myeloma cells (Gorgun et al., 2015), and the signaling axis may limit T cell activity in myeloma patients.
In vitro stimulation (48 hr) of human healthy donor CD8 + cells with anti-FcRH5/CD3 TDB in the presence of target-expressing cells resulted in significant PD-1 induction in T cells ( Figure 8A). This feedback signal was also observed in vivo. A significant increase in the number of PD-1-positive T cells was observed when cynomolgus monkeys were treated with anti-FcRH5/CD3 TDB at all tested dose levels. PD-1 induction was detected in both CD8 + and CD4 + T cells in blood, spleen, lymph nodes, and bone marrow (Figures 8B and S6). The ability of anti-FcRH5/ CD3 TDB-primed CD8 + T cells to kill PD-L1-expressing target cells was modest ( Figure 8C); however, blocking PD-1/PD-L1 signaling using an anti-PD-L1 antibody in combination with anti-FcRH5/CD3 significantly increased the efficiency of killing (Figure 8C). Together, these results show that anti-FcRH5/CD3 TDB-mediated activation of T cells leads to induction of PD-1 in T cells in vitro and in vivo. Although PD-1/PD-L1 signaling can limit anti-FcRH5/CD3 TDB-mediated killing, PD-L1 blockade can overcome this inhibition and lead to enhanced activity of the anti-FcRH5/CD3 TDB. Our data provide strong scientific rationale for the clinical assessment of anti-FcRH5/CD3 TDB combined with anti-PD-L1 blockade in MM patients.
DISCUSSION
Several CD3-bispecific antibodies or antibody fragmentbased molecules are in clinical development and clinical proof-of-concept has been established by blinatumomab in hematological malignancies (Bargou et al., 2008). Optimally, the CD3-bispecific molecules show extremely potent preclinical activity regardless of the target molecule or the molecule format. In our studies, using a plethora of tumor targets and antibody clones, we have detected extreme variability in the activity of the molecules that could not explained by the affinities of the molecules. In this study we describe the molecular events in the synapse induced by a CD3-bispecific full-length antibody and show that the events closely follow the principles of kinetic segregation, and are not significantly different compared with TCR activation by pMHC ligation. Our studies confirm that FcRH5 is suboptimal for bispecific antibody-mediated T cell triggering due to its large ECD that interferes with efficient synapse formation. By using antibodies that bind to various different epitopes we show that the efficiency of synapse formation correlates with the proximity of binding epitope to cell membrane. The difference between the epitopes in the end result (killing of myeloma cells) is dramatic: membrane-proximal antibodies kill with picomolar concentrations, whereas more distal antibodies are essentially inert. Potency in the context of BiTE-mediated killing has been correlated with similar structural features of the target molecule (Bluemel et al., 2010) complementing our molecular studies of synapse and mechanism of T cell triggering.
Anti-FcRH5/CD3 TDB killed patient-derived myeloma cells and healthy donor-derived plasma cells at picomolar concentrations. Non-clinical pharmacology studies with anti-FcRH5/CD3 TDB in mice are challenging. First, anti-FcRH5/CD3 TDB is not reactive to mouse CD3. Second, an FcRH5 ortholog does not exist in the mouse; thus genetically engineered mouse models are not suitable for testing the molecule. The only available mouse tumor model that can be used is human myeloma cell line (MOLP-2) xenografted to immune-compromised mice supplemented with human T cells. As these tumors are grafted subcutaneously, the MOLP-2 xenografts do not model activity in the bone marrow environment. A further clear limitation of this type of xenograft model is that the immune system engrafted in the mice likely does not exactly recapitulate the adult human im-mune system, potentially not accounting for the contribution of other cell types, such as T reg cells, to TDB treatment. In contrast, cynomolgus monkeys can be used as a compelling efficacy model to demonstrate in vivo activity in the bone marrow compartment. A single dose of the anti-FcRH5/CD3 TDB depleted plasma cells and B cells from tissues and led to expected reduction of serum IgG levels.
Cytokine release has been reported consistently across CD3targeting bispecific molecule platforms with variable frequency and severity. Clinical cytokine release syndrome (CRS) has been reported in CD19-targeting agents, e.g., blinatumomab and CD19 CAR-T cells. As expected, the anti-FcRH5/CD3 TDB induced mild/moderate cytokine release immediately after dose administration, but no extensive or prolonged cytokine release was observed in cynomolgus monkeys. The predictive value of the detected cytokine levels in primates to CRS in myeloma patients is unclear. However, several potential mitigation strategies are available for cytokine-related adverse effects (dose fractionation, corticosteroids, or IL-6 signaling blockers).
The anti-FcRH5/CD3 TDB is predicted to be broadly active in myeloma as the prevalence of the target expression is 100%, and as few as $200 copies of FcRH5 on a MM cell are sufficient to induce tumor cell killing. In addition to myeloma, evidence of frequent FcRH5 expression has been reported in multiple B cell malignancies such as chronic lymphocytic leukemia, mantle cell lymphoma, diffuse large B cell lymphoma, and follicular lymphoma (Ise et al., 2007;Li et al., 2008;Polson et al., 2006). This suggests a more general applicability for anti-FcRH5/CD3 in B cell-mediated malignancies in addition to MM.
Gain or amplification of chromosome 1q21 is one of the most commonly detected genetic abnormalities in MM and is considered a predictive marker of aggressive disease (Boyd et al., 2012). FcRH5 was originally identified in cloning of this chromosomal region and has been shown to be deregulated in cell lines with 1q21 abnormalities (Hatzivassiliou et al., 2001). Our analysis of primary myeloma samples demonstrates that FcRH5 mRNA is elevated in myeloma patients with 1q21 gain. Although it is unlikely that FcRH5 plays a functional role in the myeloma progression, its overexpression provides an intriguing diagnostic hypothesis for the anti-FcRH5/CD3 TDB. Our in vitro assays demonstrate a correlation between target expression level and activity of the molecule, suggesting that MM with gain of 1q21 may be exquisitely sensitive to the anti-FcRH5/CD3 TDB, thus providing clinical benefit to a patient population that otherwise has very limited treatment options.
T cell activation by anti-FcRH5/CD3 TDB induced upregulation of PD-1 in T cells. This negative feedback signaling has been detected with other CD3 targeting bispecific molecules regardless of the format of the molecule (Bacac et al., 2016;Junttila et al., 2014;Osada et al., 2015), and is likely a class effect for molecules with this mechanism of action. As PD-1/PD-L1 signaling can inhibit the killing activity of T cells, optimal clinical use of T cell-recruiting bispecific antibodies may include combination with inhibitors for this pathway.
Antibodies
Antibodies were from BD Biosciences unless otherwise mentioned. Anti-human PD-1 was from Affymetrix. Goat anti-human IgG and goat anti-mouse IgG were from Jackson ImmunoResearch. Anti-PC-FITC (clone Vs38c) was from Dako. SLP-76 from Cell Signaling Technology. p-SLP76 (Ser376) and anti-PD-L1 antibodies were generated at Genentech.
Fluorescent Labeling of Antibodies
For detection of FcRH5 from MM samples and healthy donor plasma and B cells by flow cytometry, anti-FcRH5 antibody 1G7 was labeled with phycoerythrin by SouthernBiotech. For microscopy, the TDBs were labeled with Alexa Fluor 647 using the appropriate protein labeling kit (Thermo Fisher Scientific) according to the manufacturer's instructions. TDBs were dialyzed into PBS, pH 7.2, prior to labeling and a dye/protein ratio of $4 was routinely achieved.
Stable Cell Lines
To evaluate the immunological synapse formation, SVT2 cells were infected with retrovirus encoding full-length FcRH5 with N-terminal gD expression tag or truncated FcRH5 (deletion of AA1-744) with N-terminal gD tag. To evaluate the target dependency of TDB killing, FOX-NY cells were infected with lentivirus encoding full-length FcRH5 and single-cell-derived clones with differential expression level of FcRH5 were selected. To evaluate the effect of PD-1/PD-L1 signaling to TDB activity, 293 cells were infected with lentivirus encoding FcRH5 followed by transfection of human PD-L1 encoding plasmid using Lipofectamine (Invitrogen).
Vectors and Transient Transfection for Microscopy
FcRH5 with N-terminal gD expression tag was fused to the fluorescent protein mRuby2 by first inserting FcRH5 into the pHR-SIN lentiviral vector, before ligating the mRuby2 DNA sequence into this vector, creating pHR-FcRH5-mRuby2. The SFFV promoter in this vector was subsequently replaced with the mHSP promoter, creating pHRI-FcRH5-mRuby2, which utilizes a weaker promoter than pHR, allowing more physiological expression levels of FcRH5-Ruby. Vectors expressing LCK, ZAP70, CSK/CBP, and CD45 have been described previously (James and Vale, 2012). The CD45 construct used was either the RO isoform or a construct containing the cytoplasmic domain of CD45 with the transmembrane and ECDs of CD43, which is known to mimic the function of CD45. Vectors were transiently transfected at appropriate ratios using GeneJuice (Novagen), Cells were used in experiments 24-48 hr after transfection.
Microscopy Imaging and Analysis
To image cell conjugates, 3 3 10 5 cells were harvested from culture and resuspended in 100 mL of 20 nM TDB in RPMI-1640 (without phenol red). After 20-30 min incubation to allow cell conjugation, cells were washed with PBS, resuspended in DMEM gfp2 imaging medium (Evrogen) and added to 35 mm imaging dishes (Mattek). An Andor spinning disc confocal microscope system was used to image the cells at 37 C. All images were analyzed and all presented images were manipulated in an equivalent manner using ImageJ. The presented images were background subtracted and then cropped to focus on the pair of cells and the contrast was optimized. The degree of protein clustering and segregation was determined by using the intensity of fluorescently labeled proteins in the plasma membrane. The plasma membrane was selected by manually drawing a line and the average fluorescence intensity of the plasma membrane within the cell-cell interface was divided by the average fluorescence intensity of the plasma membrane outside the cell-cell interface to calculate the degree of clustering or segregation. To generate an image of the interface of a pair of cells conjugated by TDBs from a z stack, the image stack was first deconvolved and then cropped to highlight the interface region using Huygens software.
Production of TDBs
Full-length bispecific antibodies were produced as described elsewhere (Junttila et al., 2014;Sun et al., 2015). In brief, the two half antibodies containing the ''knob'' or the ''hole'' mutations in the CH3 domains were expressed by transient transfection of CHO cells and then affinity purified with Protein A. Equal amounts of the two half antibodies were incubated with a 200 molar excess of reduced glutathione at pH 8.5 overnight at 32 C to drive the formation of the knob-hole disulfide bonds. The assembled bispecific antibody was purified from contaminants through hydrophobic interaction chromatography.
In Vitro Cytotoxicity Assay: Human Plasma Cells and Primary MM Samples Human BMMCs from MM patients were procured from Conversant Bio. Human bone marrow aspirates of healthy donors were procured from AllCells. All human biospecimens were collected, processed, and distributed in full ethical and regulatory compliance with the sites from which human biospecimens were collected. This includes independent ethical review, institutional review board approval (where appropriate), independent regulatory review, and ethical review for collection sites. All sites were located in the US and the EU. In vitro experiments using human healthy donor blood, bone marrow, or vendor-procured live tumor material are routinely performed at Genentech and do not require approval by an internal ethical review committee.
Human bone marrow aspirates of healthy donors were diluted in PBS and BMMCs were isolated by conventional gradient separation (Lymphoprep, STEMCELL). Flow cytometry viability assay was used to test the effect of 72 hr anti-FcRH5/CD3 TDB treatment on BMMC plasma cells. Myeloma BMMCs were mixed with freshly isolated healthy donor CD8 + T cells and co-culture treated with anti-FcRH5/CD3 TDB for 72 hr. PI-negative CD38 + CD138 + cells were counted by flow cytometry. The killing activity was calculated as: {(number of live target cells without TDB À number of live target cells with TDB)/(number of live target cells without TDB)} 3 100%.
RNA Expression in Normal Tissues and Myeloma
Samples mRNA expression was analyzed in the GTEx RNA sequencing sample set (Ardlie et al., 2015) consisting of 8,555 samples from 544 donors over 53 tissues and in the NCBI GEO: GSE39754 dataset (Affymetrix GeneChip Human Exon 1.0 ST Array) from the NCBI GEO repository (Chauhan et al., 2012). CD138 purified plasma cell samples in the NCBI GEO: GSE39754 dataset represent newly diagnosed patients with MM before initiation of primary treatment.
RNA Isolation, cDNA Synthesis, and Gene Expression Analysis Total RNA was extracted from decalcified formalin-fixed paraffin-embedded (FFPE) bone marrow biopsy tissues collected from MM patients. Two reference genes, SDHA and VPS33B, were evaluated for each sample and used to calculate expression of FcRH5. Gene expression of FcRH5 was determined by using the delta Ct (dCt) method (Ct gene of interest -Ct geometric mean of reference genes ). Detailed description of method in the supplement.
Cytogenetic Fluorescent In Situ Hybridization 1q21 + copy control 1 fluorescent in situ hybridization (FISH) probe (Biocare Medical; previously CymoGen Dx) was used to analyze the 1q21 region. The 1q21 probe covers the chromosomal band 1q21.3 while the control probe is located in the peri-centromeric 1p12 region of chromosome 1. FISH analysis on FFPE tissue was performed as described previously (Koeppen et al., 2014;O'Brien et al., 2008). A minimum of 100 non-overlapping tumor cells from each sample was enumerated. Cutoff of gain was 3 or more copies in >20% of the tumor cells. Detailed description of method in the supplement.
huNSG/MOLP-2 Mouse Xenograft Model
All mouse experimental procedures conformed to the guiding principles of the American Physiology Society and were approved by Genentech's Institutional Animal Care and Use Committee (IACUC). Female huNSG mice were obtained from The Jackson Laboratory. Animals were inoculated with 20 million MOLP-2 tumor cells in Hank's balanced salt solution/Matrigel, subcutaneously. Treatments were administered intravenously once a week 34. Detailed description of method in the supplement.
Cynomolgus Monkey Study
The PK and PD properties of anti-FcRH5/CD3 TDB were evaluated in naive, male cynomolgus monkeys (cynos) at Charles River Laboratories (CRL). Cynos were treated with a single-dose, intravenous infusion (1 hr) of vehicle, 1, 2, or 4 mg/kg anti-FcRH5/anti-CD3 TDB blood samples were collected by venipuncture via the femoral vein pre-study and at selected time points for 7 days after dosing for analyses of hematology, serum chemistry, coagulation, and PK and PD endpoints (cytokines, flow cytometry of T lymphocytes, B lymphocytes, activated T lymphocytes, and PD-1 and circulating cyno IgG). Bone marrow was collected in anesthetized animals by aspiration from the humerus pre-study and on day 8 for evaluation of B lymphocytes and plasma cells by flow cytometry. The study was terminated at day 8. All procedures were approved by the CRL IACUC and were performed in compliance with the Animal Welfare Act, the Guide for the Care and Use of Laboratory Animals, and the Office of Laboratory Animal welfare.
Anti-FcRH5/CD3 Pharmacokinetics in Cyno
Anti-FcRH5/CD3 TDB in serum were determined by generic ELISA. Sheep anti-human IgG antibody was used as the capturing reagent and sheep antihuman IgG conjugated to horseradish peroxidase (HRP) was used as the detection reagent. Serum concentration-time data from available samples were analyzed by a non-compartmental with IV bolus input model (Phoenix WinNonlin, Version 6.3; Pharsight Corporation). Nominal sample collection time and nominal dose concentrations were used in the data analysis. All TK analyses were based on individual animal data.
ELISA Analysis for Cyno IgG Level
Total cyno serum IgG was quantified using standard colorimetric-based sandwiched ELISA. A goat anti-monkey IgG (Bethyl Laboratories, A140-202A) and a HRP conjugated goat anti-monkey IgG (Bethyl Laboratories, A140-202P) were used as the capture and detection antibody, respectively. Cyno IgG (Cell Sciences CSI20163A) was used as the protein quantification standard.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures and six figures and can be found with this article online at http://dx.doi.org/ 10.1016/j.ccell.2017.02.001.
|
2018-04-03T02:18:40.745Z
|
2017-03-13T00:00:00.000
|
{
"year": 2017,
"sha1": "8f54fe437ef11ee0f8ccf1ea3eed004f7c772252",
"oa_license": "CCBY",
"oa_url": "http://www.cell.com/article/S1535610817300156/pdf",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "5addcf646a97dffbed4e4eb663fdd947558b2362",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Biology",
"Medicine"
]
}
|
262075713
|
pes2o/s2orc
|
v3-fos-license
|
Risk of temporal lobe necrosis between proton beam and volumetric modulated arc therapies in patients with different head and neck cancers
Background To investigate the frequency of temporal lobe necrosis (TLN) soon after radiotherapy (RT) and identify differences among patients with various types of head and neck cancer (HNC) and between different RT methods. Methods We retrospectively reviewed 483 patients with HNC who had completed RT in our hospital after January, 2015. These patients were followed-up at the radio-oncology department and received contrast-enhanced magnetic resonance imaging (MRI) or computed tomography (CT) to identify metastases or recurrence of cancer at regular intervals. Meanwhile, the occurrence of TLN, graded according to the Common Terminology Criteria for Adverse Events V5.0, was recorded. We categorized the patients into nasopharyngeal carcinoma (NPC) and non-NPC groups and compared the cumulative occurrence of TLN between the groups using Kaplan–Meier and Cox regression analyses. We further compared the cumulative occurrence of TLN between proton beam therapy (PBT) and volumetric modulated arc therapy (VMAT) in patients with any HNC, NPC, and non-NPC HNC. Results Compared with the non-NPC group, the NPC group had a higher frequency of TLN (5.6% vs. 0.4%, p < 0.01) and were more commonly associated with TLN in the Kaplan–Meier analysis (p < 0.01) and the Cox regression model after covariates were adjusted for (adjusted hazard ratio: 13.35, 95% confidence interval: 1.37–130.61) during the follow-up period. Furthermore, the frequency of TLN was similar between patients receiving PBT and those receiving VMAT (PBT vs. VMAT: 4.7% vs. 6.3%, p = 0.76). Kaplan–Meier analysis revealed that the accumulated risks of TLN were similar between PBT and VMAT in patients with any HNC (p = 0.44), NPC (p = 0.84), and non-NPC HNC (p = 0.70). Conclusion Our study demonstrated that patients with NPC are susceptible to TLN during the early period after RT. In addition, PBT may be associated with an equivalent risk of TLN when compared with VMAT in patients with NPC or other HNCs. Supplementary Information The online version contains supplementary material available at 10.1186/s13014-023-02344-y.
Introduction
Radiotherapy (RT) is the standard treatment for head and neck cancer (HNC).Late effects on "bystander" organs have become increasingly prevalent in survivors of HNC.Among radiation injuries with long-term consequences, radiation vasculopathy and hypothyroidism are relatively common [1][2][3].Patients treated with radical RT for HNC may receive significant radiation doses to large volumes of brain tissue.Patients with HNC, particularly advanced-stage nasopharyngeal carcinoma (NPC) may be at increased risk of adverse late brain effects including temporal lobe necrosis (TLN) after concurrent chemoradiation therapy [4,5].The mechanisms of TLN include microvascular injury, cell injury, and inflammatory and free radical injury [6].Patients with TLN may develop epilepsy or cognitive decline.However, available data regarding the long-term prognosis of TLN remain scarce.
The application of proton beam therapy (PBT) in the treatment of HNC has been growing in the past few years [7].The physical properties of the Bragg peak allow for precise dose delivery, thus minimizing or preventing an exit dose from affecting normal tissues located beyond the target.PBT is much more sensitive to tissue density than photon therapy.So far, intensitymodulated radiation therapy (IMRT) and PBT have been demonstrated to have similar treatment effects on local tumor control in patients with NPC [8,9].The accuracy achieved with PBT allows further widening of the therapeutic window, as dose escalation for radioresistant tumors is possible without jeopardizing treatment tolerance [10].Comparative studies of PBT versus photon radiotherapy have demonstrated diverse results [11][12][13][14][15].Some studies showed a lower regional toxicity in patients receiving PBT for various cancers [8,9,12,13,16], results corroborated by the biggest prospective PBT series published to date [12,13].However, other studies focusing on HNC only demonstrated that the risk of TLN was higher in patients receiving PBT [14,15], which could be resulted from various biological and treatment uncertainties [16].Studies mentioned above used different techniques treating different cancer around the brain, which were unable to answer the question.We used a PBT system with intensity modulation in our institution [17].Intensity modulated proton therapy have been reported to improve several aspects of dose profile [18], and has become the preferred technique for HNC The current study focus on HNC treated by PBT with intensity modulation.Our work investigated the frequency of TLN during the early period after RT for HNC and the association between novel PBT and the risk of TLN in these patients.
Patient recruitment and demographic data
We reviewed patients with HNC who received RT after January 1, 2015, which was the date on which PBT was established in our institution.All enrolled patients received regular follow-ups with brain imaging studies to detect the presence of tumor recurrence at the radio-oncology department of our hospital.Patients were referred to the neurology department for evaluation of the risk of radiation vasculopathy or other neurovascular complications.We focused on TLN during the early period after RT.The median latency for TLN detection was 30 months with a range between 6 and 56 months after RT in previous study [19]; therefore, patients with a time interval of > 5 years between the date of their final RT and the date of their latest follow-up were excluded in this study.Data on demographic characteristics and RT of all recruited patients were recorded.Laboratory data, such as glycated hemoglobin, high and low-density lipoprotein cholesterol, and high-sensitivity C reactive protein (hs-CRP) levels, were registered (Fig. 1).The study was approved by the Ethics Institutional Review Board of our hospital (202101981B0 and 202200464B0).
Cancer and RT data
We recorded the pathological types, locations, and tumor stages of HNC, method of RT [volumetric modulated arc therapy (VMAT) or PBT], interval from latest RT to study enrollment, and accumulated total doses of RT of each patient.Both PBT and VMAT treatment plans were generated using the Eclipse planning system (version 13.7; Varian Medical Systems, Palo Alto, CA, USA).Pencil beam scanning with 3 beam angles was used for fullfield PBT plans [9].The planning target volume in VMAT was expansion of 3-5 mm around the clinical target.For PBT, worst-case robust optimization was used for CTV coverage without PTV expansion.Dose constraints and algorithm used for optimization in the PBT were the same as those used in VMAT.The relative biological equivalent value of 1.1 was assumed for protons and calculated during optimization [20].Prescription consisted of 6000-6600 centi-grays (cGy) 30-33 fractions for postoperative radiotherapy and 6996 cGy in 33 fractions for primary radiotherapy over 6-7 weeks (5 fractions per week).All targets were treated simultaneously [21].
Grouping
Patients were categorized into two groups according to cancer pathology: an NPC group and a non-NPC group.The non-NPC group consisted of patients with oral cavity, oropharyngeal, laryngeal, and hypopharyngeal cancers.The NPC and non-NPC groups were further subdivided into PBT and VMAT subgroups based on the method of RT (Fig. 1).
Follow-ups
Patients received regular follow-up at the radio-oncology department at least every 6 months and received at least one comprehensive neurological examination at the neurology department.Patients received contrast-enhanced magnetic resonance imaging (MRI) and/or computed tomography (CT) every 6 months to identify potential distal metastasis or recurrence of the primary tumor and to record the occurrence, type, and progression of TLN.CT of the head and neck region was performed using a multidetector CT scanner.Thin-slice CT images (3 mm in thickness) were reconstructed in the axial, coronal, and sagittal directions.MRI was performed using a 1.5 or 3.0 Tesla MR scanner by using a standard head and neck coil.Pre-contrast T1-weighted images, T2-weighted fat-saturated images, and postcontrast T1-weighted fat-saturated images were acquired in the axial, coronal, and sagittal planes.The section thicknesses were 5 mm with a 2.5mm intersection gap in the axial plane and 4 mm with a 1-mm gap in the sagittal and coronal planes.Screening for TLN-associated neurological symptoms and detailed neurological examinations were performed during outpatient visits to the neurology department.
Outcomes
The primary outcome in this study was the diagnosis of TLN.TLN was classified according to the pattern into three types: edema, enhancement, and necrosis [22].The severity of TLN was graded according to the Common Terminology Criteria for Adverse Events (CTCAE) V5.0 [23].
Radiation dose to the temporal lobe and dosimetry analysis
For patients who had TLN and were treated by photons radiotherapy (XRT) with single VMAT plan, dose statistics of bilateral temporal lobes were retrieved from VMAT plan used for treatment.In other patients with TLN, either treated by photons or protons, had adaptive replanning during treatment course to adaptive anatomical change during fractionated radiotherapy.The primary plans were registered to second plan with dose deformation and summation by VelocityTM oncology imaging informatics system (VARIAN, Palo Alto).We compared the dosimetry data among NPC patients receiving different method of RT and NPC patients with/without TLN.We also compared the dosimetry data of the diseased versus non-diseased lobes in NPC patients with TLN.Since only a small part of temporal lobe would be irradiated by PBT, maximal dose (Dmax) and highest dose delivered to small specific volume (0.5, 1, and 2 c.c.) should be more comparable.Shroeder and colleagues reported that D1cc is the most important dosimetry factor correlated to TLN [15].The QUANTEC analysis also showed that maximal dose is predictive for TLN [24].So Dmax, D0.5 cc, D1cc, and D2cc were reported and analyzed.On the other hand, spread of low and intermediate radiation dose is the main disadvantage of VMAT.Therefore, the volumes of temporal lobe that exposed to radiation doses of 2000 (V20Gy) and 5000 cGy (V50Gy) or more were also analyzed [25][26][27].
Statistical analysis
We used SPSS 22.0 (SPSS, Chicago, IL, USA) to analyze the clinical data.Parameters are presented as mean ± standard deviation or frequency (%).We used an independent two-sample t test to identify differences in the continuous variables between the study groups.Categorical variables were compared using a chi-square test or Fisher's exact test.Radiation dose and follow-up duration were analyzed using Mann-Whitney tests and were presented as median (quartile 1, quartile 3).Intergroup differences in event risk and time to TLN were compared using Kaplan-Meier analysis and the log-rank test.Cox regression analysis was used to compare risks between the study groups, in which covariates [age, tumor staging (T3 or T4), cancer type (NPC or non-NPC), RT doses, time interval to the end of RT, type of RT (PBT or VMAT), and reirradiation) were adjusted.Continuous variables (RT dose) were analyzed as continuous data instead of being categorized into groups.We used a univariate Cox regression model and a multivariable Cox regression model with backward selection to investigate the correlations between these risk factors, NPC, and the risk of TLN.Since the definition of tumor stage is different between NPC and non-NPC, which could extraordinary be a potential confounding factor to the study results.Cochran-Mantel-Haenszel test was used to assess the conditional independence of cancer types associated with TLN.Significance was indicated by p < 0.05.
Results
A total of 652 patients with HNC who finished RT after January 1, 2015, received regular post-RT follow-ups, and had evaluation for neurovascular complications screening by neurologists in our hospital were initially reviewed.Of these patients, 167 whose time interval between the date of their final RT and the date of their latest follow-up was > 5 years were excluded.Two patients with missing information were also excluded.In total, 483 patients with HNC were recruited and categorized into two groups: 198 (41.0%) were in the NPC group and 285 (59.0%) were in the non-NPC group.In the NPC group, 86 (43.4%) patients received PBT and 112 (56.6%) patients received VMAT.In the non-NPC group, 46 (16.1%) patients received PBT and 239 (83.9%) patients received VMAT (Fig. 1).
The mean interval from RT completion to the last follow-up was slightly longer in the NPC group [NPC vs. non-NPC: 35 (21, 46) vs. 30 (18,42) months, p = 0.03].Compared with the non-NPC group, the patients in the NPC group were more likely to receive a diagnosis of TLN (NPC vs. non-NPC: 5.6% vs. 0.4%, p < 0.01; Table 2) and were at a higher risk of developing TLN during the follow-up according to the Kaplan-Meier analysis (logrank test p < 0.01; Fig. 2) and the Cox regression model (adjusted hazard ratio: 13.35, 95% confidence interval: 1.37-130.61,p = 0.03) after adjustment for covariables.Although the definition of advanced T stage could be different between NPC and non-NPC, both the NPC patients in the T0-T2 stage (NPC vs. non-NPC: 4.7% vs. 0.0%, p = 0.04; Additional file 1: Table 1) and those in the T3-T4 stage (NPC vs. non-NPC: 6.4% vs. 0.6%, p < 0.01) yielded higher risks of TLN development.Cochran-Mantel-Haenszel test (p < 0.01) also showed that patients with NPC remained have a higher odds ratio of the TLN occurrence after excluding the effect of tumor stages.
Regarding the influence of the method of RT, the frequency of TLN was similar between patients receiving PBT and VMAT (PBT vs. VMAT: 4.7% vs. 6.3%, p = 0.76).Kaplan-Meier analysis indicated that the accumulated risks of developing TLN between patients who received PBT and those who received VMAT were similar among patients with any HNC (log-rank test: p = 0.44; Fig. 3A), patients with NPC (log-rank test: p = 0.84; Fig. 3B), and patients with non-NPC HNC (log-rank test: p = 0.70; Fig. 3C).The radiation doses to the temporal lobes in the NPC patients receiving different method of RT were shown in the Table 3.Compared to NPC patients undergoing VMAT therapy, NPC patients receiving PBT had significantly lower V20Gy and V50Gy of the temporal lobes.D1cc and D2cc of left temporal lobe were also significantly lower in patients under PBT (Table 3).
In total, 12 patients (2.5%) developed TLN.The details of the RT dose-volume parameters of the 12 patients who developed TLN are listed in Additional file 1: Table 2.Among these patients, 3 (25%) had bilateral involvement, 11 (91.7%)developed an edematous pattern, and 6 (50%) had contrast enhanced lesions.Only 1 (8.3%) patient had a seizure before the latest follow-up.Ten (83.3%) of the 12 TLN patients were grade 1 (asymptomatic) in severity based on CTCAE v5.0 criteria, while 2 (Patient 4 and 6, 16.7%;Additional file 1: Table 2) were graded 2 and received steroid treatment.The dose distributions of select patients with NPC with and without TLN receiving different RT methods are shown in the Supplementary Figure .Of notes, all dosimetry parameters (Dmax, D0.5 cc, D1cc, D2cc, V50Gy, and V20Gy) in bilateral temporal lobes of the NPC patients with TLN were higher compared to other NPC patients without TLN (Table 3).Totally 15 necrotic temporal lobes were observed in NPC patients with TLN, all the dosimetry parameters except V20Gy of the necrotic temporal lobes were higher than those of the non-necrotic temporal lobes with statistical significance (Additional file 1: Table 3).
Discussion
Our study demonstrated that patients with NPC were more susceptible to TLN development during the early period after RT than patients with non-NPC HNC.In addition, PBT is not associated with an increased risk of TLN compared with VMAT in patients with any HNC, NPC, or non-NPC HNC.The incidence rate of TLN varied from 4.6 to 7% during the early period after RT [26].NPC is known to be a risk factor for TLN; however, the prevalence of TLN in patients with non-NPC HNC remains uncertain.For patients with non-NPC HNCs, exposure of high radiation dose over temporal lobes is not rare.The treatment volume of RT would extend to skull base or temporal lobes if there is extensive perineural invasion in pathological samples or if patients are at risk of retropharyngeal lymph node metastasis.One previous study investigated the association of RT with the development of TLN in patients with parotid gland tumors [28]; One of our patients with parotid gland tumor also developed TLN during follow-up.However, our results revealed the risk of TLN is relatively lower in patients with non-NPC HNC.It is possible that the prescription dose to skull base for the conditions described above could be definitively lower than the radiation dose prescribed for the patients with overt tumor extension to skull base.Therefore, screening for TLN in non-NPC population may not be cost-effective.For patients with NPC, an important question is whether the RT method moderates the risk of TLN, particularly in the current era, where several advanced RT methods are available.A study demonstrated that IMRT significantly reduced the risk of TLN compared with two-dimensional RT [29].One study revealed that the prevalence rate of 2-year TLN was approximately 4.6% in patients with HNC after PBT [26].Most of the studies exploring the cause of TLN showed that highest dose in a small area is more critical in the prediction of TLN than the total volumes of intermediate radiation dose.However, the most significant advantage of PBT could be the decreasing "dose spillage", which means the intermediate and low dose region spreading everywhere in the body.This could be the reason why PBT may not reduce the risk of TLN in these studies [26].However, another study demonstrated that the risk of developing TLN could be even higher in patients with NPC who received PBT (10%) than in those who received VMAT (4%), and this may arouse the attention to possible adverse effects of PBT use in NPC patients [11].Compared with the patients undergoing VMAT treatment, our NPC patients who received PBT had similar risks of TLN development during follow-up, and had lower V20Gy and V50Gy radiation doses bilaterally.Our findings might mitigate some safety concerns that individuals may have regarded PBT as a risk factor for TLN in patients with NPC, and may provide some evidence of post-RT complication for the share-decision-making process before choosing the treatment strategy.Advanced T staging, RT dose-volume parameters, radiation fraction, and reirradiation are common risk factors for TLN development [30].However, there were still 4 NPC patients with T2 stage developed TLN in our study.In these patients, the extents of parapharyngeal space invasion of the primary tumor were close to the cranial cavity, and therefore the radio-oncologists in-charge extended the treatment volume into the cranial cavity and along the foramen ovale to eradicate possible subclinical tumor extension.The radiation doses to the lowest part of the temporal lobes were more than 7000 cGy in these patients, and this could be a possible reason why these NPC patients with T2 stage still developed TLN after RT.The current study showed that higher radiation doses to the temporal lobe increased the risk of TLN, suggesting the radiation dose is a more determinative factor rather than the method of radiation.However, this study was not able to identify the most decisive dose level for predicting TLN.Different methods of dose-response analysis would be required to solve this problem.To avoid TLN, we should still follow the QUANTEC report for dose prescription to reduce maximal dose to temporal lobe currently [25].
The median follow-up time was approximately 3.43 years in TLN patients after RT [31].Among patients who developed TLN, 21.5% developed symptomatic epilepsy during follow-up [31].Another retrospective review found that 88.5% of patients developed radiation necrosis after RT, of which 16.5% developed epilepsy [32].However, the incidence rates of TLN and seizure in our study were lower than those in previous studies.The pathogenesis of RT-related epilepsy can be explained by various potential mechanisms: (1) RT can cause endothelial cell and blood-brain barrier damage and aggravate brain edema, and (2) inflammatory cytokines released after RT may be associated with epilepsy development [33].Detection of epilepsy is important, because seizures can lead to status epilepticus, sudden unexplained death in epilepsy, conscious disturbance, cognitive impairment, and significant morbidity.Notably, temporal lobe epilepsy should theoretically be higher in patients after RT due to its anatomical correlation.Our patient with epilepsy developed an altered sense of smell, which was considered as grade 1 seizure based on CTCAE v5.0 [23]; this symptom and interictal discharges on electroencephalogram disappeared after treatment with levetiracetam.Seizure symptoms are sometimes vague in these TLN patients.Cooperation between different medical specialties may help to discover the occurrence of seizure.Cognitive impairment is another complication in patients with TLN [34][35][36].Cerebral microbleeds, hippocampal atrophy, and TLN have been proposed as possible mechanisms leading to cognitive impairment after RT [34][35][36]; however, our retrospective study was unable to obtain comprehensive neuropsychiatric assessments in these patients.Future prospective studies may help to illustrate whether the method of RT affects cognitive or hippocampus preservation [37,38].
A previous study showed that TLN usually begins with white matter lesions, followed by contrastenhanced lesions and cystic formation.These lesions can be progressed, regressed, static, or fluctuated [39].In our patients, 92% and 50% exhibited white matter and contrast-enhanced lesions, but none had cystic formation, which is consistent with previous reports.Treatment of TLN remains an area of uncertainty.Glucocorticoid, anticoagulant, pentoxifylline with vitamin E, bevacizumab, hyperbaric oxygen therapy, and surgical treatment have all been proposed for TLN treatment [40][41][42].The efficacy of steroid treatment varies, and previous reports have shown that 19.4% and 15.3% of patients with TLN achieve complete response and partial response, respectively, after oral steroid treatment [30,43].However, steroid use is not a standard treatment of TLN in our hospital; thus, this study was unable to investigate the responses to the aforementioned treatments.
This study has several limitations.First, the sample size of enrolled patients may not be sufficient to yield results of clinical significance.Our hospital is a large medical center in East Asia providing PBT to patients with HNC and may provide valuable pilot data in this field.A multicenter prospective study is warranted to evaluate the accuracy of our findings [38].Second, both contrast-enhanced CT and MRI studies were arranged to screen for cancer recurrence in these patients.MRI was more sensitive to TLN diagnosis than was CT.The retrospective nature of our study may have underestimated the true prevalence of TLN in patients with HNC.Third, dosimetry data were not available for all patients.This may have affected the determination of the causal relationship between TLN and the method of RT.However, the purpose of this study was to investigate the associations of TLN with different types of HNC and RT methods instead of determining causal relationships.Fourth, a wide range of confounding factors was present.Selection and reporting bias due to influencing factors and missing data may also have confounded the results of the present study.Finally, the generalizability of our results to patients of other ethnicities remains uncertain.In addition, the capacity of PBT may still be limited and therefore patient selection remains crucial.Also, access is not equitable in most countries and it needs to be clear that in a wide territory with a high incidence of NPC, patients can still be treated with gold-standard XRT without jeopardizing their outcomes [44,45].
Conclusions
Our single-center retrospective study demonstrated that patients with NPC were more susceptible to TLN development during the early period after RT than patients with non-NPC HNC.Among advanced RT techniques, PBT was associated with an equivalent risk of TLN compared with VMAT in patients with NPC.These findings could potentially mitigate the safety concerns regarding the use of PBT in these patients.
Fig. 2
Fig. 2 Cumulative risks of TLN in patients with NPC and non-NPC head and neck cancer.Kaplan-Meier analysis comparing the cumulative risks of TLN between the NPC and non-NPC groups.Frequency of TLN was significantly higher in the NPC group than in the non-NPC group.NPC Nasopharyngeal carcinoma; TLN Temporal lobe necrosis
Fig. 3 3
Fig. 3 Cumulative risks of TLN of patients with any HNC, NPC, and non-NPC HNC receiving PBT or VMAT.Kaplan-Meier analysis comparing the cumulative risks of TLN between PBT and VMAT in patients with any HNC (A), patients with NPC (B), and patients with non-NPC HNC (C).The risks of TLN were comparable between PBT and VMAT in all the subgroups.HNC Head and neck cancer; NPC Nasopharyngeal carcinoma; PBT Proton beam therapy; TLN Temporal lobe necrosis; VMAT Volumetric modulated arc therapy
Table 1
Baseline characteristics of the study groups CRP c-reactive protein; NPC Nasopharyngeal carcinoma; RT Radiotherapy † p < 0.05 # Data were examined using Mann-Whitney exams and presented as median (quartile 1, quartile 3)Other data were examined using two-sample t tests (continuous variables) and chi-square tests (categorical variables)
Table 2
Primary study outcome and associated factors of TLN in Cox regression models TLN Temporal lobe necrosis; CI Confidence interval; HR Hazard ratio; NPC Nasopharyngeal carcinoma # Data was analyzed using the Chi-square test † Data was analyzed using a multivariable Cox regression model with backward selection
|
2023-09-21T13:58:28.385Z
|
2023-09-21T00:00:00.000
|
{
"year": 2023,
"sha1": "d369302973e70ce22fd234b513fb9c8a86626e12",
"oa_license": "CCBY",
"oa_url": "https://ro-journal.biomedcentral.com/counter/pdf/10.1186/s13014-023-02344-y",
"oa_status": "GOLD",
"pdf_src": "Springer",
"pdf_hash": "dda576702bbbb1b06f98ef27af5d02ad193e28ac",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
87662467
|
pes2o/s2orc
|
v3-fos-license
|
19th and 20th Century Plant Hunters
The latter part of the 19th and the first several decades of the 20th century can be described as a ‘‘golden age’’ for plant exploration and collecting. During the initial years of this period, agricultural scientists from the United States and elsewhere devoted considerable resources to collecting potential new crops for farmers as well as superior plants or cultivars of the species that farmers were already growing. Over time, there was a shift toward collecting unadapted germplasm, or raw material that possessed traits that plant breeders and other scientists could use for cultivar improvement and other types of research. Although many institutions and individuals were involved in plant collecting during this period, the creation of the U.S. Department of Agriculture (USDA) Office of Seed and Plant Introduction in 1898, resulted in the largest single program devoted to plant exploration. This office employed many individuals, including David Fairchild, P.H. Dorsett, Frank Meyer, Walter Swingle, and Wilson Popenoe. These and many other individuals collected—and introduced into the United States—seeds and plants of thousands of fruits, vegetables, nuts, ornamentals, cereals, forages, oilseeds, and other types of crops. Although the mission of most of the plant explorations during this period was to collect any plants that appeared interesting or potentially useful, others focused on collecting targeted species. Much of the material collected during this era is still maintained by the U.S. National Plant Germplasm System (NPGS), and much more of it shows up in the pedigrees of cultivars grown by farmers and gardeners today. In addition to collecting plants for immediate and future use, scientists of this era, such as Nicolai I. Vavilov and Jack Harlan, contributed greatly to the understanding of the evolution of plants and plant genetic diversity, and the interdependence of plants and civilization. Plant introductions (PIs) into the United States have come from thousands of sources: immigrants, tourists, government officials. The first official PI activities by the United States government were conducted in the late 1700s and early 1800s when naval officers and diplomats sent germplasm back to the United States. The latter part of the 19th and the first several decades of the 20th century can be described as a ‘‘golden age’’ for plant exploration and collecting. During the initial years of this period, agricultural scientists from the United States and elsewhere devoted considerable resources to collecting potential new crops for farmers, as well as superior plants or cultivars of species that farmers were already growing. This paper will discuss the earliest USDA explorers through those of the middle of the 20th century. MIDDLE TO LATE 1800S: FIRST UNITED STATES PLANT EXPLORERS The first official federal United States foreign plant exploration was conducted in 1858. On this trip, Robert Fortune collected tea, Camellia sinensis L., from China. This trip occurred before the establishment of the U.S. Department of Agriculture (USDA). Mr. Fortune was sponsored by the U.S. Commissioner of Patents. Private collections and botanical gardens were interested in exotic plants. Professor Charles S. Sargent was the first Director of the Arnold Arboretum, Jamaica Plain, Mass. In 1892, he inaugurated the Arboretum’s interest in eastern Asia by visiting Japan. The Arnold Arboretum subsequently provided the foundation plant material for the introduction of many Japanese plants into American gardens. Ernest H. Wilson, also under the employ of the Arnold Arboretum, undertook four major expeditions to China between 1899 and 1911. He spent most of the second decade of the 20th century exploring for plants in Japan, Korea, and Taiwan. Although many institutions and individuals were involved in plant collecting during this period, the creation of the USDA Section of Foreign Seed and Plant Introduction (SPI) resulted in the largest single program devoted to plant collection and exploration in the United States (White and Briggs, 1989). James ‘‘Tama Jim’’ Wilson, previously the Director of the Iowa Experiment Station, was appointed to be Secretary of the USDA. He appointed David Fairchild, who was at that time 22 years old, to be Chief of the Section and Professor Niles Ebbesen Hansen to be the first official plant explorer for the USDA. Secretary Wilson stated, ‘‘I have twelve thousand men under me, but none who knows how to work like Hansen. There is only one Hansen.’’ Indeed Professor Hansen was a very productive collector. Fairchild (1944) states that, ‘‘Almost the moment I was settled in our first location, tons of seeds and plants began to pour in upon me from Russia. Hansen felt that he had been sent out to collect, and he collected everything and collected it in quantity. It was all most embarrassing, as no provision whatever had been made to take care of his shipments.’’ Professor Hansen was supported on three USDA plant collecting trips to Russia for 10 months in 1897, to Turkistan for 6 months in 1906, and in Europe and Siberia for 9 months in 1908. On these trips, he collected alfalfa and other forage germplasm, plus apples, pears, crabapples, and small fruits. The plant material that he collected was rich in cold-hardiness and drought tolerance. EARLY 1900S PLANT EXPLORERS Mark A. Carleton, a classmate of David Fairchild’s, was a USDA wheat breeder who was working to discover rust resistance in cereals. He had been studying Russian, and in the course of his reading he concluded that the Black Lands of Russia could contain resistant strains of wheat. Fairchild asked him to travel to Russia in 1898 to collect wheat, barley, and other cereals. Carleton brought back ‘‘durum wheat,’’ a special class of early wheat, which prefers dry conditions and is very productive. The spectacular success of durum wheat focused public attention on the value of PI. It demonstrated that farmers could achieve high profits as a result of the introduction of a new variety. Two years after the introduction of durum wheat, 60,000 bushels were produced; only 5 years later, 20,000,000 bushels were grown (Fairchild, 1944). Fairchild himself traveled to distant reaches of every continent but Antarctica between 1896 and 1933. He searched for fruits, vegetables, cereals, and other economic plants, including mangos, dates, nectarines, and flowering cherries. During his career he and those who worked under him introduced more than 80,000 accessions into the United States. Many USDA employees from the Section of Plant Pathology were also involved with The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper therefore must be hereby marked advertisement solely to indicate this fact. Supervisory Horticulturist, Retired. Research Leader/Curator; E-mail khummer@ars-grin.gov HORTSCIENCE VOL. 42(2) APRIL 2007 197 PI. Some of these included Walter T. Swingle, Frank N. Meyer, Fredrick Wilson Popenoe, and P. Howard Dorsett. As Harry V. Harlan (Harlan and Harlan, 2005) describes, ‘‘The pioneers of plant industry were as amazing a lot as those who had pushed back our frontiers. They were striking individualists with no two having the same background or any distant resemblances.each brilliant in one way or another—but where in God’s world did they find them?’’ Walter T. Swingle delivered his first scientific paper at the age of 16. He received a B.S. from Kansas State University at 19 years of age. He joined USDA in 1891 and became an agricultural explorer in 1898. He made many trips to southern Europe, North Africa, the Near East, China, Japan, and the Philippines. He introduced many citrus varieties and unusual species relatives. Frank N. Meyer traveled over a thousand miles along the routes of Marco Polo across Asia (Cunningham, 1984) He spent 13 years in eastern and northern China collecting samples of any plant that looked interesting. However, he primarily focused on what he thought would be economically useful. He sent over 2500 PIs to the United States as live cuttings and thousands of sacks of seed. His introductions included plants from alfalfa to zoysia grass. During his long periods of travel, Meyer had to deal with threatening robber brigands, wolf packs, revolutionary soldiers on the prowl, interpreters who refused to go on, carts that shattered on lonely mountainsides, inadequate food, poor shelter, and vermin that infested Asian villages. Meyer returned ‘‘tired, but satisfied,’’ from a successful expedition in the high mountains of Shanxi Province in northern China on 26 Feb. 1908. Frank Meyer unexpectedly died while traveling on the Yangtze River in China during his final expedition in 1918. The inscription on the tombstone on Meyer’s grave at Bubbling Wells Cemetery, Shanghai, China, reads, ‘‘In the glorious luxuriance of the hundred plants he takes delight.’’ The Meyer Medal, an annual award presented by the Crop Science Society of America, originated in 1918. It is presented in honor of Frank N. Meyer, as a result of funds bequeathed to the Plant Introduction Section in his will. This medal is presented to recognize an individual with stellar meritorious work in the field of Plant Genetic Resources. Frederick Wilson Popenoe was a USDA agricultural explorer from 1912 to 1925. He had an eye for economic crops with potential value to the United States. He specialized in South American crops and was largely responsible for popularizing avocadoes in America. Palemon Howard Dorsett, another USDA agricultural explorer, collected nearly 3000 soybean accessions in China, Japan, Manchuria, and Korea between 1929 and 1930. Explorations by Dorsett and others during the years 1906–1932 yielded 5534 soybean accessions. None of the original 114 PIs from 1906 to 1917 were saved; all were distributed to users. At that time, no germplasm repositories existed for long term maintenance of originally collected germplasm. Now, 24 PIderived cultivars are part of the United States collection (GRIN, 2005). Harry V. Harlan joined the USDA, Bureau of Plant Industry, in 1910 and was ‘‘young, eager, and thirsty for new discoveries.’’ Harlan collected barley in Russia, Ethiopia, North Africa, and other areas. ‘‘The desire to explore and to collect is as inherent in us and as much a part of us as toenails and far less transitory than hair. Who has not felt the urge to go places and do things? Any boy old enough to walk spends most of his time exploring, and his mother is in a perpetual state of dismay over the treasure he collects. My urge to explore the world and the desire to collect barleys need no word of explanation. I wanted to go so badly that I would not have blamed those in power if they had harbored a suspicion that my eagerness was as much for personal pleasure as for the good of the cause.’’ Harlan stated that ‘‘I never for a moment doubted the material thus obtained would justify the expense of the trip many times over.’’ He thought that, ‘‘The whole record of American agriculture was made up of a succession of crops that came from a handful of seed-crops with annual revenues to the farmers valued in the hundreds of millions.’’ Harlan was also aware that, ‘‘Unfortunately too few Americans realize what introductions of foreign crops has meant to the American way of life.’’ Although we have been discussing American plant explorers, one Russian, Nicolai Ivanovich Vavilov, was extremely instrumental at defining the need for plant exploration as a foundation for agricultural development. Between 1916 and 1940, Vavilov made expeditions to five continents in search of new agricultural plants and confirmation of his theories on plant genetic diversity. His principal work was a scientific survey of his travels and explorations (Vavilov, 1997). He participated in more than 100 plant-collecting expeditions. He obtained plants from Afghanistan, Uzbekistan, the Mediterranean Region, Germany, China, Japan, Korea, United States, Canada, Iran, Cuba, Central and South America, the Caucasus, and Middle Asia. Unfortunately, due to political differences, Vavilov was arrested in 1940. By that time, the collection of cultivated plants in the Soviet Institute of Plant Industry amounted to 200,000 specimens due to his efforts. MID-1900S USDA PLANT EXPLORERS Jack R. Harlan, son of Harry Harlan, devoted his career to studying the evolution of plants, plant exploration, and the interdependence of plants and civilization. His collected work rivaled that of Vavilov, and his insights on centers of crop diversity clarified, corrected, and extended Vavilov’s concepts. Jack Harlan stated, ‘‘For the sake of future generations, we MUST collect and study wild and weedy relatives of our cultivated plants as well as the domesticated races. These sources of germplasm have been dangerously neglected in the past, but the future may not be so tolerant. In the plant breeding programs of tomorrow, we cannot afford to ignore any source of usable genes’’ (Harlan and Harlan, 2005). While the history of recent plant exploration is rightly dominated by a handful of legendary collectors, many others have made important contributions. In 1948, Jack Harlan and Turkish colleague Osman Tosun collected a wheat sample from a field in Turkey that looked terrible. It lodged, had no winterhardiness, and was susceptible to leaf rust. Harlan wrote of the wheat, PI 178383, ‘‘it was a hopelessly useless wheat but was dutifully conserved’’ (Harlan and Harlan, 2005). However, 15 years later, when United States wheat breeders were looking for resistance to a stripe rust outbreak, PI 178383 was found to have resistance to 4 races of stripe rust, 35 races of common bunt, and 10 races of dwarf bunt, as well as tolerance to flag smut and snow mold. Today PI 178383 appears in the pedigree of virtually all of the wheat grown in the Pacific Northwest. Joseph J. Rock, Professor of Chinese and Botany at University of Hawaii from 1907 to 1920, collected economic plants for the USDA from 1920 to 1949. He was sent by the USDA to India to find Taraktogenos kurzii, a tree called the ‘‘kalaw’’ by native tribesmen. A dark-colored oil called ‘‘chaulmoogra’’ reputed to be a cure for leprosy was extracted from the tree. During his expedition, he faced a maneating tiger, a ‘‘rouge’’ elephant, and a typhoon, but he came back with the seed. In the 1930s, Howard Scott Gentry collected and mapped the plants of northwest Mexico. He searched for rubber substitutes in the early 1940s. As a USDA botanist and agricultural explorer from 1950 to 1971, he collected a wide diversity of plant germplasm, including beans and agave from Mexico, cereals and peas from Ethiopia, new crops from South America and the United States, and forage species from the Balkans. Many explorers continued to collect germplasm during the 1930s. Due to world conflicts in the 1940s, the number of plant exploration trips dropped significantly. The 1950s brought a new surge of explorations. Harold Olmo was a Professor of the University of California, Davis. He was hired after prohibition to conduct viticultural research. While conducting research to develop improved grape varieties, he traveled the world to study and collect grape germplasm. He introduced hundreds of grape cultivars from Greece, Portugal, Spain, and the Middle East to the United States. Plants with ornamental value, as well as food and fiber crops were introduced by plant explorers into the United States. For example, in 1957, John Creech introduced Japanese mums for the USDA and Longwood Gardens. 198 HORTSCIENCE VOL. 42(2) APRIL 2007 In 1970, Harold Winters introduced the New Guinea Impatiens into the United States. Dr. Charles M. Rick, Professor in the Department of Vegetable Crops at the University of California, Davis, did significant plant exploration. During a career that spanned more than 60 years, Dr. Rick traveled extensively throughout the Andean region of western South America collecting the wild relatives of the tomato. The intercontinental transfer and exchange of crops between Old and New Worlds has radically changed the nature of global agriculture over many centuries. Through the 19th century and the early decades of the 20th century, exploration was focused on broad goals. Although some explorations were focused on specific crops or crop groups, typically any plant of potential value was a target. Increasingly, explorations focused on specific crops and toward collecting germplasm with specific plant characteristics for use in cultivar improvement and other types of research.
Plant introductions (PIs) into the United
States have come from thousands of sources: immigrants, tourists, government officials. The first official PI activities by the United States government were conducted in the late 1700s and early 1800s when naval officers and diplomats sent germplasm back to the United States. The latter part of the 19th and the first several decades of the 20th century can be described as a ''golden age'' for plant exploration and collecting. During the initial years of this period, agricultural scientists from the United States and elsewhere devoted considerable resources to collecting potential new crops for farmers, as well as superior plants or cultivars of species that farmers were already growing. This paper will discuss the earliest USDA explorers through those of the middle of the 20th century.
MIDDLE TO LATE 1800S: FIRST UNITED STATES PLANT EXPLORERS
The first official federal United States foreign plant exploration was conducted in 1858. On this trip, Robert Fortune collected tea, Camellia sinensis L., from China. This trip occurred before the establishment of the U.S. Department of Agriculture (USDA). Mr. Fortune was sponsored by the U.S. Commissioner of Patents.
Private collections and botanical gardens were interested in exotic plants. Professor Charles S. Sargent was the first Director of the Arnold Arboretum, Jamaica Plain, Mass. In 1892, he inaugurated the Arboretum's interest in eastern Asia by visiting Japan. The Arnold Arboretum subsequently provided the foundation plant material for the introduction of many Japanese plants into American gardens. Ernest H. Wilson, also under the employ of the Arnold Arboretum, undertook four major expeditions to China between 1899 and 1911. He spent most of the second decade of the 20th century exploring for plants in Japan, Korea, and Taiwan.
Although many institutions and individuals were involved in plant collecting during this period, the creation of the USDA Section of Foreign Seed and Plant Introduction (SPI) resulted in the largest single program devoted to plant collection and exploration in the United States (White and Briggs, 1989). James ''Tama Jim'' Wilson, previously the Director of the Iowa Experiment Station, was appointed to be Secretary of the USDA. He appointed David Fairchild, who was at that time 22 years old, to be Chief of the Section and Professor Niles Ebbesen Hansen to be the first official plant explorer for the USDA. Secretary Wilson stated, ''I have twelve thousand men under me, but none who knows how to work like Hansen. There is only one Hansen.'' Indeed Professor Hansen was a very productive collector. Fairchild (1944) states that, ''Almost the moment I was settled in our first location, tons of seeds and plants began to pour in upon me from Russia. Hansen felt that he had been sent out to collect, and he collected everything and collected it in quantity. It was all most embarrassing, as no provision whatever had been made to take care of his shipments.'' Professor Hansen was supported on three USDA plant collecting trips to Russia for 10 months in 1897, to Turkistan for 6 months in 1906, and in Europe and Siberia for 9 months in 1908. On these trips, he collected alfalfa and other forage germplasm, plus apples, pears, crabapples, and small fruits. The plant material that he collected was rich in cold-hardiness and drought tolerance.
EARLY 1900S PLANT EXPLORERS
Mark A. Carleton, a classmate of David Fairchild's, was a USDA wheat breeder who was working to discover rust resistance in cereals. He had been studying Russian, and in the course of his reading he concluded that the Black Lands of Russia could contain resistant strains of wheat. Fairchild asked him to travel to Russia in 1898 to collect wheat, barley, and other cereals. Carleton brought back ''durum wheat,'' a special class of early wheat, which prefers dry conditions and is very productive. The spectacular success of durum wheat focused public attention on the value of PI. It demonstrated that farmers could achieve high profits as a result of the introduction of a new variety. Two years after the introduction of durum wheat, 60,000 bushels were produced; only 5 years later, 20,000,000 bushels were grown (Fairchild, 1944).
Fairchild himself traveled to distant reaches of every continent but Antarctica between 1896 and 1933. He searched for fruits, vegetables, cereals, and other economic plants, including mangos, dates, nectarines, and flowering cherries. During his career he and those who worked under him introduced more than 80,000 accessions into the United States.
Many USDA employees from the Section of Plant Pathology were also involved with The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper therefore must be hereby marked advertisement solely to indicate this fact. 1 Supervisory Horticulturist, Retired. 2 Research Leader/Curator; E-mail khummer@ars-grin.gov PI. Some of these included Walter T. Swingle, Frank N. Meyer, Fredrick Wilson Popenoe, and P. Howard Dorsett. As Harry V. Harlan (Harlan and Harlan, 2005) describes, ''The pioneers of plant industry were as amazing a lot as those who had pushed back our frontiers. They were striking individualists with no two having the same background or any distant resemblances.each brilliant in one way or another-but where in God's world did they find them?'' Walter T. Swingle delivered his first scientific paper at the age of 16. He received a B.S. from Kansas State University at 19 years of age. He joined USDA in 1891 and became an agricultural explorer in 1898. He made many trips to southern Europe, North Africa, the Near East, China, Japan, and the Philippines. He introduced many citrus varieties and unusual species relatives.
Frank N. Meyer traveled over a thousand miles along the routes of Marco Polo across Asia (Cunningham, 1984) ''The desire to explore and to collect is as inherent in us and as much a part of us as toenails and far less transitory than hair. Who has not felt the urge to go places and do things? Any boy old enough to walk spends most of his time exploring, and his mother is in a perpetual state of dismay over the treasure he collects. My urge to explore the world and the desire to collect barleys need no word of explanation. I wanted to go so badly that I would not have blamed those in power if they had harbored a suspicion that my eagerness was as much for personal pleasure as for the good of the cause.'' Harlan stated that ''I never for a moment doubted the material thus obtained would justify the expense of the trip many times over.'' He thought that, ''The whole record of American agriculture was made up of a succession of crops that came from a handful of seed-crops with annual revenues to the farmers valued in the hundreds of millions.'' Harlan was also aware that, ''Unfortunately too few Americans realize what introductions of foreign crops has meant to the American way of life.'' Although we have been discussing American plant explorers, one Russian, Nicolai Ivanovich Vavilov, was extremely instrumental at defining the need for plant exploration as a foundation for agricultural development. Between 1916 and 1940, Vavilov made expeditions to five continents in search of new agricultural plants and confirmation of his theories on plant genetic diversity. His principal work was a scientific survey of his travels and explorations (Vavilov, 1997). He participated in more than 100 plant-collecting expeditions. He obtained plants from Afghanistan, Uzbekistan, the Mediterranean Region, Germany, China, Japan, Korea, United States, Canada, Iran, Cuba, Central and South America, the Caucasus, and Middle Asia. Unfortunately, due to political differences, Vavilov was arrested in 1940. By that time, the collection of cultivated plants in the Soviet Institute of Plant Industry amounted to 200,000 specimens due to his efforts.
MID-1900S USDA PLANT EXPLORERS
Jack R. Harlan, son of Harry Harlan, devoted his career to studying the evolution of plants, plant exploration, and the interdependence of plants and civilization. His collected work rivaled that of Vavilov, and his insights on centers of crop diversity clarified, corrected, and extended Vavilov's concepts. Jack Harlan stated, ''For the sake of future generations, we MUST collect and study wild and weedy relatives of our cultivated plants as well as the domesticated races. These sources of germplasm have been dangerously neglected in the past, but the future may not be so tolerant. In the plant breeding programs of tomorrow, we cannot afford to ignore any source of usable genes'' (Harlan and Harlan, 2005).
While the history of recent plant exploration is rightly dominated by a handful of legendary collectors, many others have made important contributions. In 1948, Jack Harlan and Turkish colleague Osman Tosun collected a wheat sample from a field in Turkey that looked terrible. It lodged, had no winterhardiness, and was susceptible to leaf rust. Harlan wrote of the wheat, PI 178383, ''it was a hopelessly useless wheat but was dutifully conserved'' (Harlan and Harlan, 2005). However, 15 years later, when United States wheat breeders were looking for resistance to a stripe rust outbreak, PI 178383 was found to have resistance to 4 races of stripe rust, 35 races of common bunt, and 10 races of dwarf bunt, as well as tolerance to flag smut and snow mold. Today PI 178383 appears in the pedigree of virtually all of the wheat grown in the Pacific Northwest.
Joseph J. Rock, Professor of Chinese and Botany at University of Hawaii from 1907 to 1920, collected economic plants for the USDA from 1920 to 1949. He was sent by the USDA to India to find Taraktogenos kurzii, a tree called the ''kalaw'' by native tribesmen. A dark-colored oil called ''chaulmoogra'' reputed to be a cure for leprosy was extracted from the tree.
During his expedition, he faced a maneating tiger, a ''rouge'' elephant, and a typhoon, but he came back with the seed.
In the 1930s, Howard Scott Gentry collected and mapped the plants of northwest Mexico. He searched for rubber substitutes in the early 1940s. As a USDA botanist and agricultural explorer from 1950 to 1971, he collected a wide diversity of plant germplasm, including beans and agave from Mexico, cereals and peas from Ethiopia, new crops from South America and the United States, and forage species from the Balkans.
Many explorers continued to collect germplasm during the 1930s. Due to world conflicts in the 1940s, the number of plant exploration trips dropped significantly. The 1950s brought a new surge of explorations.
Harold Olmo was a Professor of the University of California, Davis. He was hired after prohibition to conduct viticultural research. While conducting research to develop improved grape varieties, he traveled the world to study and collect grape germplasm. He introduced hundreds of grape cultivars from Greece, Portugal, Spain, and the Middle East to the United States.
Plants with ornamental value, as well as food and fiber crops were introduced by plant explorers into the United States. For example, in 1957, John Creech introduced Japanese mums for the USDA and Longwood Gardens.
In 1970, Harold Winters introduced the New Guinea Impatiens into the United States.
Dr. Charles M. Rick, Professor in the Department of Vegetable Crops at the University of California, Davis, did significant plant exploration. During a career that spanned more than 60 years, Dr. Rick traveled extensively throughout the Andean region of western South America collecting the wild relatives of the tomato.
The intercontinental transfer and exchange of crops between Old and New Worlds has radically changed the nature of global agriculture over many centuries. Through the 19th century and the early decades of the 20th century, exploration was focused on broad goals. Although some explorations were focused on specific crops or crop groups, typically any plant of potential value was a target. Increasingly, explorations focused on specific crops and toward collecting germplasm with specific plant characteristics for use in cultivar improvement and other types of research.
|
2018-10-09T03:16:49.755Z
|
2007-04-01T00:00:00.000
|
{
"year": 2007,
"sha1": "772a7d91394a70e4bae96d4a85d7df5208ad01ac",
"oa_license": null,
"oa_url": "https://journals.ashs.org/downloadpdf/journals/hortsci/42/2/article-p197.pdf",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "686ea13ed978d31903c1a29fa306fa05a1bb95c9",
"s2fieldsofstudy": [
"Agricultural and Food Sciences",
"History"
],
"extfieldsofstudy": [
"Biology"
]
}
|
244782427
|
pes2o/s2orc
|
v3-fos-license
|
Psychosocial factors associated with malaria care-seeking in rural Ethiopia
Background Ethiopia’s National Malaria Control and Elimination Program aims to diagnose all suspected malaria cases within 24 h of fever onset and provide prompt treatment for confirmed cases. This study explored psychosocial factors associated with no-, delayed- and prompt- care-seeking among female caregivers of children under five years with fever in rural Ethiopia. Methods Household surveys were conducted from 2016–2019 among female caregivers (N = 479) of children under five years old with fever in Oromia; Amhara; Southern Nations, Nationalities, and Peoples Region (SNNPR); and Tigray. Prompt and delayed care-seeking were defined as seeking treatment within ≤ 24 h or > 24 h of symptom onset respectively. Contextual factors explored included sociodemographic factors, household supply of bed nets, exposure to health messages, and household vulnerability (a measure of financial access to food, shelter, schooling, and medical treatment). Ideational factors included psychosocial factors related to care-seeking (knowledge, self-efficacy, response efficacy, attitudes, involvement in decision-making, and household social support). Results The prevalence of fever among children under five years was 18% (ranging from 9% in Tigray to 34% in SNNPR. Overall, 45% of caregivers of children with fever sought care promptly, while 23% delayed care-seeking and 32% sought no care. Prompt care-seeking rates were higher among caregivers with positive attitudes toward prompt care-seeking (48%), involved in decision-making (48%) or perceived equitable gender norms in the community (65%). Caregivers with a high care-seeking ideation had increased odds of prompt care-seeking (aOR: 2.65; 95% CI: 1.74–4.02). Significant contextual factors included residence in the Oromia region (aOR: 2.99; 95% CI:1.40–6.41), caregivers age 35–49 years (aOR: 0.49; 95% CI: 0.26–0.95), residence in vulnerable households (aOR: 2.01; 95% CI: 1.28–3.18). Conclusions Among this rural Ethiopian population, prompt care-seeking was low but positively influenced by both ideational and contextual psychosocial factors occurring at the caregiver level. Multi-sectoral interventions at the individual, community, and health facility levels are needed to improve prompt care-seeking. These include social behavior change interventions to improve ideation, complemented by health facility interventions to ensure provision of high-quality services and structural interventions to increase educational attainment in these rural settings. Supplementary Information The online version contains supplementary material available at 10.1186/s12889-022-13862-x.
with 27 million people living in high-transmission zones (more than one case per 1,000 population) [2] and over 250,000 confirmed cases of malaria in 2019 [1].
Appropriate management of fever cases, insecticidetreated net (ITN) use, and intermittent presumptive treatment of malaria in pregnancy remain key to malaria control. The National Malaria Control and Elimination Program (NMCEP) has made great strides in reducing malaria burden and charting a path to elimination, including the roll-out of rapid diagnostic testing (RDT) and artemisinin-based combination therapy at the community level [2]. Prompt care-seeking, defined as seeking care within the same day or next day of fever onset [3], is a key behavior to ensuring proper case management of suspected and confirmed malaria cases [4].
NMCEP aims to ensure all suspected malaria cases are diagnosed using either RDT or microscopy within 24 h of fever onset and all households living in malaria-endemic areas have the knowledge, attitudes, and practices necessary for adopting appropriate health-seeking behavior for malaria prevention and control [2]. While the 2015 National Malaria Indicator Survey demonstrated that advice or treatment was sought for half of all children with fever, the number of children with fever for whom care was promptly sought is unknown [3]. Considering NMCEP's priorities and the critical role of prompt careseeking in ensuring proper diagnosis and treatment of malaria, relevant stakeholders need to understand the factors influencing care-seeking behaviors for children with fever, particularly at the caregiver level [2].
Of particular interest is the role of caregivers' psychosocial factors, defined as social, cultural, environmental phenomena, and other influences that may affect their behavior [5]. Psychosocial factors play an indubitable role in health and behavior change [6][7][8]. Recent studies in Ethiopia have explored the psychosocial factors associated with prompt care-seeking [9][10][11][12]. A facilitybased study of a predominantly adult patient population in the Amhara region revealed that about half of febrile patient participants sought treatment within 24 h of fever onset, and this was higher among patients who were knowledgeable about malaria prevention and transmission, close to a health center, or had a family size of less than five household members [12]. In Tigray, care-seeking was delayed among low-income patients and those without health insurance [10]. In Dera, northwest Ethiopia, income, community-based health insurance, previous history of malaria infection, decision-making and distance from the facility were determinants of delay in seeking treatment for malaria [11]. In addition, a community based study in Jimma found higher rates of prompt care-seeking among Muslim and uneducated respondents [9].
However, few studies have specifically explored factors influencing prompt care-seeking among caregivers of febrile children under five years of age. This warrants deeper research, as children under five years remain more vulnerable to malaria complications. Community studies demonstrated low levels of caretakers' understanding of malaria in specific localities. Specifically, in the Mandura district of West Ethiopia, a considerable number of caregivers first consulted traditional healers and tried home treatment and thus, sought treatment late [13]. Caregivers' ages, malaria knowledge, attitudes, perceived susceptibility to malaria, and perceived barrier to seek treatment, and if caregivers lived in rural villages, were important factors in seeking health care [13].
This study seeks to explore the spectrum of prompt care-seeking behavior and associated psychosocial factors among female caregivers of children under five years with fever in rural Ethiopia: Oromia; Amhara; Southern Nations, Nationalities, and Peoples Region (SNNPR); and Tigray. The study objective is to identify the extent to which Ethiopian caregivers who seek care promptly, differ from those who delay seeking care, and from those who never seek care at all. Study findings can inform the design of evidence-based community engagement approaches that appropriately target caregivers in these regions and other similar settings.
Several conceptual frameworks have explored psychosocial factors influencing health behavior. As shown in Fig. 1, the ideation framework developed by Kincaid serves as the underpinning for this manuscript [14,15]. The concept of ideation has proven useful in understanding individual-level psychosocial factors affecting malaria-specific outcomes including ITN use [16,17] and appropriate care-seeking [18]. The ideation framework is drawn from various behavioral theories and recognizes that most behavioral decisions are driven by psychosocial factors including cognitive, emotional, and social factors [15]. Cognitive factors include knowledge of disease symptoms; transmission and prevention; and attitudes, values, and attitudes related to proposed actions. Emotional factors include perceived severity and susceptibility to disease, perceived self-efficacy and belief in the efficacy of proposed actions. Social factors include social support, social influence, spousal/partner communication, and personal advocacy.
This study uses the ideation framework to understand the contextual and ideational psychosocial factors associated with care-seeking for malaria among caregivers of febrile children under five years in Ethiopia The study explores a limited number of ideational psychosocial factors related to care-seeking (malaria-related knowledge, self-efficacy, response-efficacy, attitudes, decision-making agency, community norms and social support in the household) against the backdrop of overarching contextual psychosocial factors (demographics, physical environment, and access to resources) seen among caregivers in Ethiopia.
Study background
The data presented in this manuscript are drawn from survey data for the Communication for Health (C4H) project-a five-year (2015-2020) project in Ethiopia funded by the United States Agency for International Development focusing on integrated social and behavior change communication (SBCC) in multiple health areas including malaria [16]. The goal of the C4H project was to improve health practices through standardized SBCC interventions in four major regions of Ethiopia: Amhara, Oromia, Tigray, and SNNPR. The project conducted formative studies, including a baseline survey in 2016 targeting all woredas in the four study regions and a follow-up survey in 2019 assessing the exposure and effectiveness of the interventions among 160 woredas where the interventions had been implemented. The C4H project study population was women of reproductive ages (15-49 years) who were the primary beneficiaries of the project's social and behavior change interventions. The C4H project implemented integrated social and behavior change activities including mass-media and interpersonal communication interventions targeting the general population. Health messages promoting timely care-seeking were included in radio spots, a weekly drama/education radio series; vans mounted with loudspeakers disseminate messages in intervention areas; offline videos focusing on postnatal care, essential newborn care, seeking care for newborn illnesses, immunization, and malaria prevention and treatment; a mobile application that contain key health messages including prevention and care seeking and complementary training of community health workers to conduct effective house to house education and outreach activities such as cue-cards and family health guide.
Sampling strategy
The study data are drawn from two cross sectional studies in 2016 and 2019. In each round, C4H applied probability proportionate to size (PPS) sampling to select project woredas and enumeration areas (EAs). In each EAs, all households were listed to identify eligible households (with women aged 15-49) from which 35 eligible households were randomly selected. The C4H data collectors interviewed all women aged 15-49 face to face from the eligible households using electronic data collection tools. A total of 2,770 women were interviewed at baseline and 1,773 at the follow-up survey, because the intervention was implemented in fewer communities than the potential number determined at baseline. Among all women surveyed, 2453 were caregivers of children under five years out of which 479 caregivers had a child or children with a fever in the two weeks prior to the survey. See the study flow chart in Supplemental Fig. 1 for more details.
Power calculation
Assuming a 5% type I error, the sample of 479 women caregivers of children under five years old who reported fever results in 86% power to estimate the percent of caregivers who sought care promptly with 7% precision (estimate ranging from 44 to 51%).
Ethical considerations
The study received ethical approval from the Ethiopian Public Health Institute (EPHI) ethical review committee, Addis Ababa, Ethiopia, and the institutional review board (IRB) of the Johns Hopkins Bloomberg School of Public Health (JHSPH), Baltimore, Maryland, USA (JHSPH IRB # 00,007,138 and EPHI-IRB-173-2019). All study procedures were performed in accordance with the Declaration of Helsinki. Informed consent was obtained from all study participants. Participants less than 18 years provided informed assent after parental consent was obtained.
Data collection
The study interviewed a total of 4,543 women aged 15-49 (94% response rate). Of these, 2,770 (92% response rate) were interviewed at baseline; because the intervention was implemented in fewer communities than the potential number determined at baseline, 1,773 (97% response rate) were interviewed at follow-up surveys.
The questionnaires aimed to collect information on common themes, including sociodemographic characteristics, malaria related behavior and psychosocial factors, gender equitable norms, and access to health education messages. The study administered questionnaires in the Amharic, Oromia, and Tigrigna languages.
Data analysis
The study measured participants' care-seeking behaviors and psychosocial factors accounting for contextual factors, including demographic characteristics, vulnerability, and gender equitable norms. The survey asked all caregivers, "Has [child's name] been ill with a fever at any time in the last two weeks?" Any care-seeking was defined using the question, "Did you seek advice or treatment for the fever from any source?" This study defined overall care-seeking as "prompt, " "delayed, " or "none, " based on the time elapsed before seeking care.
The key outcome variable was prompt care-seeking for fever; explored by the survey question, "How long after the onset of fever did you seek treatment?" This study defined prompt care-seeking as seeking treatment within 24 h or less and Delayed care-seeking as seeking treatment more than 24 h after the onset of fever.
Psychosocial factors explored in this study included ideational factors relevant to care-seeking as well as overarching contextual factors. Ideational factors include cognitive, social and emotional constructs, which are quantitatively assessed using a battery of Likert scale questions to capture these latent variables [19]. A limited number of ideational psychosocial factors related to care-seeking (knowledge, self-efficacy, response efficacy, attitudes, involvement in decision-making, and social support in the household) were explored as follows: • Knowledge: The study assessed this by exploring participants' awareness of malaria, signs and symptoms, cause, and prevention measures. Specific questions included the following: • "Have you ever heard of an illness called malaria?" • "What signs or symptoms would lead you to think a person has malaria?" • "What do you think is the cause of malaria?" • "How can someone protect themselves against malaria?" • Self-efficacy: This refers to an individual's belief in one's capacity to execute behaviors necessary to produce specific performance attainments [18]. This study assessed participants' self-efficacy for careseeking and malaria prevention using the following statement: "I can take my child to treatment within 24 h of onset fevers, and I am able to have children under five years sleep under an ITN each night. " • Response efficacy: This refers to a person's attitudes as to whether the recommended action step will avoid the threat. The study sought to understand if participants understood the benefits gained if they engaged in prompt care-seeking or malaria prevention behavior. This study proposed statements to participants like, "Having my children sleep under an ITN each night will prevent malaria, " and, "Seeking treatment for my under five children within 24 h of onset fever improves chances of recovery and survival. " • Attitudes towards care-seeking: This study assessed this using the statement, "I should seek treatment for children under five years within 24 h of onset of fever. " Responses for self-efficacy, response efficacy, and attitudinal questions were recorded on a fourpoint Likert scale of "Strongly agree, " "Agree, " "Disagree, " and "Strongly disagree. " • Involvement in decision-making: this study explored this using the question, "Who usually makes deci- This study condensed all ideational psychosocial variables into a composite score, referred to as care-seeking ideation, ranging from 0-12 (Cronbach's alpha of 0.7). Caregivers with a care-seeking ideation score of 9 or more (greater than the median) were categorized as having a high care-seeking ideation, while those with a score of 8 or less were categorized as having low care-seeking ideation. Supplemental Table 1 summarizes the psychosocial variables used in generating the malaria care-seeking ideation.
Access to ITNs within the household was defined as the number of bed nets per household member. Given the assumption that a bed net can be used by up to two people [21], a value of ≥ 0.5 bed nets per household member was defined as adequate ITN supply, while < 0.5 was classified as inadequate. Exposure to health communication messages was assessed by asking whether caregivers had the Family Health Guide (FHG), a national health communication tool given to all families. FHG comprises more than 79 messages with cues to action and illustrations focusing on multiple health areas including malaria.
Additionally, a vulnerability index was assessed in the survey using four states the participant said they experienced in the last 12 months: (1) lacked adequate nutrition, (2) lacked shelter/house to stay in, (3) was unable to afford to send children to school, and (4) lacked sufficient money to buy medicines/medical treatment. Responses included never (coded as 1), rarely (coded as 2), sometimes (coded as 3), and often (coded as 4). We computed a composite score from the sum of the four responses. The vulnerability index was divided into two categories: non-vulnerable (score ≤ 7) and vulnerable (8)(9)(10)(11)(12)(13)(14)(15)(16), with high scores indicating a greater level of poverty. Finally, the study included cross-sectional survey timing (2016 baseline versus 2019 follow-up) as a contextual variable to account for temporal changes that might explain inter relationships between psychosocial factors and care-seeking.
Since we performed analysis on children under five years with a fever in the two weeks preceding the survey, we explored any clustering effect that might occur with multiple children under five years in the same household/ caregiver. No evidence of clustering in the data (intraclass correlation = 0.015) appeared. The study used chisquare, t-tests, Analyses of Variance for bivariable tests of association and exploratory analysis, to compare psychosocial factors across the spectrum of care-seeking. We employed multivariable logistic regressions to explore psychosocial factors associated with both care-seeking and prompt care-seeking. Data management and analysis was performed using Stata software, version 16 (Stata Corporation, College Station, TX, USA) and Excel 2016 (Microsoft Corp, Seattle, WA, USA). We weighted the data using the svyset command in Stata created from the probabilities of woreda, EA and household selection to ensure the data was representative of the study population. Overall, 45% of caregivers of children with fever sought care promptly while 23% delayed care-seeking and 32% sought no care. Care-seeking behavior differed significantly by some contextual psychosocial factors. Prompt care-seekers were more observed in Oromia (37%, 79/214) and SNNPR (35%, 75/214), compared to Amhara (20%) and Tigray (8%) regions. Caregivers' education significantly influenced care-seeking behavior as 41% (89/214 of prompt care-seekers had formal education (only about a quarter of women who delayed or sought no care had a formal education). There were no significant differences in the spectrum of care-seeking by malaria transmission, age, religion, marital status, wealth index, household net supply, vulnerability status, ownership of the health guide or survey timing.
Ideational psychosocial factors related to care-seeking
Ideational psychosocial factors explored included knowledge, self-efficacy, response efficacy, attitudes, decision making involvement and social support in the household. Table 2 shows the spectrum of care-seeking and associated ideational psychosocial factors. While the majority (95%) of caregivers were aware of malaria, only 52% knew malaria symptoms, 60% knew the cause of malaria and 24% knew measures to prevent malaria. Most caregivers had positive attitudes towards prompt care-seeking (90%). In addition, most caregivers had positive perceptions related to malaria including perceived response efficacy for prompt care-seeking (96%) and malaria prevention (88%), perceived self-efficacy to seek care promptly (87%) and prevent malaria (77%). Many (70%) caregivers reported involvement in decision-making, but few noted social support in the household from their partners (44%). Overall, less than half (44%) of caregivers possessed high care-seeking ideation.
Specific ideational psychosocial factors significantly associated with care-seeking included self-efficacy for prompt care-seeking, response efficacy of malaria prevention, positive attitudes toward prompt care-seeking, involvement in decision-making, and perceived equitable gender norms in the community. Specifically, the overwhelming majority of caregivers who sought care promptly had perceived self-efficacy for prompt careseeking (96%, 205/214) compared to their counterparts who delayed (80%,90/112) or did not seek any care (79%, 120/153). Similarly, caregivers who sought care promptly were more likely to have perceived response efficacy of malaria prevention (92%, 196/214) compared to those who did not seek care (82%, 124/153). The majority of caregivers who sought care promptly for their children with fever had positive attitudes towards care-seeking (97%, 208/214)) compared to those delayed (87%) or no (84%) care-seeking. Caregivers who sought care promptly were more likely to be involved in household decisionmaking (74%, 160/214) or perceived equitable gender norms in their community (20%, 44/214) compared to caregivers who delayed care-seeking or sought no care. Overall, the study observed significantly higher rates of overall high care-seeking ideation among caregivers with prompt care-seeking (55%, 117/214) compared to caregivers with delayed or no care-seeking (35% respectively). Table 3 highlights the contextual psychosocial factors associated with high prompt care-seeking ideation. These included malaria transmission zone, age, and marital status. Specifically, while caregivers in SNNPR had reduced odds of a high care-seeking ideation (adjusted odds ratio (aOR): 0.39; 95% confidence interval (CI): 0.19-0.77), caregivers in high transmission zones had increased odds of having high levels of care-seeking ideation (aOR: 1.85; 95% CI: 1.03-3.30). Older caregivers aged 25-34 years (aOR: 2.41; 95% CI: 1.32-4.38) and 35-49 years (aOR: 2.04 (1.05-3.93) old had increased odds of a high care-seeking ideation compared to caregivers aged 15-24 years. In addition, caregivers with a primary education (aOR: 2.24; 95% CI: 1.42-3.55) had increased odds of a high care-seeking ideation. Table 4 details the inter-relationship between prompt care-seeking behavior and contextual psychosocial factors. Caregivers with a high care-seeking ideation had increased odds of prompt care-seeking (aOR: 2.65; 95% CI: 1.74-4.02) compared to their counterparts with low care-seeking ideation. Contextual psychosocial factors associated with prompt care-seeking include residence in Oromia region (aOR: 2.99; 95% CI: 1.40-6.41), being educated (aOR: 1.71; 95% CI: 1.07-2.72), and residence in a vulnerable household (aOR: 2.01; 95% CI: 1.28-3.18). On the other hand, older caregivers aged 35-49 years old had lower odds of prompt care-seeking (aOR: 0.49; 95% CI: 0.26-0.95).
Principal findings
This study is of significance as it is one of few studies exploring factors associated with prompt care-seeking for children under five years with fever in rural Ethiopia. Among the rural Ethiopian population, prompt care-seeking rates were low but positively associated with both ideational and contextual psychosocial factors occurring at the caregiver level. The study found an association between high care-seeking ideation and higher rates of prompt care-seeking. Specific ideational factors of note in this setting included perceived self-efficacy, response efficacy, attitudes toward prompt care-seeking, and involvement in household decision-making. Contextual psychosocial factors associated with influencing prompt care-seeking in rural Ethiopia included region of residence, education, and household vulnerability. This study reported similar rates of care-seeking ideation and behavior across the baseline and follow-up surveys. Of note, a C4H program evaluation report demonstrated significantly higher rates of prompt care-seeking, knowledge, and perceived efficacy among caregivers exposed to the program interventions [22]. The findings are consistent with similar studies in Ethiopia. With regards to ideational factors such as knowledge, perceived self-efficacy and response efficacy, studies in Southwest Ethiopia demonstrated that mothers who have poor knowledge [23], negative attitudes about malaria treatment [24] or misperceptions about the effectiveness of antimalarial drugs [9] were more likely to delay treatment for children under five with fever. Furthermore, in Northwest Ethiopia, decisions to seek health care were found to be taken mostly by fathers, which in turn resulted in delayed care seeking [10]. Study findings on contextual factors such as region, education and household status influencing prompt care seeking in this study were corroborated with studies in West and Northwest Ethiopia affirming that place of residence [13,25], and education [10] were key determinants of health seeking. In other settings across sub-Saharan Africa, women's education [26], household income [27], and reduced perceived efficacy of treatment [27] were significant correlates of care seeking. To the best of our knowledge, this is the first study exploring the interconnections between care-seeking and both ideational psychosocial or contextual factors in Ethiopia, employing a larger perspective and elucidating the complementarity of both psychosocial and structural factors associated with malaria related behavior.
Study implications
Study findings can inform relevant behavior change, structural and research interventions to reverse this trend in rural Ethiopia. This study demonstrated the overall role of psychosocial ideational factors. A key premise behind the ideation model is that people do not ordinarily take action, especially new action, until they have gained sufficient knowledge about it and its consequences, until they have a positive attitude toward it, and until they have talked to others about it and feel right about it [14]. Communication can directly influence such cognitive, emotional, and social factors of ideation separately, as well as jointly as these ideational factors often are interdependent and often occur simultaneously. Considering this, social and behavior change interventions should employ messaging that not only improves communities' knowledge about prompt care-seeking and its benefits but also promotes a positive attitude about prompt careseeking, generates conversation and decision-making regarding prompt care-seeking, and enables communities to feel right about seeking care promptly. Behavior change approaches should aim to promote supportive norms within the community and remove barriers to caregivers' self-efficacy and response efficacy regarding prompt care-seeking and other malaria related behavior. Such behavior change approaches to promote utilization of health services are more impactful when complemented with interventions to ensure high quality service delivery.
The role of contextual and ideational factors suggests interventions should employ a holistic approach to health and quality of life, not just focus on malaria. This is of critical importance, as educational achievement and gender equality norms remain key factors in rural Ethiopia. Holistic approaches should employ a multi-sectoral lens and combine health, education, and gender empowerment interventions to improve the lives of community members. Investing in women, and, more specifically, girl's education, has several positive effects on children, women, and communities at large, combating poverty and fostering economic growth [28,29]. At the heart of achieving gender equality is the education of girls and women and the removal of barriers to education and opportunities for their advancement.
Strengths and limitations
A key strength of the study is the use of a theory-driven, comprehensive exploration of psychosocial factors specific to a high-risk population. We expect that our study findings might be generalizable to similar rural contexts in sub-Saharan Africa. However, some study limitations include its cross-sectional design, which limits the ability to make causal inferences. The use of self-reported data from multiple survey rounds may be limited by recall or social desirability bias. While the study employed the ideation model as part of its theoretical underpinnings, some aspects of this theory such as social influence, spousal/partner communication, and personal advocacy were not explored in this study. In addition, the study did not ask about health provider-or facility-level factors that could influence care-seeking, such as either perceived quality of care or stock out of commodities. Future research should seek to use prospective studies to validate findings with more complete measures of the ideational constructs.
Conclusion
This study identified psychosocial factors among female caregivers of children under five years with fever seeking care in Ethiopia. Structural factors included caregivers' level of education while ideational factors included caregivers' self-efficacy and response efficacy, attitudes related to prompt care-seeking, involvement in decisionmaking, and perceptions of gender equitable norms in the community. Such factors should be accounted for when designing and implementing multi-sectoral interventions at community level.
|
2021-12-02T16:27:26.493Z
|
2021-11-30T00:00:00.000
|
{
"year": 2022,
"sha1": "123bfa2eadcf49da9642d9918343db1b9f8fb947",
"oa_license": "CCBY",
"oa_url": "https://www.researchsquare.com/article/rs-1062042/latest.pdf",
"oa_status": "GREEN",
"pdf_src": "Springer",
"pdf_hash": "00fb4f116a13329d821003f74d82532e2a153183",
"s2fieldsofstudy": [
"Medicine",
"Psychology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
260531916
|
pes2o/s2orc
|
v3-fos-license
|
Global, regional, and national estimates of the population at increased risk of severe COVID-19 due to underlying health conditions in 2020: a modelling study
Summary Background The risk of severe COVID-19 if an individual becomes infected is known to be higher in older individuals and those with underlying health conditions. Understanding the number of individuals at increased risk of severe COVID-19 and how this varies between countries should inform the design of possible strategies to shield or vaccinate those at highest risk. Methods We estimated the number of individuals at increased risk of severe disease (defined as those with at least one condition listed as “at increased risk of severe COVID-19” in current guidelines) by age (5-year age groups), sex, and country for 188 countries using prevalence data from the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2017 and UN population estimates for 2020. The list of underlying conditions relevant to COVID-19 was determined by mapping the conditions listed in GBD 2017 to those listed in guidelines published by WHO and public health agencies in the UK and the USA. We analysed data from two large multimorbidity studies to determine appropriate adjustment factors for clustering and multimorbidity. To help interpretation of the degree of risk among those at increased risk, we also estimated the number of individuals at high risk (defined as those that would require hospital admission if infected) using age-specific infection–hospitalisation ratios for COVID-19 estimated for mainland China and making adjustments to reflect country-specific differences in the prevalence of underlying conditions and frailty. We assumed males were twice at likely as females to be at high risk. We also calculated the number of individuals without an underlying condition that could be considered at increased risk because of their age, using minimum ages from 50 to 70 years. We generated uncertainty intervals (UIs) for our estimates by running low and high scenarios using the lower and upper 95% confidence limits for country population size, disease prevalences, multimorbidity fractions, and infection–hospitalisation ratios, and plausible low and high estimates for the degree of clustering, informed by multimorbidity studies. Findings We estimated that 1·7 billion (UI 1·0–2·4) people, comprising 22% (UI 15–28) of the global population, have at least one underlying condition that puts them at increased risk of severe COVID-19 if infected (ranging from <5% of those younger than 20 years to >66% of those aged 70 years or older). We estimated that 349 million (186–787) people (4% [3–9] of the global population) are at high risk of severe COVID-19 and would require hospital admission if infected (ranging from <1% of those younger than 20 years to approximately 20% of those aged 70 years or older). We estimated 6% (3–12) of males to be at high risk compared with 3% (2–7) of females. The share of the population at increased risk was highest in countries with older populations, African countries with high HIV/AIDS prevalence, and small island nations with high diabetes prevalence. Estimates of the number of individuals at increased risk were most sensitive to the prevalence of chronic kidney disease, diabetes, cardiovascular disease, and chronic respiratory disease. Interpretation About one in five individuals worldwide could be at increased risk of severe COVID-19, should they become infected, due to underlying health conditions, but this risk varies considerably by age. Our estimates are uncertain, and focus on underlying conditions rather than other risk factors such as ethnicity, socioeconomic deprivation, and obesity, but provide a starting point for considering the number of individuals that might need to be shielded or vaccinated as the global pandemic unfolds. Funding UK Department for International Development, Wellcome Trust, Health Data Research UK, Medical Research Council, and National Institute for Health Research.
Methods used to calculate the proportion of individuals with at least one underlying condition
Note: Some of the text from the main paper is repeated here for convenience.
Proportion with at least one underlying condition relevant to severe COVID-19 disease
The GBD study provides prevalence estimates for each disease category separately, but not what we needed, which was the prevalence of people in at least 1 of these categories. Diseases may cluster, for example if they are causally related. To deal with this, we first calculated , which is the expected proportion of individuals with at least one condition assuming no clustering and that the various prevalences are independent (e.g. the fact that someone has diabetes does not affect their risk of getting cancer) as 1 minus the probability of not having any of the conditions c1, c2, c3….i.e. 1 -(1 -p_c1) x (1 -p_c2) x (1 -p_c3)….
We then estimated the proportion who have at least one underlying condition as = × , where is the ratio between the observed and expected percentage of individuals with at least one condition. We based on evidence from large cross-sectional multimorbidity studies in Scotland 1 and Southern China. 2 The ratio was broadly consistent by age, sex and study (see Figure 1 overleaf). For the analysis of both males and females combined, the mean of all age-specific values of r was 0.92 (range 0.86 to 0.99) in Scotland and 0.92 (range (0.75 -1.15) in China. When extrapolating this value to other countries, we used a ratio of 0.9 for all age groups varied this between 0.7 and 1.0 for transparency. The resulting national estimates of were constrained to be no less than each country's single most prevalent condition. We conducted sensitivity analyses to explore the impact on results of using the observed age-specific values of rather than the same value for all ages, but this had a very small impact on the share of the population estimated to be at increased risk i.e. the share of the population at increased risk changed from 22.5% to 22.8%.
Adjustment for multimorbidity
In addition to providing estimates for , the studies in Scotland and Southern China were also used to calculate the multimorbidity fraction i.e. the proportion of individuals with multiple (two or more) underlying conditions among those with at least one, by age group and sex. All analyses were done using disease categories that matched as closely as possible to the COVID-19-relevant categories defined in our analysis. In both studies this included: CVD (defined as the presence of one or more of coronary heart disease, hypertension, cerebrovascular disease, peripheral arterial disease, heart failure, or atrial fibrillation); chronic neurological disease (defined as one or more of dementia, multiple sclerosis and Parkinson's disease); and CRD (defined as one or both of chronic obstructive pulmonary disease and bronchiectasis). Other COVID-related conditions listed in the main analysis were counted separately. The GBD provide separate estimates for hypertensive heart disease and CKD due to hypertension, but it was not possible to make this distinction in the multimorbidity datasets, so all hypertension was included in the CVD category.
We calculated pooled estimates of the multimorbidity fraction by age and sex and extrapolated these pooled estimates to all countries included in the analysis (see Figure 1 overleaf).
Figure 1. Empirical estimates of the ratio (left panel) and the multimorbidity fraction among those with at least one underlying health condition relevant to COVID-19 (right panel) from cross-sectional studies in Scotland and Southern China
The top row shows results for females and males combined. The middle row shows results for females only and the bottom row shows results for males only.
The left panel/column shows the ratio between the observed and expected % of individuals with at least one condition by age. Expected estimates were calculated by assuming the prevalences of COVID-19 underling conditions are independent (e.g. the fact that someone has diabetes does not affect their risk of getting cancer) as 1 minus the probability of not having any of the conditions c1, c2, c3….i.e. 1 -(1 -p_c1) x (1 -p_c2) x (1 -p_c3)…. This was then compared to the observed value of the % of individuals with at least one condition based on the same dataset (either Scotland or Southern China). The ratios between expected and observed are shown on the left panel/column below. Both studies indicate that the expected value based on the assumption of independence would provide reasonable estimates of the observed value. In our main analysis we assumed the ratio was 0.9 but varied this between 0.7 and 1.0 in uncertainty analysis.
The right panel/column shows the proportion of those with at least one underlying condition relevant to COVID-19 with multimorbidity (two or more conditions). As expected, this percentage increases with age in both studies. The grey lines represent pooled estimates and 95% confidence intervals based on a 2 nd order polynomial model fitted to all data points. Pooled estimates were extrapolated to all countries included in the analysis by age and sex. The lower and upper CI values were used in our low and high estimates in the main paper.
Infection hospitalisation ratios
To estimate the number of individuals at high risk (those that would require hospital admission if infected) we applied country-level UN estimates of the number of individuals alive in each 5-year age group 3 to age-specific infection hospitalisation ratios (IHRs) recently estimated for mainland China by Verity et al. 4 IHRs represent the proportion of people who are infected that would have symptoms severe enough to require hospital admission. The term 'require hospital admission' is consistent with the WHO definition for severe cases. 5 Whether or not these individuals actually receive hospital care will depend on the health system in the country concerned but is beyond the scope of this analysis.
Adjustments to IHRs
We made two adjustments to account for differences between IHRs in China and other countries. The first was designed to capture the effect on IHRs of national variations in prevalence mix. The second was to adjust for infections in given age groups being more severe in higher mortality settings.
1. Adjusting for underlying conditions. For each 5-year age group and sex, the prevalence rates for each underlying condition were multiplied by their respective relative risks (RRs) for hospitalisation. RRs of 3.0 were assumed for CKD, diabetes and CVD and of 1.5 for the eight other conditions. The RRs we used were informed by a rapid review of what is currently known about the strength of association between different variables and COVID-19 hospital admission (see table 2 below). The totals were then summed across all 11 conditions and added to the proportion of individuals without underlying conditions, to create a risk score for each 5-year age group. IHRs were then adjusted to account for the ratio of the risk score for the country of interest and China. For example, for males aged 55-59 years the risk score was 2.63 in Afghanistan and 1.95 in China, so the IHR for this age group was multiplied by a ratio of 1.35 (2.63/1.95). This adjustment assumes that for each underlying condition, the RR of hospital admission is the same for every country.
2.
Adjusting for age-based frailty. Direct interpretation of the first adjustment relies on constant RRs of admission for each condition across countries. To address the likelihood of more severe infections at given ages in higher mortality settings we went further. For each 5-year age group and sex, we divided UN estimates of age-specific life expectancy 3 for China by the equivalent estimate for the country of interest. This ratio (proxy for difference in age-based frailty) was then applied to the IHR for same age group. For example, the life expectancy for males at age 55 years is 22.8 and 19.1 years in China and Afghanistan, respectively, so the IHR for this age group was multiplied by a ratio of 1.19 (22.8/19.1) to generate the estimate for Afghanistan. Country-specific differences in life expectancy may be more extreme than country-specific differences in infection severity because life expectancy depend not only general health status, but also on access to effective health care. Also, the extent to which infections will be more severe in higher mortality settings is still very uncertain. We therefore show results with and without this adjustment.
Thus, for males aged 55-59 years in Afghanistan, separate adjustments for underlying conditions and age-based frailty had the effect of increasing the IHR by 1.61 (IHR x 1.35 x 1.19), from 9% to 15%.
Several studies are emerging on the risk of mortality among those already hospitalised 6,7 but we restricted our analysis of RRs to studies that allowed comparison of hospitalised and non-hospitalised COVID-19 cases. This was because our analysis focused on the risk of severe disease (requiring hospital admission) rather than the risk of death. Three studies from the USA met these criteria. The first contained descriptive data on 6,637 COVID-19 cases reported to the CDC as of 28 th March, 2020. For this study, we derived crude univariable RRs for each condition from the reported case counts. 8 The second was a multivariable analysis of 4,103 COVID-19 cases in New York City, between 1 st March 2020 and 2 nd April 2020. 9 The third was a multivariable study from Northern California with 1,052 confirmed cases of COVID-19 captured between 1 st January and 8 th April, 2020. 10 For conditions where evidence was weak or missing, we included studies that did not meet our initial inclusion criteria, either because they were not yet published (i.e. multivariable analyses in Italy 11 and the UK 12 ) or because many of the COVID-19 cases in the less severe group were hospitalised i.e. one published metaanalysis of 4 studies from China. 13 Using the evidence from these studies, we summarised the strength of the association with hospital admission (low, moderate, high) and graded our confidence in the strength of that association (low, moderate high).
This rapid review aimed to summarise what is currently known about the strength of association between different variables and COVID-19 hospital admission, but this is unlikely to be exhaustive. We therefore chose to use a very simple stratification of risk for the 11 conditions (either RR= 3 or RR = 1.5) based on the range of RRs reported for these conditions. For three conditions (CKD, diabetes and CVD) we had moderate confidence that the strength of the association would be high based on univariable and multivariable analysis, so assumed a RR of 3.0. For the eight other conditions we assumed a RR of 1.5 because there was insufficient evidence of a significant independent association with hospital admission from multivariable analyses.
RR values for each condition were varied one at a time (assuming low and high values of 1 and 10 respectively) to assess the impact of these changes on the total population at high risk (p 11). Gender is not included in current guidelines. In one multivariable analysis in New York City (n=4103) male gender was a significant independent predictor of hospitalisation (OR = 2.80, 95% CI 2.38 -3.30, p <0.001). 9 Males were twice as likely to be admitted to hospital as females (OR = 1.9, p =0.001) based on multivariable analysis from Northern California (n=1,052). 10 In a large study of hospitalised patients in the UK, 60% of hospitalised patients were male and this effect was seen across all ages (n=16,749). 6 Males had a RR of 1.64 (95% CI 1.25 -2.14, p <0.001) in a multivariable study that compared a cohort that did and did not have a test-positive hospital admission in the UK (n=428,225). 12 Data from an unpublished multivariable analysis in Italy (n=1866) found an association between male gender and hospital admission (HR = 1.4 [95% CI 1.2-1.6]). 11 High (for simplicity we assumed males were twice as likely as females to be at high risk i.e. to require hospital admission if infected).
Pregnancy
The prevalence of pregnancy was the same among COVID-19 hospital admissions (6%) and the wider community in a large study in the UK (n=16749), 6 but this is likely to remain in current guidelines until more evidence emerges e.g. on the risks at different stages of pregnancy. Table 3 (overleaf) summarises the share of the population estimated to be at high risk (those that would require hospital admission if infected) for the base case scenario and for a range of alternative scenarios.
Sensitivity analysis for estimates of the number of individuals at high risk
The following assumptions were varied: 1. Low and high credible intervals of IHRsscenarios based on the low and high credible interval values of the IHRs reported in Verity et al, 16 were influential. The global population at high risk (4.5%) decreased to 2.7% with low IHRs and increased to 9.1% with high IHRs; 2. Adjustments for underlying conditionsremoving this adjustment reduced the global population at high risk from 4.5% to 4.0% (and from 3.1% to 2.7% in Africa). In some countries removing this adjustment had a more substantial impact, due to very high prevalence of specific conditions relative to the same conditions in China e.g. diabetes in Fiji, HIV/AIDs in Swaziland; 3. Adjustments for age-based frailty -as expected, removing the age-based frailty adjustment resulted in a lower population at high risk in Africa (from 3.1% to 2.6%) and a higher population at high risk in Europe and other high-income settings; 4. Altering the maximum proportion of the population infectedour central estimates of the numbers at high risk of severe COVID-19 disease assume there is no theoretical maximum to the proportion of the population that could ever be infected. Thus, the total number of individuals at high risk is calculated by multiplying the total population in each age group by the IHR for the same age group. However, as our understanding of COVID-19 transmission dynamics improves, empirical data may begin to emerge on the scale and duration of immunity acquired from natural infections, and on the proportion of the population that could ever be infected given widespread community transmission. If this is lower than our current assumption of 100%, then fewer individuals would be at high risk using our method. When we varied this value, the global population at high risk decreased from 4.5% assuming all individuals could ever be infected to 4.0%, 3.6%, 3.1% and 2.7% when assuming this value could not exceed 90%, 80%, 70% and 60%, respectively; and,
5.
Altering the RRs for specific conditionschanging the RR values for each condition one at a time (assuming low and high values of 1 and 10 respectively) had a limited impact on the total population at high risk (<5% increase/decrease). Results were most sensitive to CVD, CKD, diabetes and liver disease (due to its higher prevalence in China relative to many other settings). Increasing the RR for HIV/AIDS from 1.5 to 10.0 was influential in Africa and increased the share of the population at high risk from 3.1% (42 million) to 3.7% (49 million).
Table 4. Number of individuals in millions at increased risk of severe COVID-19 illness by age, number of conditions, region, and age threshold: low scenario estimates
For numbers at increased risk, the low estimates were based on a scenario assuming the lower 95% CI values for the age-sex-specific population estimates, disease prevalence rates, and multimorbidity fraction, and assume an r ratio of 0.7. Figure 3 in the main paper shows the share of the population at risk in different countries based on real-world differences in population structure and disease prevalence. This information is important when calculating the numbers that might need to be shielded or vaccinated but does not allow direct comparison of risks at equivalent ages in different countries. In this alternative version (see below), circles have been added to show the age-standardised share of the population at high risk (black circles) and increased risk (open circles). These assume each country has the same WHO standard reference population. 17 A low age-adjusted population at risk in countries with older populations (eg, Japan, Europe and Puerto Rico) helps to confirm that older age is the main reason why these countries have a high unadjusted population at risk. Similarly, a high age-adjusted population at risk in African countries with high HIV prevalence (eg, eSwatini, Lesotho) and small island nations with high diabetes prevalence (eg. Fiji, Mauritius) explains why these countries have a high unadjusted population at risk, despite having younger populations. Differences in demography can mask important differences in age-specific risks that may be revealed by age-standardisation. For example, in eSwatini and New Zealand the population at high risk is 5% in both countries, but when risks are compared for equivalent age groups (within the spreadsheet tool) the age-specific risks in eSwatini are more than double those in New Zealand (consistent with eSwatini having a higher age-adjusted population at high risk ie, 8% vs 3%). Thus, although younger populations will tend to have a lower share of the population at risk than older populations, their risk at equivalent ages could still be higher.
|
2020-06-16T13:05:52.115Z
|
2020-06-15T00:00:00.000
|
{
"year": 2020,
"sha1": "7977d01d35246dbdf2f2af56bf94106d8cc09177",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1016/s2214-109x(20)30264-3",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "55fb3fad86f3d9f0c39cb794a55aa342b3951f83",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
108692037
|
pes2o/s2orc
|
v3-fos-license
|
The finite element models of thin-walled branched structures in heat transfer problems
At present engineers are often faced with the task of designing thin-walled branched structures working under conditions of intensive heat interchange with the environment. In the junction zone of the plates of such structures, great temperature oscillations take place, which has an influence on the analysis of the stress–strain state of the structure [1]. In solving this problem, first of all we have to evaluate accurately the values of the temperature in the plate junction zone. Transient heat conduction in an anisotropic material is described by the differential equation
Introduction
At present engineers are often faced with the task of designing thin-walled branched structures working under conditions of intensive heat interchange with the environment.In the junction zone of the plates of such structures, great temperature oscillations take place, which has an influence on the analysis of the stress-strain state of the structure [1].In solving this problem, first of all we have to evaluate accurately the values of the temperature in the plate junction zone.
Transient heat conduction in an anisotropic material is described by the differential equation here ,, x y z k k k are anisotropic thermal conductivity coeffi- cients; is the mass density of the material; c is the spe- cific heat capacity; ,, Q x y z is the heat generated within the body; ,, T x y z is the temperature; t is the time.To solve the Eq. ( 1), initial and boundary conditions are introduced where ,, x y z n n n are the direction cosines, ,, q x y z is the heat flux, α is the convection coefficient over the surface, T ∞ is the fluid temperature.
Because the analysed structure is of a complex geometrical shape, it is convenient to solve the problem of the temperature field by the finite element method.All mathematical dependences of the method are easily formalised and generated by computer technologies.
An Eq. ( 1) with initial conditions (2) and with boundary conditions (3) in accordance with [2] is written as the system of differential equations where K , C and F are respectively thermal conduc- tivity matrix, heat capacity matrix and vector of thermal load.
The finite element model of the examined structure is composed of finite elements of several types: plane (triangular or quadrangular) and connective quadrangular prismatic finite elements.That is why the problem of modelling plate junction zones with finite elements appears.It is analysed in research works [3][4][5].Plane finite elements are joined with triangular or quadrangular prismatic connective finite elements.
Practical calculations have shown that in those cases when connective finite elements join other finite elements in the nodes that are in the middle-surface of the plates, it is necessary to use the middle nodes of the prism base instead of corner nodes of the prism.To calculate temperature values in such nodes special transformation matrices are necessary.
In this work, transformation formulas of coordinates that translate the middle side nodes of the bases of a quadrangular prism to the corner nodes of the prism, and formulas that translate the temperature values of the corner nodes into the temperature values of the middle side nodes of the bases of the finite element are found.The problem is solved for the connective triangular prismatic finite element [3].
The aim of the present article is to assess the transient temperature field of the plate junction zone of a thinwalled branched structure of an anisotropic material, when the boundary conditions on the lateral surfaces of the plate can be expressed in three different forms: convection, heat source, and heat flow.The junction zone of the plates is going to be modelled by a connective quadrangular prismatic finite element, the main mathematical expressions of which for the isotropic material are described in [4].
Matrices of the connective quadrangular prismatic finite element
To make a discrete model of the junction zone of the plates, a connective quadrangular finite element of an equal cross-section is going to be used; with a quadrangle as the base (Fig. 1).The element has 8 degrees of freedom.where [N] is the matrix of the shape functions of the element, {T e } is temperature values in the nodes of the element.
The shape functions, satisfying properties in the nodes of the element , , 0, when 1, 2,3, ,8 where S is the cross-sectional area of the connective finite element, h is the height of the element.The values of a and b are calculated by using the formulae to calculate the distance between two points.The contribution of the connective finite element to the matrices K , C and vector F of the Eq. ( 4) is expressed by the following formulae here V is the volume of the finite element, S is the lateral area of the element, [B] is a matrix of the derivatives of the shape functions of the element in the directions x, y and z, [D] is a matrix of the thermal conductivities in the directions x, y and z.
The matrix [D] in the discussed case is entered as follows 00 00 00 By differentiating matrix [N] (7) with respect to x, y and z, we form matrix [B], the inverse matrix of which is entered as follows While calculating matrices of the finite element, let us assume that The finite element thermal conductivity matrix [K] (8) has two parts: − the first integral describes the thermal conductivity of the finite element, determined by the thermal conductivity coefficients of the material; − the second integral describes the thermal conductivity of the finite element, determined by the convection heat exchange with environment via the lateral surfaces and the ends of the finite element.
The thermal conductivity matrix of the element always has the first component, while the second has to be evaluated only when the lateral surfaces and the ends of the finite element are open, i.e. when they have a contact with the environment.
While calculating the first integral of the thermal conductivity matrix of the element (8), the values of matrices [B], [B] T and [D] are inserted in its expression.It can be seen that post-integral expressions are rather complex.That is why it is difficult to calculate them manually; besides, it is easy to make mistakes.That is why mathematical transformations are made by using computer algebra systems MAPLE and MATHEMATICA.
Having integrated the first integral of the thermal conductivity matrix of the connective finite element, we get The value of the second component of the thermal conductivity matrix of the connective finite element (8) depends on the surface of the element (lateral or end) where the heat transfer with convection takes place.The value of this integral does not depend on the thermal conductivity coefficients, that is why by employing the outcomes of [4], we can calculate the value of integral (15) the surface, e.g.
1 : , , The values of integral (15) are written accordingly for surfaces S 2 , S 3 , S 4 , S 5 , S 6 , S 7 , S 8 , where the values of the elements of matrix [ ' i S ] depend on the numbers of the nodes belonging to lateral or end surfaces.
The matrix of the thermal capacity of the connective finite element (9) according to [3] has the following form Because the integrals of the thermal load vector (10) do not depend on the thermal conductivity coefficients, on the basis of [4], the value of the first integral is
The values of the other two integrals of the thermal load vector (10) depend on the numbers of the nodes of the lateral or end surfaces of the finite element.For instance, for surface
The transformation formulae of the coordinate nodes of the connective finite element
We shall now analyse the base (quadrangle) of the connective finite element (Fig. 2) with nodal points at middle side points, numbered counter-clockwise from the first freely chosen node.We shall transfer the middle side nodes of the base 2 , , , x y z problem, the equations of lines going via given points, known from analytical geometry, are going to be used.Because the branched structure of a complex shape is analysed, the connective finite element can be in various positions in its finite elements scheme.That is why it is necessary to analyse a few cases of the positions of the nodes of the connective finite elements in space., , , , The rank of matrix
The transformation formulae of the temperature values of the connective finite element
In the case when the scheme of the finite elements of the structure is composed only from connective quadrangular finite elements, it is not necessary to translate the temperature values of the corner nodes to the middle side nodes of bases.Otherwise, when connective finite elements in the structure join plane finite elements, it is necessary to translate the temperature values of the corner nodes to the middle side nodes of bases.Having evaluated previously formulated conditions (7), we get the following transformation matrix of temperature values
Numerical results
We shall analyse the temperature field in a structure of an anisotropic material in the case of a transient heat transfer process.The algorithm was written by FORTRAN.
A cooled branched structure is given [4].The air moving across the upper surface has a temperature of 20ºC, convection coefficient is 8.1 W/(m 2 °C).The lower surface is cooled by a liquid of the temperature of -196ºC, convection coefficient is 3529 W/(m 2 °C).The solution schema of the finite elements of the structure with a quadrangular connective finite element is shown in Fig. 3. Fig. 3 The grid of the finite elements of the structure To analyse the temperature convergence, two finite element schemes were adapted: with and without a connective quadrangular finite element.The convergence of the results is illustrated by observing the change of temperature of a certain point of the calculated scheme in time, e.g. 31 (Fig. 3).
The view of the temperature convergence of node 31 of the two calculated schemes is presented in Fig. 4. 2. In cases of the first and second sampling of the junction zone, the curves of the temperature values of node 31 differ.According to the calculated scheme with a connective finite element, the calculated temperature values (at node 31) are closer to the control values than the values obtained according to the scheme without a connective finite element.
3. The obtained transformation formulae of the node coordinates and temperature values allow making a finite element scheme of the analysed structure both with corner and middle side nodes of a base of a connective finite element.This allows a more efficient modelling of joining a few plates with different normals to a horizontal surface.
S. Turskienė THE FINITE ELEMENT MODELS OF THIN-WALLED BRANCHED STRUCTURES IN HEAT TRANSFER PROBLEMS S u m m a r y
The paper deals with the problem of analysis of the transient temperature field in the plate junction zone of a thin-walled branched structure of an anisotropic material.In the junction zone of the plates of such structures, great temperature oscillations take place, which have an influence on the analysis of the stress-strain state of the structure.The junction zone of the plates was modelled by a connective finite element with a quadrangular base.The thermal conductivity matrix of the connective finite element was obtained in the case of an anisotropic material.The coordinate transformation and temperature value transformation formulae were derived, which allowed composing a grid of the finite elements of the examined object both with the corner and middle side nodes of the connective finite element.This allows a more efficient modelling of joining a few plates with different normals to a horizontal surface.The derived formulae were checked by the solution of a problem of the temperature field of a cooled plate.
Fig. 1
Fig.1The geometry of the connective finite element . we shall express the coordinates of nodes A, B, C, D by the coordinates of nodal points 1, 2, 3, 4. To solve this
Fig. 2
Fig. 2 Numbering of the nodes of the base of the element Thus the thermal conductivity matrix and thermal capacity matrix of the connective finite element to calculate temperature values in the middle side nodes of the base of the finite element are found as follows
Fig. 4
Fig.4 The temperature in time of node 31 of the two calculated schemesThe correctness of the formulae of the transfor- mation of node coordinates and temperatures values of the connective finite element is established by calculating temperature values in the connective nodes 11, 16, 21 and 26 (Fig.3).The results of the calculation are presented in the table below.
|
2019-04-12T13:55:49.820Z
|
2012-04-20T00:00:00.000
|
{
"year": 2012,
"sha1": "9224e86ddbbef56423aa0b7da9852b616cb83841",
"oa_license": "CCBY",
"oa_url": "https://mechanika.ktu.lt/index.php/Mech/article/download/1562/1400",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "9224e86ddbbef56423aa0b7da9852b616cb83841",
"s2fieldsofstudy": [
"Materials Science"
],
"extfieldsofstudy": [
"Materials Science"
]
}
|
226374177
|
pes2o/s2orc
|
v3-fos-license
|
The antiproliferative potential of isolated emodin and aloe-emodin from Rheum ribes on different cancer cell lines
Herbal compounds are important resources that used in the treatment of many diseases, and are inspiring in drug design and development. Rheum ribes, a perennial native plant, has potentially diverse medicinal effects, including antioxidant, antibacterial and anti-tumor activity. The key molecules associated with anticancer activity of R. ribes in tumor formation still remains unclear. The present study was aimed to elucidate the antiproliferative effects of two purified compounds, emodin and aloe-emodin, from R. ribes. In our study, chromatographic separation was used to purification of compounds. Experimental 1D and 2D 1H and 13C-NMR spectra helped us to confirm the chemical structures of emodin and aloe-emodin. The viability of human glioblastoma cell line (U373), human breast carcinoma cells (MCF-7), and human colorectal cancer cell line (HT-29) were detected by WST-1 proliferation assay as dose- and time-dependent. The isolated emodin and aloe-emodin from R. ribes, reduced viability of human glioblastoma, breast and colorectal cancer cell lines in dose- and time-dependent manner significantly. These antiproliferative effects may contribute as a possible strategy to the development of new therapeutic approaches in the treatment of cancer.
Introduction
Plants have been used for many centuries as nutrient, dietary supplementation, food additives, flavoring, and main source of drugs used in treatment of human sickness. In particular, the therapeutic properties of plants are currently being studied intensively for the development and discovery of newer therapeutics. Increasing incidence and high mortality rates have make highlighted the studies on the treatment of cancer in recent years. In cancer studies, phyto-compounds, which are compounds derived from plants, have an important position, because of their low or no cytotoxicity and clinical efficacy. Plant-derived compounds such as irinotecan, camptothecin, paclitaxel, vinblastine, combrestatin and etoposide, are used in clinical as chemotherapy drugs [1], and natural plant polyphenols, curcumin, quercetin, resveratrol, luteoline, gallic acid, hesperidine and genistein, etc., are also considered as promising adjuvants for cancer prevention and treatment [2].
Rheum species are known to have medical importance in the treatment of many diseases and disorders. Rheum ribes L., a member of the Polygonaceae family, is an important source in folk medicine. The medically important Rheum species are rich in anthracene-derived compounds [3]. R. ribes are used to treat laxative and joint disorders in Iran [4,5]. The Polygonaceae family also includes eight genera and seventy species in the Turkey. R. ribes is the only species in the family represented by the Turkey [6]. This plant grows naturally Turkey, Lebanon and Iran and it famous with names "Işgın, Kurdish Banana and Uçgun".
R. ribes' young twigs and stems consumed as the salad, fresh vegetable and appetizing by the local people especially in Turkey's eastern and south-eastern Anatolia region. Besides, roots, leaves, flowers and stems are also used by natives in this area to treat the bile ducts, stomach discomforts, diarrhea, hemorrhoids, measles and smallpox as well its use as an appetite enhancer [7,8]. Further, ethnobotanists in Jordan reported that the root of R. ribes was used by the public in diabetes, hypertension, obesity, kidney stone and sand [9]. Water decoction from the R. ribes roots was shown to reduce blood sugar in mice when given to mice diabetic with alloxan, but not hypoglycemia when administered to healthy mice forming the control group [10]. It was shown that isolates of the roots, leaves and stems of R. ribes had anti-viral and anti-microbial effect activity [11][12][13][14].
The objective of this study was to purify the bioactive compounds, emodin (E) and aloe-emodin (AE), from R. ribes, and to determine the anti-proliferative activities of these purified compounds first time. The anti-proliferative activities of E and AE was observed in the cancer cell models of human glioblastoma, breast and colorectal cancer, U373, MCF-7 and HT-29 cell lines, respectively. This study elucidates the anticancer potential of R. ribes E and AE phytochemicals as potential anti-tumor agents.
Plant material
R. ribes species, which were grown at 2000 m and higher in the natural habitats of Yelesen village of Bingöl, Turkey, were collected from the field following the flowering period during April-May months. The collection of the species was devoted to collecting without destroying nature. The species identification of the plant material was made by Alpaslan Koçak and the herbarium specimen was stored with KOÇAK 4003 registration number. R. ribes, collected in a suitable form from the field, was spread on unprinted paper and dried in a damp and shaded environment. The dried material was milled in the laboratory type mill so as not to exceed 3 cm.
Extraction, fractination and isolation process
Dried and ground plant material (5 kg) was taken in a glass flask and 10.0 L of methanol: chloroform (1:1; v/v ratio) was added and left to extraction for 2 days at the laboratory temperature. At the end of the second day, the extract was filtered with the help of the Whatman no 4 filter paper and the remaining pulp was allowed to be extracted again with the same amount and ratio of solvent system. This procedure was repeated three more times. The extracts obtained by the extraction process were combined and the solvents they contained were removed under reduced pressure at a temperature of 35°C by means of the rotary evaporator. Removal of the solvent system from the extracts yielded 523.8 g of dry extract with a yield of 10.476%. The extraction yield was calculated by the following formula; % Extract yield = [Dry extract amount (g)/dry plant quantity (g)]x100. The dry extract was stored at +4°C by taking an amber flask for anticancer activity and isolation studies. For isolation, 50 g of methanol: chloroform dry extract was dissolved in 1L of distilled water. The aqueous extract was fractionated with chloroform (8x0.6 L). At the end of the fractionation, 15.63 g of dry chloroform extract was obtained as a result of the removal of chloroforms by rotary evaporator. Isolation process was continued with this part. For the flash chromatography system, the pharmaceutical glass column (5x60cm) containing sephadex LH-20 column filler. The methanol: chloroform (35:65; v: v) solvent system was used as mobile phase and operating conditions for elution collection were adjusted to 2 mL per minute. A total of 126 sub-fractions were eluted in each elution volume of 10 mL. As a result of the chromatographic separation, the emodine (68 mg) and aloe-emodin (56 mg) were obtained in the sub-fraction 37 and 75, respectively (
Cell viability
The dose-and time-dependent antiproliferative effects of E and AE from R. ribes, was measured using the WST-1 cell proliferation assay kit according to manufacturer`s protocol (Intron Biotechnology, Gyeonggi-do, Korea), in vitro. Cells (5 × 10 3 cells/well) were seeded into 96-well culture plates and incubated at 37°C and 5% CO2 overnight. The plates were treated with various concentrations (1, 10, 25, 50, and 100 µg/mL) of E and AE for 24 and 48 hours. The control groups were exposed to 0.05% DMSO containing media. Culture medium was replaced by fresh medium and 10 μl of WST-1 reagent were pipetted into wells. After 3 h incubation period, the amount of soluble formazan dye produced by metabolically active cells was measured with an ELISA microplate reader (SpectraMax 384 Plus, Molecular Devices, USA) at 450 nm as absorbance (optical density (OD). The mixture of WST-1 reagent and culture medium without cells was evaluated as a background control. To determine percentage of cell survival that formula was used: % cell survival = ODtest/ODcontrol × 100%. The half maximal inhibitory concentration (IC50) value for each test group was calculated according to the findings.
Statistical analysis
Each assay was performed in three experiments and mean± standard deviation (SD) values were calculated. Statistical analysis of the findings were evaluated by t-test and one-way ANOVA test using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA). P values <0.05 was assumed to be statistically significant.
Chemical characterization
The 1 H and 13 C NMR spectra of E and AE compounds were analyzed using 600 MHz NMR spectrometer (Bruker AVANCE). 7 mg of each compound was taken to NMR tube and dissolved in 450 µL of DMSO-d6. The chemical shifts of compounds that consistent with the literature was documented in the table 1 [19,20]. Also chemical structures of emodin and aloe-emodin confirmed by experimental 1D and 2D 1 H and 13 C-NMR spectra.
Antiproliferative effects of emodin and aloe-emodin from R. ribes
The dose-and time-dependent antiproliferative activities of E and AE were assesed in U373, MCF-7 and HT-29 cells. The increasing concentrations of E and AE were treated cell lines. After 24 and 48 h treatment, the viability of the cells were observed with WST-1 cell proliferation assay. The GraphPad Prism 8 software was used to estimate the halfmaximal inhibitory concentrations (IC50). E and AE exhibited dose-and time-dependent effects on viability of all cell lines. As shown in Fig 2, the decreases in percentage of viability were statistically significant in all treatment groups compared to control (U373 *p<0.001; MCF-7 **p<0.001; and HT-29 ***p<0.001). Treatment with E at 1 µg/mL significantly reduced viability in all three cell lines in 24 h treatments, however, the decrease in viability of U373 and MCF-7 cells was milder at 48 hours treatment of the same concentration. The IC50 values of E, at 24 h treatment, were 29.75, 92.33 and 27.64 µg/mL for U373, MCF-7 and HT-29 cell lines, respectively (Fig 2A). The required concentrations of E to inhibite the growth of 50% of cells (IC50) for U373, MCF-7 and HT-29 were 13.94, 35.45 and 9.51 µg/mL at 48 hours (Fig 2B). These results indicate that, 48 h treatment with E reduces IC50 values by more than half, compared to 24 h treatment. On the other hand, the viability of the cells decreased significantly in treated with AE at 1 µg/mL, which was the lowest concentration, but at the highest concentration, 100 µg/mL, the viability of all cells was the lowest at 48 hours of treatment. AE treatment at 100 µg/mL in 48 h, reduced the viability of U373 human glioblastoma cells, MCF-7 breast cancer cell line and HT-29 colorectal adenocarcinoma cells to 37.1, 41.2 and 20.7%, respectively (Fig 2D). The IC50 values of AE were 65.06, 51.53, 13.23 µg/mL and 18.59, 16.56, 5.38 µg/mL in U373, MCF-7 and HT-29 cell lines, at 24 and 48 h treatment, respectively (Fig 2C, Fig 2D). IC50 values in 48 h AE treatment groups, also decreased approximately three-fold compared to 24 h treatments.
Conclusions and discussion
Cancer causes huge losses as a major health problem all over the world. Many new therapeutics and clinical treatments have been proposed to combat this disease. In this context, herbal-based treatment approaches stand out in the fight against cancer with their low toxic effects and efficacy. The therapeutic potential of medicinal plants has recently aroused a great interest in the treatment and prevention of many diseases, such as cancer [21]. The anticarcinogenic effect of Rheum genus have been shown in several studies. It was determined that R. officinale water extract was significantly growth inhibitory on A549 and MCF-7 cell lines and IC50 values were 620 and 515 μg/mL [22]. R. emodi water and methanol extracts have been reported to show anticancer activity against two human carcinomas (MDA-MB-435S breast cancer and Hep3B liver carcinoma cell lines) [23]. R. ribes ethanol, water and methanol extracts significantly reduce the viability of HCT-116, A2780, MCF-7 and PC-3 cells, especially at concentrations of 25, 50 and 100 µg/mL, and in the same cell lines the IC50 values were determined between 14.2-42.2 µg/mL [24]. In another study, ethanolic extract of R. ribes was showed significantly cytotoxic activity against SW742 and MCF-7 cells, and this extract inhibited the half of the cells at concentrations of 46.4 and 51.3 µg/mL, respectively [25]. R.ribes butanol fraction exhibited antiproliferative activity on MCF-7 human breast cancer cells, and IC50 value was determined as 36 µg/mL [26]. R.ribes is particularly rich in anthraquinone compounds, and emodin and aloeemodin are among the main components [15][16][17]. In this study, the anti-proliferative effects of two anthraquinone derivatives, emodin (1,3,8 trihydroxy-6-methylanthraquinone) and aloe-emodin (1,8-dihydroxy-3-hydroxymethylanthraquinone), which isolated from R.ribes, were investigated on different cancer cell lines.
Glioblastoma is a malignant brain tumor that affects stem/progenitor cells in the brain and it has the rapidly spread potential in adults [27]. Ismail et al. demonstrated that aloe-emodin inhibits proliferation of U87 malignant glioma cells in a dose and time dependent manner and has an IC50 value of 58.6 μg/mL in 24 h treatment and 25.0 μg/mL in 48 h treatment [28]. In accordance with this study, we found that aloe-emodin, that isolated from R. ribes, was decreased the viability of U373 glioblastoma cells in a dose and time dependent manner, and the IC50 value was 65.1 μg/mL and 18.6 μg/mL at 24 and 48 h treatment, respectively. We have also demonstrated that isolated emodine significantly inhibits the viability of U373 glioblastoma cells depending on increasing time and concentration.
Breast cancer is the most common cancer among women and accounts for 9.6% of cancer-related deaths worldwide [29]. Kang et al. determined the IC50 value of the emodin and aloe-emodin isolated from R. palmatum for 48 h in MCF-7 breast cancer cells as 16.4 and 12.6 µg/mL, respectively [30]. Similarly, we found the IC50 value as 16.6 µg/mL for R. ribes' aloe-emodin in MCF-7 cells at 48 h treatment. However, isolated emodin from R. ribes was had a milder antiproliferative effect than aloe-emodin. It has been reported that, emodin and aloe-emodin significantly inhibited the proliferation of MCF-7 cells in a dose and time dependent manner and show anticancer activity [31]. We can also say that, emodin and aloe-emodin show anticancer activity depending on dose and concentration, and these two compounds make important contributions to R.ribes anticancer potential.
Colorectal cancer is the second cause of cancer-related deaths in both men and women [32]. Suboj et al. treated WiDr, RKO, SW 480 and SW 620 colon adenocarcinoma cells for 48 h with commercially AE, and showed that AE has a dose-dependent antiproliferative effect and it was arrested the cell cycle in the G2/M phase [33]. Similarly, in our study, AE exhibited dose-dependent inhibition in the proliferation of HT-29 colorectal cancer cells. In addition, both emodin and aloe-emodin, that isolated from R.ribes, significantly inhibited the proliferation of HT-29 colorectal adenocarcinoma cells, even at very low concentrations.
Overall, these results indicate that, isolated emodin and aloe-emodin from R. ribes have a significant antiproliferative effect in human breast, colon and glioma cancer cell lines, and therewithal suggesting that these isolated compounds from R. ribes have potential therapeutic efficacy in the prevention and treatment of cancer. However, further investigations are needed to clarify the molecular mechanisms of these isolated compounds.
|
2020-11-14T01:07:11.052Z
|
2020-08-15T00:00:00.000
|
{
"year": 2020,
"sha1": "27aab700d137209ed469e7b484cc1735d642a8b3",
"oa_license": null,
"oa_url": "https://dergipark.org.tr/en/download/article-file/1241057",
"oa_status": "GOLD",
"pdf_src": "Adhoc",
"pdf_hash": "22cc6073a2c3305291fd0933f514e5b0f1bbcede",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Chemistry"
]
}
|
265905591
|
pes2o/s2orc
|
v3-fos-license
|
Excited-state intramolecular proton transfer to carbon atoms: nonadiabatic surface-hopping dynamics simulations
Excited-state intramolecular proton transfer (ESIPT) between two highly electronegative atoms, for example, oxygen and nitrogen, has been intensely studied experimentally and computationally, whereas there has been much less theoretical work on ESIPT to other atoms such as carbon. We have employed CASSCF, MS-CASPT2, RI-ADC(2), OM2/MRCI, DFT, and TDDFT methods to study the mechanistic photochemistry of 2-phenylphenol, for which such an ESIPT has been observed experimentally. According to static electronic structure calculations, irradiation of 2-phenylphenol populates the bright S 1 state, which has a rather flat potential in the Franck–Condon region (with a shallow enol minimum at the CASSCF level) and may undergo an essentially barrierless ESIPT to the more stable S 1 keto species. There are two S 1 /S 0 conical intersections that mediate relaxation to the ground state, one in the enol region and one in the keto region, with the latter one substantially lower in energy. After S 1 - S 0 internal conversion, the transient keto species can return back to the S 0 enol structure via reverse ground-state hydrogen transfer in a facile tautomerization. This mechanistic scenario is verified by OM2/MRCI-based fewest-switches surface-hopping simulations that provide detailed dynamic information. In these trajectories, ESIPT is complete within 118 fs; the corresponding S 1 excited-state lifetime is computed to be 373 fs in vacuum. Most of the trajectories decay to the ground state via the S 1 /S 0 conical intersection in the keto region (67%), and the remaining ones via the enol region (33%). The combination of static electronic structure computations and nonadiabatic dynamics simulations is expected to be generally useful for understanding the mechanistic photophysics and photochemistry of molecules with intramolecular hydrogen bonds.
Introduction
Excited-state intramolecular or intermolecular proton transfers are elementary processes occurring in many molecular and biochemical systems [1][2][3][4][5] and electronic devices, [6][7][8][9] for example, in natural and artificial photosynthesis, 10,11 water-splitting photocatalysis, 12 green fluorescent proteins, 13,14 and photoswitches. 15Understanding these excited-state proton transfer processes is important both from fundamental and technological points of view.7][18][19][20][21][22][23][24][25][26][27][28][29][30] Most of these previous studies focused on excited-state proton transfer processes between two highly electronegative atoms, e.g.27,[31][32][33][34][35][36][37] What about excited-state intramolecular proton transfer (ESIPT) in molecules without strong hydrogen bonds, for example, in alcohols or phenols?Such ESIPT processes were first investigated in the 1980s, 38,39 with proton transfer to aromatic carbon atoms being first addressed at the beginning of this century. 402][43][44][45][46][47][48] They first studied photochemical deuterium incorporation at the ortho and para positions of 2-phenylphenol in various solvent mixtures 49 and found that the predominant exchange at the ortho position is independent of water and methanol contents, implying an intramolecular process.They also investigated the photochemistry of o-hydroxybiaryls, which features not only an efficient excited-state proton transfer to the ortho carbon atom of the naphthyl ring, but also a novel ring-closing reaction. 50Flegel et al. 51 studied the photoaddition of water and alcohols to the 9-and 10-positions of the anthracene moiety of 9-(2-hydroxyphenyl)anthracene in acetonitrile and methanol mixtures and proposed a mechanism involving water-mediated excited-state proton transfer from the phenolic OH group to the anthracene fragment.Basaric ánd Wan 52 investigated the potential excited-state proton transfer in four derivatives of 9-(2-hydroxyphenyl)anthracene.Nayak and Wan 53 explored photochemical deuterium incorporation in extended orthosubstituted biaryl systems and reported the longest solvent-assisted proton-relay chain.They proposed direct and water-assisted proton transfer mechanisms to explain photohydration at the ortho and distal positions, respectively.In these experimental studies, it was generally believed that excited-state intramolecular proton transfer to ortho positions is efficient in phenols.
The underlying photophysical and photochemical mechanisms in these systems have not yet been elucidated in detail, e.g., with regard to the relevant structures, proton transfer paths, excited-state potential energy surfaces, lifetimes, and decay channels.We are aware of only one recent theoretical study in this context, 54 which employed the single-reference second-order coupled cluster (RI-CC2) method to explore direct and water-assisted excited-state proton transfer in 2-phenyl-1-naphthol.Given this situation, we decided to perform high-level multi-reference electronic structure computations and trajectory-based surface-hopping dynamics simulations to study the mechanistic photochemistry of the prototypical 2-phenylphenol molecule, with emphasis on the ESIPT process to the ortho carbon atom and the deactivation channels leading back to the ground state.
OM2/MRCI method
[79][80] During geometry optimizations, all required energies, gradients, and nonadiabatic coupling elements were computed analytically.Conical intersections were optimized using the Lagrange-Newton approach. 81,82n OM2/MRCI calculations, the restricted open-shell HF formalism was applied in the self-consistent field (SCF) treatment (i.e., the orbitals were optimized for the leading configuration of the S 1 state with two singly occupied orbitals).The active space in the MRCI calculations included 12 electrons in 10 orbitals (see ESI †).In terms of the SCF configuration, it comprised the five highest doubly occupied orbitals, two singly occupied orbitals, and the three lowest unoccupied orbitals.For the MRCI treatment, three configuration state functions were chosen as references, namely the SCF configuration and the two closed-shell configurations derived therefrom (i.e., all singlet configurations that can be generated from the HOMO and the LUMO of the closed-shell ground state).The MRCI wavefunction was built by allowing all single and double excitations from these three references.
The nonadiabatic dynamics was studied by performing 1 ps OM2/MRCI trajectory surface-hopping simulations.The initial atomic coordinates and velocities were randomly selected from 5 ps trajectories of ground-state molecular dynamics.The number of excited-state dynamics runs was then chosen according to the computed S 0 -S 1 transition probability.A total of 193 surface-hopping trajectories were run, with all relevant energies, gradients, and nonadiabatic coupling vectors being computed on-the-fly as needed.For points with an S 1 -S 0 energy gap of less than 10 kcal mol À1 , the fewest-switches criterion was applied to decide whether to hop.The time step was chosen to be 0.1 fs for nuclear motion and 0.0005 fs for electronic propagation.The unitary propagator evaluated at a mid-point was used to propagate the electronic motion.The translational and rotational motions were removed in each step.The empirical decoherence correction (0.1 a.u.) proposed by Granucci et al. was employed. 83The final evaluations were done for the 148 trajectories that finished successfully and satisfied our energy continuity criterion (no changes greater than 30 kcal mol À1 between any two consecutive MD steps).9][84][85][86]
Ground-state properties and vertical excitation energies
Apart from the most stable ground-state structure of 2-phenylphenol (S0-ENOL), there is also a minimum for the keto tautomer (S0-KETO), see Fig. 1 and Table 1.For each of the two minima, OM2/MRCI and CASSCF yield similar geometries.S0-ENOL is more stable than S0-KETO by 34.7 (33.7) kcal mol À1 at the OM2/MRCI (MS-CASPT2) level.
The computed vertical excitation energies for the first excited singlet state (S 1 ) are collected in Table 2 obtained from laser flash photolysis of 2-phenylphenol in solutions. 49A slightly lower experimental value of 4.28 eV has been reported for 2-phenyl-1-naphthol featuring more extensive conjugation. 54The S 1 state at the Franck-Condon point is spectroscopically bright; its oscillator strength is computed to be 0.155 at the TD-CAM-B3LYP level.Molecular orbital analysis shows that the S 0 -S 1 electronic transition mainly originates from the HOMO-LUMO single excitation (Fig. 2).The HOMO is mainly localized in the phenolic part, whereas the LUMO is localized in the phenyl group.Hence, the S 1 state is of charge-transfer character, which sets the stage for the subsequent excited-state proton transfer.In fact, this kind of electronic structure change has been found in many similar intramolecularly hydrogen-bonded systems. 18,20,22,26,27,30cited-state minima At the CASSCF level, there is a shallow S 1 minimum in the Franck-Condon region of 2-phenylphenol (S1-ENOL), which is computed to lie 103.0 kcal mol À1 above the S0-ENOL minimum in single-point MS-CASPT2 calculations.At the OM2/MRCI level, no such S 1 minimum could be located since all minimizations starting from the S0-ENOL equilibrium geometry led directly to the S 1 keto species (S1-KETO), see the left panel of Fig. 3. Likewise, the minimum-energy path for proton transfer computed at the RI-ADC(2)/def2-SVP level indicates an essentially barrierless excited-state enol-keto tautomerization, see the right panel of Fig. 3.This is also verified by OM2/MRCI nonadiabatic dynamics simulations (vide infra).CASSCF optimization yields another S 1 minimum in the keto region (S1-KETO, see Fig. 1).At this geometry, there is no significant charge transfer in the S 1 state; the S 0 -S 1 transition involves mostly the central C3QC5 double-bond region, causing an elongation of this bond from 1.396 to 1.480 Å (S 0 versus S 1 keto minimum, CASSCF values).Other geometric parameters change only slightly (Fig. 1).According to single-point MS-CASPT2 calculations, S1-KETO lies 74.4 kcal mol À1 above S0-ENOL and 28.6 kcal mol À1 below S1-ENOL (Table 1).Thus, the excited-state proton transfer that yields the keto species is highly exothermic; in other words, 2-phenylphenol is a strong photoacid in S 1 .As already mentioned, this proton transfer is computed to be essentially barrierless and is thus expected to be ultrafast.In terms of excited-state topology, our present results are consistent with recent RI-CC2 computations on a similar system, 2-phenyl-1naphthol. 54
Conical intersections
At the OM2/MRCI level, we were able to locate two S 1 /S 0 minimum-energy conical intersections (S1S0-ENOL and S1S0-KETO).Selected bond lengths and dihedral angles are given in Fig. 1 and in Table 1, respectively.In S1S0-ENOL, the H10 atom is still attached to the O9 atom (phenol species) but the OH group is extruded out of the ring plane, with a C4C3C2O9 dihedral angle of À831 (OM2/MRCI).This strong out-of-plane deformation significantly increases the S 0 energy, thus closing the S 0 -S 1 energy gap and reaching an S 1 /S 0 conical intersection.In S1S0-KETO, the H10 atom is already bonded to the C7 atom (keto species); the two rings are not coplanar with a C2C3C5C7 dihedral angle of 581.Energetically, S1S0-ENOL [S1S0-KETO] is computed to lie 92.1 [73.7] kcal mol À1 above the S0-ENOL ground state, and 21.4 kcal mol À1 [39.8 kcal mol À1 ] below the S 1 energy at the Franck-Condon point (113.5 kcal mol À1 ); hence, these two conical intersections are energetically accessible.Taking into consideration that S1S0-KETO is more stable than S1S0-ENOL by 18.4 kcal mol À1 , the former is expected to play a more vital role in excited-state deactivation.
It is worth stressing that OM2/MRCI and the ab initio methods give similar structures and energies for the two S 1 /S 0 conical intersections (Table 1).Taking S1S0-ENOL as an example, the dihedral angles C2C3C5C7, C4C3C2O9, and C3C1C2O9 are computed to be À281, À831, and 1411 at the OM2/MRCI level, compared with À351, À971, and 1551 at the CASSCF level, respectively, (see Fig. 1).The relative energies from OM2/MRCI and single-point MS-CASPT2 calculations are also reasonably close to each other: the values of S1S0-ENOL [S1S0-KETO] are 92.1 [73.7] kcal mol À1 for OM2/MRCI, and 98.8/100.7 [69.8/73.9]kcal mol À1 for MS-CASPT2.In the latter case, the quoted S 0 and S 1 state energies differ slightly because they come from single-point MS-CASPT2 calculations at CASSCF-optimized geometries.
Excited-state decay paths
The preceding static electronic structure computations suggest the following scenario for the photoinduced processes in 2-phenylphenol.Upon irradiation, the spectroscopically bright S 1 state is populated in the Franck-Condon region, from which the S 1 /S 0 conical intersection with an intact phenol moiety is energetically accessible (with relaxation to the ground state via S1S0-ENOL).A competitive process involves an essentially barrierless excited-state proton transfer yielding an S 1 keto minimum, which can decay to the ground state via the S 1 /S 0 conical intersection in the keto region (S1S0-KETO); back in the S 0 state, the keto species S0-KETO can return to the more stable tautomer S0-ENOL via reverse ground-state hydrogen transfer.
To verify this mechanism and to explore the timescales of the underlying photophysical and photochemical events, we have performed trajectory-based fewest-switches surface-hopping dynamics simulations starting in the S 1 state of 2-phenylphenol.
Hopping-point distribution
The S 1 -S 0 hopping-point distribution extracted from all surfacehopping trajectories reflects the topology of the conical intersection seam. 84The two types of S 1 /S 0 conical intersections in 2-phenylphenol, S1S0-ENOL and S1S0-KETO, clearly govern our nonadiabatic dynamics simulations.Fig. 4 depicts the distributions of the C7H10 distance and the C4C3C2O9 dihedral angle at all S 1 -S 0 hopping points.Obviously, there are two main hopping regions, which cluster around two minimum-energy S 1 /S 0 conical intersections S1S0-ENOL and S1S0-KETO.A closer examination of the C7H10 distance distribution at all hopping points in Fig. 4 shows that most of the trajectories (67%) hop to the S 0 state via the keto conical intersection seam.This preference arises from two factors: first, the S 1 proton transfer is essentially barrierless so that the S 1 keto species is generated easily, and second, S1S0-KETO is thermodynamically favored over S1S0-ENOL because its potential energy is lower by 18.4 kcal mol À1 (OM2/MRCI).Hence, it is not surprising that most trajectories decay to the S 0 state via S1S0-KETO in our dynamics simulations.
S 1 lifetime
In our simulations, 118 of 148 (80%) trajectories have reached the S 0 state at the end of the 1 ps nonadiabatic dynamics runs.As shown in the left panel of Fig. 5, most of the S 1 -S 0 hops happen between 100 and 400 fs (only 4 hops after 400 fs).Again, this ultrafast decay is consistent with the excited-state topological features, i.e. an almost barrierless proton transfer and two efficient deactivation channels (vide supra).
The S 1 excited-state deactivation can be viewed as a firstorder elementary reaction.The S 1 state population is thus ruled by the following rate equation: where k is the corresponding rate constant; p 0 is the S 1 population at the end of the run (0.2 in this work); and t 0 is the initial delay time (57 fs).After fitting the time-dependent state population in Fig. 4 to eqn (1), we obtain an S 1 excitedstate lifetime of 373 fs.One should note that the S 1 excited-state lifetime may be expected to increase in the condensed phase, in particular in a rigid environment.
Product distribution
Fig. 6 shows the product distribution at the end of the 1 ps nonadiabatic simulations.Overall, there are four kinds of products, namely S 1 enol (13%) and keto (5%), and S 0 enol (47%) and keto (35%); the enol : keto ratio is estimated to be 3 : 2. The top panel illustrates the distribution of the resulting phenol conformers.Most of the trajectories ending up in the phenol region have the H10 atom bonded to the O9 atom.However, there are also some S 0 phenol products that have the H8 atom bonded to the O9 atom, not the H10 atom (see H8O9 distribution).In these trajectories, the excited-state proton transfer and the reverse ground-state hydrogen transfer involve two different hydrogen atoms (H10 and H8, respectively).The bottom panel depicts the distribution of the keto conformers after the 1 ps simulations.There are ca.90 trajectories with the H8 atom bonded to the C7 atom, and ca.40 trajectories with the H10 atom bonded to the C7 atom.
Typical trajectories
In our nonadiabatic dynamics simulations, we see three different photocycles that start from S1-ENOL and end up at S0-ENOL: (I) the S 1 state decays directly to the ground state, without excitedstate intramolecular proton transfer (ESIPT); (II) the S 1 state first evolves towards the S 1 keto species via an ultrafast barrierless ESIPT and then decays to the ground state in the keto region followed by a reverse ground-state hydrogen transfer (GSHT) involving the same migrating hydrogen atom; (III) the photocycle is the same as in case (II) except that different hydrogen atoms are involved in ESIPT and GSHT.In the following, we present for each photocycle pattern a representative trajectory to illustrate the main photophysical and photochemical events.Fig. 7 shows a typical trajectory for case (I) with direct decay via the S1S0-ENOL conical intersection.Within the first 400 fs, the system starts to rotate around its central C3C5 bond (strong changes in the C2C3C5C6 and C2C3C5C7 dihedral angles; only small fluctuations in the C4C3C2O9 and C3C1C2O9 dihedral angles).During this process, the nonadiabatic coupling remains small and the S 1 -S 0 energy gap remains large, so there is no nonadiabatic transition.After about 400 fs, the C4C3C2O9 dihedral angle starts to decrease from 1801 to 401 at ca. 600 fs.The S 1 and S 0 states now become energetically close to each other (within 4 kcal mol À1 ) and there is a large nonadiabatic coupling; thus, a nonadiabatic S 1 -S 0 hop takes place, with relaxation of S 1 to the S 0 state.Thereafter, the C4C3C2O9 and C3C1C2O9 dihedral angles move back towards their original values (from twisted to a more planar arrangement).There is no ESIPT in this trajectory.We emphasize that this photocycle pattern occurs only rarely in our trajectories.
Fig. 8 depicts a typical trajectory for case (II) with deactivation to the S 0 state via the S1S0-KETO conical intersection.In the initial stage of this trajectory, the O9H10 and C7H10 distances quickly increase and decrease, respectively.At ca.50 fs, the ESIPT is complete and the S 1 keto species S1-KETO is formed, which remains in the S 1 state for another 150 fs (while retaining a rather short O9H10 distance indicative of excited-state hydrogen bonding interactions).Thereafter, it decays to the S 0 state at a point where the S 1 -S 0 nonadiabatic coupling becomes very large (panel c) and the S 1 -S 0 gap is very small (panel d).Interestingly, the generated S 0 keto species does not return back to the enol region immediately; instead, it roams the keto region for additional 500 fs.Then, a reverse groundstate hydrogen transfer takes place, regenerating the S 0 enol conformer and completing the photocycle.The rotation around the C3-C5 bond starts after ca.700 fs (see the C2C3C5C6 and C2C3C5C7 dihedral angles in panel b) while the C4C3C2O9 and C3C1C2O9 dihedral angles do not vary much.
Fig. 9 presents a typical trajectory for case (III).Here, the O9H10 and C7H10 distances fluctuate around their equilibrium positions in the first 100 fs; then, they start to increase and decrease quickly.At about 110 fs, the S 1 keto species S1-KETO is formed, which stays in the S 1 state for ca.50 fs and then decays to the S 0 state at 165 fs, when the keto S 1 /S 0 conical intersection is encountered.The generated keto species roams the keto region in the S 0 state for a longer time (640 fs).After ca.800 fs, the most stable S 0 phenol conformer is regenerated via a reverse ground-state hydrogen transfer.The rotation around the C3-C5 bond starts after ca.900 fs (see the C2C3C5C6 and C2C3C5C7 dihedral angles in panel b).Interestingly, the H10 atom bonded to the O9 atom is transferred to the C7 atom in the ESIPT process, while the H8 atom originally bonded to the C7 atom is transferred to the O9 atom in the final GSHT step.
Discussion
Our results are consistent with the experiments available for 2-phenylphenol.Lukeman and Wan 49 argued that singlet reactivity is major for 2-phenylphenol, which is consistent with our computations.The S 1 excited-state proton transfer is nearly barrierless and ultrafast, so it is impossible for the system to efficiently populate triplet states in the Franck-Condon region.In addition, the S 1 -T 1 intersystem crossing in the keto region is not expected to be competitive with the efficient internal conversion from the S 1 keto species to the S 0 state.However, this intersystem crossing could become more probable in a rigid environment because the internal conversion involves a large conformational change that could be impeded by steric interactions with the environment.Furthermore, there is experimental evidence that 2-phenylphenol is a strong photoacid in the S 1 state.This point is supported by the MS-CASPT2 results (see Fig. 10), which confirm that the S 1 excited-state proton transfer is highly exothermic -S1-KETO lies 28.6 kcal mol À1 below S1-ENOL and 40.6 kcal mol À1 below the initially populated S 1 Franck-Condon point.
We emphasize in this context that our present computations are carried out in vacuum and thus only consider the intrinsic photochemistry of 2-phenylphenol, for example, direct ESIPT processes to the ortho-position, without accounting for solventassisted intermolecular proton transfer to remote sites such as para-positions.Previous electronic structure computations on a similar system 54 showed that there exists an efficient S 1 /S 0 conical intersection near the keto region, but without optimizing its structure.In this work we precisely located this kind of minimum-energy conical intersection in 2-phenylphenol, both at the OM2/MRCI and CASSCF levels (S1S0-KETO), and we explored its dynamical role in the S 1 photodynamics of 2-phenylphenol using full-dimensional surface-hopping dynamics simulations.We find that 67% trajectories decay to the S 0 state via this conical intersection in the keto region.In addition, we optimized the S 1 /S 0 conical intersection in the Franck-Condon region (S1S0-ENOL), which also plays an important role in the S 1 deactivation (33%).Thus, both conical intersections need to be considered in order to correctly understand the mechanistic photochemistry of 2-phenylphenol and its variants.
How is the S 0 isomer S0-ENOL-1 (see the bottom of Fig. 10) generated in the photodynamics of 2-phenylphenol?Experimentally, Lukeman and Wan 49 assumed that this species comes from S0-KETO via a concerted reverse hydrogen transfer and 1,5hydrogen shift.Our present dynamics simulations do not support this scenario -we do not see any 1,5-hydrogen shift in any of the trajectories.Instead, S0-ENOL-1 is generated by a simple singlebond rotation, after the hydrogen atom originally bonded to the phenyl ring has been transferred to the oxygen atom (Fig. 10).
The S 0 keto species has not yet been detected spectroscopically when using nanosecond laser flash photolysis. 49,54B3LYP calculations indicate that the tautomerization of S0-KETO to the most stable phenol conformer S0-ENOL has to overcome a small barrier of 4.0 kcal mol À1 .It might thus be possible to observe this predicted transient species using ultrafast timeresolved transient spectroscopy.
Summary
With the use of electronic structure computations and trajectory-based surface-hopping dynamics simulations, we have for the first time explored the mechanistic photochemistry of 2-phenylphenol.We have simulated the S 1 excited-state proton transfer and deactivation as well as the reverse hydrogen transfer in the S 0 state.Mechanistically, some trajectories directly evolve from the Franck-Condon region toward an enol-type S 1 /S 0 conical intersection, followed by an S 1 -S 0 internal conversion to the ground-state minimum.Most of the trajectories proceed from the Franck-Condon region to the S 1 keto species via an essentially barrierless ESIPT; the transient S 1 keto species is then de-excited to the ground state via a second S 1 /S 0 conical intersection in the keto region, followed by a quick relaxation back to the most stable phenol minimum via a reverse GSHT process (barrier of ca. 4 kcal mol À1 at the B3LYP level).The nonadiabatic dynamics simulations predict an average lifetime of 118 fs for the ESIPT process. 40,49In these simulations, 67% of the trajectories decay to the S 0 state via the keto S 1 /S 0 conical intersection, and 33% decay via the S 1 /S 0 conical intersection in the Franck-Condon region.According to the computed time-dependent state populations, the S 1 excitedstate lifetime is estimated to be 373 fs in vacuum.We hope that these computational results and mechanistic insights will stimulate further experimental work on 2-phenylphenol, especially by ultrafast time-resolved transient spectroscopy.
Fig. 3 (
Fig. 3 (left) OM2/MRCI optimization path starting from the enol minimum, which leads directly to a keto species after about 50 steps; (right) RI-ADC(2)/def2-SVP computed minimum-energy reaction path with respect to the C7-H10 proton transfer reaction coordinate.See text for discussion.
Fig. 4
Fig. 4 Distribution of the C7H10 distance and the C4C3C2C9 dihedral angle at all S 1 -S 0 hopping points.See text for detailed discussion.
Fig. 5
Fig. 5 Distribution of the S 1 -S 0 hopping times (left) and time-dependent S 1 and S 0 state populations (right).See text for detailed discussion.
Fig. 6
Fig. 6 Distribution of the C7H10, C7H8, H8O9, and O9H10 bond lengths at the end of 1 ps simulations.See text for detailed discussion.
Table 2
49mputed vertical excitation energies to the first excited singlet state of 2-phenylphenol and the experimental band maximum from laser flash photolysis in solution49 This journal is © the Owner Societies 2015 Phys.Chem.Chem.Phys., 2015, 17, 9687--9697 | 9695
|
2017-10-23T12:42:19.585Z
|
2015-04-01T00:00:00.000
|
{
"year": 2015,
"sha1": "8c26953a9c79ebcac97019f1474e960c17d25e80",
"oa_license": "CCBY",
"oa_url": "https://pubs.rsc.org/en/content/articlepdf/2015/cp/c5cp00101c",
"oa_status": "HYBRID",
"pdf_src": "Anansi",
"pdf_hash": "aaab1c594b3018b3ca74f931e92f0f0b461e194b",
"s2fieldsofstudy": [
"Chemistry"
],
"extfieldsofstudy": [
"Medicine",
"Chemistry"
]
}
|
225077092
|
pes2o/s2orc
|
v3-fos-license
|
Impact of Oscillating Positive Expiratory Pressure Device Use on Post-Discharge Hospitalizations: A Retrospective Cohort Study Comparing Patients with COPD or Chronic Bronchitis Using the Aerobika® and Acapella® Devices
Purpose Managing and preventing disease exacerbations are key goals of COPD care. Oscillating positive expiratory pressure (OPEP) devices have been shown to improve clinical outcomes when added to COPD standard of care. This retrospective database study compared real-world resource use and disease exacerbation among patients with COPD or chronic bronchitis prescribed either of two commonly used OPEP devices. Patients and methods Patients using the Aerobika® (Trudell Medical International, London, ON, Canada) or Acapella® (Smiths Medical, Wampsville, New York, USA) OPEP device for COPD or chronic bronchitis were identified from hospital claims linked to medical and prescription claims between September 2013 and April 2018; the index date was the first hospital visit with an OPEP device. Severe disease exacerbation, defined as an inpatient visit with a COPD or chronic bronchitis diagnosis, and all-cause healthcare resource utilization over 30 days and 12 months post-discharge were compared in propensity score (PS)-matched Aerobika device and Acapella device users. Results In total, 619 Aerobika device and 1857 Acapella device users remained after PS matching. After discharge from the index visit, Aerobika device users were less likely to have ≥1 severe exacerbation within 30 days (12.0% vs 17.4%, p=0.01) and/or 12 months (39.6% vs 45.3%, p=0.01) and had fewer 12-month severe exacerbations (mean, 0.7 vs 0.9 per patient per year, p=0.01), with significantly longer time to first severe exacerbation than Acapella users (log-rank p=0.01). Aerobika device users were also less likely to have ≥1 all-cause inpatient visit within 30 days (13.9% vs 20.3%, p<0.001) and 12 months (44.9% vs 51.8%, p=0.003) than Acapella users. Conclusion Patients receiving the Aerobika OPEP device, compared to the Acapella device, had lower rates of subsequent severe disease exacerbation and all-cause inpatient admission. This suggests that Aerobika OPEP device may be a beneficial add-on to usual care and that OPEP devices may vary in clinical effectiveness.
Introduction
Chronic obstructive pulmonary disease (COPD) is a major source of morbidity and affects 6% of adults over the age of 40 and 10% of adults 65 years or older in the United States. 1 The clinical course of COPD is punctuated by exacerbations, characterized by acute worsening of symptoms such as increased sputum production, airway inflammation, and breathlessness that require additional treatment. 2,3 Severe cases require emergency department (ED) visits and/or hospitalizations. 3 COPD exacerbations are common but often underreported, since as many as half of patients experiencing an exacerbation may not seek medical attention. 4 One community-based study found that COPD patients had 1.5 disease exacerbations per year based on clinic visits, but 2.7 exacerbations per patient per year after including symptom patterns recorded in daily diary cards. 4 COPD exacerbations are associated with long-term implications, including worsened quality of life, 4,5 faster decline in lung function, 6 and increased risk of allcause mortality. 7,8 The burden of COPD exacerbations becomes even greater among patients with longer recovery times 9 and those with frequent exacerbations. 6 Previous exacerbations are the most important predictor of future ones, 10,11 and about one in five COPD patients hospitalized for an exacerbation will require re-hospitalization within 30 days of discharge. 12 Thus, preventing disease exacerbations and appropriately treating exacerbations to minimize their consequences are key goals of COPD management. 3,13,14 Current treatment guidelines recommend rescue inhalers, corticosteroids, and antibiotics for the treatment of severe exacerbations and maintenance medications for the management of COPD when stable. 3,13,14 Oscillatory positive expiratory pressure (OPEP) devices, such as the Aerobika ® device (Trudell Medical International, London, Ontario, Canada) and Acapella ® (Smiths Medical, Wampsville, New York, USA), are non-pharmacological interventions that ease sputum clearance through positive pressure to hold airways open combined with airway oscillations that thin mucus [15][16][17] and have low safety risks compared to pharmacologic therapies. 18 Clinical evidence suggests that some OPEP devices not only help clear mucus but also improve short-term lung capacity, exercise capacity, and quality of life in patients with stable COPD 17,19,20 and cystic fibrosis. 21 While adjunct therapy with OPEP devices can potentially provide additional benefit to the management of COPD, evidence from large clinical trials or real-world studies on the benefit of these devices in relation to COPD disease exacerbations is limited. One database study reported that the Aerobika OPEP device reduces 30-day exacerbations by nearly 30% compared to no positive expiratory pressure (PEP) or OPEP device use following a hospital visit, 22 and a clinical trial reported that patients who used an adjunct Acapella device had a 1-day reduction of hospital length of stay for acute exacerbation compared to those who received physician's choice of standard COPD management. 23 Moreover, there are functional differences among a few existing OPEP devices, as revealed by several laboratory studies using simulated lung models, which may impact clinical efficacy. In particular, the Aerobika OPEP device has been shown to have the most consistent pressure amplitude, highest mean pressure amplitude at higher resistance settings, 24,25 and highest total pressure pulse impact 26 compared to Acapella and other OPEP devices. However, clinical studies have been limited to comparisons of the efficacy of adjunct OPEP devices to usual care in COPD, rather than head-to-head comparisons. 18 To that end, this real-world study aimed to compare healthcare resource use (HRU) and severe disease exacerbations 30 days and 12 months post-discharge in patients with COPD or chronic bronchitis, a subset of COPD characterized by mucus hypersecretion, 27 treated with two of the most commonly used OPEP devices in the hospital setting.
Data Sources
This retrospective cohort study used IQVIA's Hospital Charge Detail Master (CDM) hospital database from September 1, 2012 to April 30, 2019 (study period) linked to professional medical claims (Dx) and longitudinal prescription claims (LRx) databases (IQVIA, Inc., Durham, NC, USA). The CDM database includes service order records from over 650 hospitals, covering 7 million annual inpatient stays and 60 million annual outpatient visits. Patient-level data include healthcare services from hospital departments (inpatient, outpatient, ED, pharmacy), with detailed drug, procedure, and diagnosis codes from each encounter for the entire stay; each visit has both an admission date and discharge date. The Dx database captures over 1 billion pre-adjudicated claims and 3 billion records obtained annually from approximately 800,000 officebased physicians and specialists (75% of American Medical Association providers captured). Medical claims from ambulatory and general health care sites (ie, outpatient clinics associated with hospitals such as rehabilitation, same-day surgery, chemotherapy centers) are also included. The LRx database captures information on dispensed prescriptions with 92% coverage of prescriptions from the retail channel, 72% coverage of standard mail service, and 76% coverage of long-term care facilities. The study dataset was created based on Health Insurance Portability and Accountability Act (HIPAA)-compliant linking processes using IQVIA's patented and proprietary encryption algorithm. [28][29][30] As this retrospective cohort analysis was conducted using de-identified HIPAA-compliant data, Institutional Review Board (IRB) review was not required for this study. .X and J44.X, not including emphysema given that it is not characterized by sputum production 3 ). As a proxy of continuous enrollment, patient eligibility as well as pharmacy stability in LRx and provider stability in Dx in the 12-month pre-index (ie, baseline) and post-index (ie, follow-up) periods were required. Patients were excluded if they had evidence of the index OPEP device in the baseline period, non-index OPEP devices in the follow-up period (including the index date), diagnosis of asthma on or before the index date, OPEP device use in a post-operative setting (identified with a surgical procedure of interest 30 days before the index visit admission date or during the index visit), or incomplete data.
Study Population
Patients who met all inclusion and exclusion criteria were stratified into two mutually exclusive study cohorts based on the OPEP device used on the index date: patients treated with Aerobika OPEP device (Aerobika device cohort) or Acapella OPEP device (Acapella cohort); patients using other OPEP devices were not included in the analysis due to a possibility of misclassification when using hospital billing descriptions. The cohorts were then matched using propensity score (PS) matching at a 1:3 ratio with a greedy nearest-neighbor matching algorithm without replacement. The logistic regression model to generate the PS included age, sex, payer type, comorbid conditions, COPD-related medication use, and history of exacerbations (Tables 1 and 2 list these variables in detail).
Index year and geographic region were excluded from the PS model to limit misclassification of the index OPEP device and to maximize cohort sample sizes after matching.
Study Measures
The baseline demographics and clinical characteristics, including history of severe and moderate disease exacerbations, were evaluated during the fixed 12-month baseline period. A severe disease exacerbation was defined as an inpatient visit with COPD or chronic bronchitis diagnosis in any position and moderate disease exacerbation was defined as an ED visit with COPD or chronic bronchitis diagnosis or prescription of an oral corticosteroid (OCS) within 14 days of an office visit with a COPD or chronic bronchitis diagnosis in any position (that did not lead to an inpatient visit). 11,22,31 The follow-up measures included index visit characteristics and clinical and HRU outcomes among the matched cohorts. The index measures, evaluated from the admission date to the discharge date, were the initial care setting, length of stay, COPD-related medications and respiratory support. The follow-up measures, evaluated from the day after the discharge date to the end of 30-day post-discharge or 12-month follow-up, were severe exacerbations, moderate exacerbations, and all-cause HRU. Time following the index visit discharge date until the end of the 12-month follow-up period was also reported for each cohort as it varied depending on the duration of the index hospital visit (all patients had 12-month follow-up data following the index date [ie, the admission date on the index hospital record]); therefore, when reporting 12-month outcomes, counts for HRU categories were standardized and reported as per patient per year (PPPY).
Statistical Analyses
Baseline patient characteristics between the cohorts before and after PS matching were assessed using standardized difference, applying a commonly used threshold of >10% in absolute standardized difference to determine imbalance. 32 Generalized estimating equations for generalized linear models without additional covariate adjustment were used to compare continuous and categorical outcome measures, respectively, between the PS-matched cohorts. Kaplan-Meier survival analysis with log-rank test was also used to compare time from index visit discharge to first severe disease exacerbation between the PS-matched cohorts. Generalized estimating equations for generalized linear models were used to estimate the association between index OPEP device and the number of severe disease exacerbations PPPY (with normal distribution and log link function) and between index OPEP device and the odds of experiencing at least one 12-month severe disease exacerbation (with binomial distribution and logit link function); the models additionally adjusted for the Notes: COPD and chronic bronchitis diagnoses were identified using ICD-9 codes 491.XX and 496 and ICD-10 codes J41.X and J44.X, not including emphysema given that it is not characterized by sputum production. 3 Pharmacy and provider stability was used as a proxy for continuous enrollment, which was defined as stability of reporting claims in every month for at least one pharmacy in LRx or provider in Dx as well as at least one LRx claim for any medication and one Dx claim in the baseline period and the 12-month period after the index date. There were no patients excluded for incomplete data.
Patient Characteristics
In total, 5029 patients met the inclusion criteria (662 Aerobika device and 4367 Acapella users; Figure 1). After 1:3 PS matching, the remaining study population consisted of 619 patients receiving the Aerobika device matched to 1857 patients receiving the Acapella device.
In the cohorts before PS matching, the baseline patient demographic and clinical characteristics were generally balanced. The cohorts were similar in terms of age (mean age, 72.5 years and 71.8 years for Aerobika device cohort and Acapella cohort, respectively), payer type (74.3% and 72.8% had Medicare), and Charlson Comorbidity Index (CCI) score (mean, 3.7 and 3.8; all absolute standardized differences ≤10%). Further details of the cohorts before matching are described in Supplemental Tables 1 and 2. After matching, all baseline characteristics of the Aerobika device and Acapella users used in the PS model were well balanced. Patients in both cohorts were 72 years old on average, and more than half of the patients (54-55%) were female (Table 1). Geographic region and index year distribution (not included in the PS model) differed between the matched cohorts, with the majority of the Aerobika device cohort (55.4%) residing in Midwestern states and the majority of the Acapella cohort residing in Southern states (51.3%). All baseline clinical characteristics and HRU were well balanced after PS matching. The mean CCI score 33 was 3.7, and the most common comorbidities of interest were hyperlipidemia (56.7% of the Aerobika device cohort and 57.3% of the Acapella cohort), cardiovascular disease (44.1% and 43.2%), and anxiety/depression (35.9% and 35.1%) ( Table 2). During the baseline period, about half of patients in each cohort experienced at least one severe (45.4% of the Aerobika device cohort and 44.5% of the Acapella cohort) or moderate (46.0% and 45.7%) disease exacerbation. About one-fourth of patients also had oxygen therapy in the baseline period (29.4% of the Aerobika device cohort and 26.6% of the Acapella cohort).
Index Visit Characteristics
Most patients in both cohorts visited the ED first as part of the index visit (82.4% vs 72.5% of matched Aerobika device and Acapella users, respectively, p<0.001); the remaining patients were admitted to inpatient directly (13.2% vs 23.2%, p<0.001) or visited an outpatient department (4.4% and 4.3%, p=0.95) (Figure 2A). Nearly all patients, regardless of initial care setting during the index visit, eventually had an inpatient admission during the index visit (91.8% [n=568] of Aerobika device users and 94.9% [n=1762] of Acapella users; p=0.004) ( Figure 2B). Among those with an inpatient stay, the length of stay was a week on average (mean days, 7.0 for the Aerobika device users vs 7.5 days for Acapella users, p=0.09). There were some differences in disease-and exacerbationrelated treatments received during patients' inpatient stays. For example, the Aerobika device users had lower use of SABA (44.5% vs 53.5%), antibiotics (83.1% vs 88.1%), OCS (29.9% vs 50.8%), and LAMA (11.4% vs 20.4%), but higher use of LABA (10.9% vs 5.7%) and oxygen therapy (93.0% vs 74.0%; all p-values<0.05) (Figure 3).
Disease Exacerbations Post-Discharge
The duration of the "12-month post-discharge" period, from the day after the index visit discharge date to the end of the 12-month follow-up period, was on average 354.5 (standard deviation, 5.5) days for the Aerobika device users and 353.9 (5.2) days for the Acapella users (p=0.01). The proportion of patients who experienced at Notes: All baseline clinical characteristics and healthcare resource utilization were balanced after PS matching. a Use of combination therapy ICS/LABA/LAMA was also assessed but not reported due to occurrence in less than 1% of patients. b Defined as an inpatient admission with COPD or chronic bronchitis diagnosis, not including the index date. c Defined as an emergency department visit with COPD or chronic bronchitis diagnosis or prescription for an oral corticosteroid within 14 days of an office visit with COPD or chronic bronchitis diagnosis, not including the index date. Table 3). The significant findings for differences in severe exacerbations persisted in regression models which adjusted for a few additional baseline and index visit characteristics (see Table 4 for these variables). Compared to Acapella users, Aerobika device users had 17% fewer severe exacerbations PPPY (estimate, 0.83; Notes: a Severe exacerbation was defined as an inpatient admission with COPD or chronic bronchitis diagnosis, anytime during the follow-up period, not including the index visit b Defined as inpatient admission within 30 days following index hospitalization discharge date, assessed among patients with index inpatient visit (n=568 Aerobika device users and n=1762 Acapella users). Abbreviations: ED, emergency department; HRU, healthcare resource utilization; PPPY, per patient per year; SD, standard deviation.
95% CI, 0.71-0.96) ( Table 4) and had 20% lower odds of 12-month severe disease exacerbation (odds ratio, 0.80; 95% CI, 0.66-0.98) ( Table 5). Index hospital visit being inpatient (whether admitted directly or from ED) compared to ED or outpatient hospital visit without inpatient admission and OCS use during the index visit were also significant predictors of increased number of 12-month severe disease exacerbations PPPY (Table 4) and having at least one post-discharge severe disease exacerbation (Table 5). Furthermore, time from the index visit discharge date to the first severe exacerbation was similar between Aerobika device users and matched Acapella users (102.0 vs 96.6 days, p=0.44). When accounting for the duration of follow-up in Kaplan-Meier survival analysis, Aerobika device users had significantly longer times to severe disease exacerbation compared to Acapella users (log-rank p=0.01) (Figure 4).
All-Cause Post-Discharge HRU
Those who received the Aerobika device during the index visit were less likely to have at least one early (within 30 days post-discharge) all-cause inpatient readmission (13.9% for Aerobika device cohort vs 20.3% for Acapella cohort, p<0.001) and at least one all-cause inpatient admission 12 months post-discharge (44.9% vs 51.8%, p=0.003). Aerobika device users also had fewer all-cause 12-month inpatient visits (mean PPPY, 0.9 vs 1.1, p=0.003); the mean length of stay was 8 days for both cohorts (p=0.28). There was no significant difference in all-cause ED visits or outpatient visits between the cohorts (Table 3).
Discussion
To our knowledge, this is the first study to compare the impact of two commonly used OPEP devices on short-and long-term clinical and resource use outcomes in patients with COPD or chronic bronchitis. In the cohorts both before and after matching, patients with the Aerobika device or Acapella device were typically elderly (about half were 75 years of age or older), were predominantly female, had multiple Charlson comorbidities, and had a history of at least one severe or moderate disease exacerbation during the baseline period. After matching to ensure outcomes were compared in similar patient populations, we observed that during the index visit with OPEP device use and COPD or chronic bronchitis diagnosis, more than 90% of patients had an inpatient admission, consistent with the definition of a severe disease exacerbation used in this study. Following the index visit, those who received the Aerobika device had significantly lower rates of allcause hospitalization and severe disease exacerbation within 30 days of the index visit discharge date, with sustained differences over 12 months of follow-up, and fewer moderate exacerbations over 12 months of followup. Our findings are supported by previous, albeit limited, research suggesting the clinical benefits of the Aerobika OPEP device in the treatment of COPD and differences in efficacy across different OPEP devices. In a small crossover study, 27 COPD patients who used the Aerobika OPEP device daily for 4 weeks had subsequent improvements in Patient Evaluation Questionnaire (PEQ)-ease-bringing-up-sputum score and forced expiratory volume in 1 s (FEV 1 ) related to improved ventilation. 17 These improvements in airway clearance may also improve disease exacerbation outcomes demonstrated in a previous database study by Burudpakdee et al,22 where Aerobika device users had significantly lower rates of 30-day severe disease exacerbations (13.8%, consistent with the 12.0% reported in our study) compared to matched controls who did not receive an OPEP device (vs 19.0%). To our knowledge, the only study of Acapella effectiveness in treating acute COPD exacerbations was a clinical trial comparing patients who received an adjunct Acapella device (PEP or OPEP) to matched retrospective controls who received standard of care alone; this study found that patients who used Acapella had a 1-day reduction in the hospital length of stay. 23 Although OPEP devices share the general objective of opening airways and using airway oscillations to loosen mucus, the device mechanisms and consequently the pressure pulse waveforms that they generate differ, and several studies have reported better performance characteristics related to airway clearance for the Aerobika device compared to Acapella. [24][25][26] Therefore, the present study provides new evidence on how functional differences between the devices may translate to differences in preventing severe disease exacerbations, illustrated by the lower rate of subsequent inpatient admissions due to severe disease exacerbation in Aerobika device users compared to Acapella users in the PS-matched comparative analyses and further confirmed in adjusted analyses, as well as preventing moderate exacerbations and allcause hospitalizations that may be due to worsening of overall quality of life 4,5 and worsening comorbid conditions (eg, hospitalizations due to cardiac events 34 ). Considering previous COPD exacerbations are the most important predictor of future events, 10,11 this study demonstrates that the Aerobika device may mitigate the clinical and economic impact of exacerbations when used as an adjunct therapy in a population known to have a high economic burden. 35 Aside from functional differences, high patient satisfaction with the Aerobika device and patient-reported ease of use, as revealed in a survey of Aerobika device users, 36 may support real-world adherence to the Aerobika OPEP device and contribute to improved outcomes. However, considering the lack of published data, future research comparing adjunct OPEP device use to standard of care, to other commercially available OPEP devices, and to PEP devices is needed to provide guidance on how to incorporate OPEP devices into clinical practice. The strengths of this study are the use of three linked databases to obtain a comprehensive view of HRU and treatments before and after the initial visit with OPEP device use and the statistical methodology to limit bias in the results. PS methods were used to mimic the selection process of clinical trials and create comparable cohorts of Aerobika device and Acapella device users for evaluating outcomes. 37 Then, our findings that Aerobika device users were less likely to have a severe disease exacerbation over 12 months of follow-up compared to Acapella users persisted after adjusting for geographic region and characteristics of the index visit in regression models, which added to the internal validity of the findings.
With that said, a few limitations should be noted. First, patients with COPD and chronic bronchitis were identified using ICD diagnosis codes, and confirmation of diagnosis would require clinical measures not available in claims databases. To address this limitation, we required a diagnosis during the hospital visit with OPEP device therapy and excluded those with evidence of asthma or surgical procedure for which the OPEP device could also be used. Second, data on disease severity were not available, which would provide context on the generalizability of the study population. For example, some physicians may selectively prescribe OPEP devices to patients with a history of frequent severe exacerbations, 38 but severe patients at risk of post-discharge mortality may be underrepresented, given our study requirement of 12 months of follow-up data. Thirdly, it is not known from the claims data who administered the OPEP device, whether patients used the OPEP devices with the proper techniques or with the appropriate frequency, if the devices were used with nebulizers, or what other physiotherapy treatments (eg, breathing programs) may have been administered, which could impact device efficacy and outcomes. Lastly, the claims data are not able to identify factors leading to OPEP device use such as potential regional and healthcare facility-level differences in practice patterns that may lead to residual confounding.
Conclusion
This study demonstrates that the Aerobika OPEP device significantly reduces all-cause inpatient visits and severe disease exacerbations, including 30-day inpatient readmissions and 12-month inpatient visits, compared to Acapella, when added to standard of care among patients with COPD and chronic bronchitis. Combined with previous clinical and real-world data, these findings further support the use of the Aerobika OPEP device as an add-on to usual care for the treatment of severe COPD exacerbations and highlights the benefit of the Aerobika device versus an alternative OPEP device.
Funding
This study was funded by Monaghan Medical Corporation and Trudell Medical International.
|
2020-10-28T05:08:52.970Z
|
2020-10-19T00:00:00.000
|
{
"year": 2020,
"sha1": "dff0e6fc47e50147da7092c5f227d6490a734de9",
"oa_license": "CCBYNC",
"oa_url": "https://www.dovepress.com/getfile.php?fileID=62662",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "dff0e6fc47e50147da7092c5f227d6490a734de9",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
159444890
|
pes2o/s2orc
|
v3-fos-license
|
Examining the Factors Behind the Success and Sustainability of China ’ s Creative Research Group : An Extension of the Teamwork Quality Model
The Creative Research Group (CRG) is the special high-level scientific and innovation team funded by the National Natural Science Foundation of China to promote basic research at the frontiers of science. In general, there are problems of “structure anomie” and “cooperation inefficiency” in the operations and teamwork dynamics within the nationwide CRG project. By extending and developing the two-stage teamwork quality model, this study aims to depict and analyze the impact factors behind the success and sustainability of the Creative Research Group (SSCRG), and reveal the relationships among them. Herein, rationality of team structure (ROTS) is used to describe the rationality of team formation and structure collocation, teamwork quality (TWQ) is used to describe the process of team members’ integration and cooperation, and SSCRG includes team performance, personal success and team comprehensive impacts. The results show that ROTS and TWQ have different influences on SSCRG, while TWQ is the key mediation factor between ROTS and SSCRG. In summary, the model built describes the complex phenomena and relationships in the teamwork of the CRG, and helps us to understand and solve the problems of structure and cooperation. Although the limitations lie in the specific samples and research methods, the extension and migration of classical models and theories would help to further deepen such research and contribute to the governance and development of such innovation teams.
Introduction
Although basic research in China has made considerable progress, there is still a gap in its overall strength compared with major developed countries, according to the report of the National Natural Science Foundation of China (NSFC) and the "NSFC Strategic Plan for 2016-2020 (The 13th Five-Year Plan of NSFC)".It has been claimed that, "Firstly, there are few significant original achievements with international influence and there is a lack of the ability to create important new disciplines and directions; secondly, there is a shortage of world-class scientists leading the scientific trends and the environment for the growth of young talent needs to be improved; thirdly, the role of basic research in promoting economic and social development and safeguarding national security needs to be enhanced [1]."The lack of major original achievements and the top talent team has become a key factor restricting the development of basic research and the construction of an innovative country in China.Original achievements are achieved by scientific researchers, so the key is to pay attention to the training of scientific researchers and top teams.Yang Wei, the former director of the NSFC wrote in the journal Nature in 2016 that China should pay more attention to basic research and improving the quality of basic research should be the focus of China's efforts to promote national innovation [2].In terms of innovation, policy and institutional arrangements are particularly important: China is paying greater attention to the development of basic research and the cultivation of young talent as the core of its innovation-driven development strategy.The projects funded by relevant government departments include the Innovation Team Project of the Ministry of Education, the Innovation Team Project of Chinese Academy of Sciences, and the Creative Research Group (CRG) Project of the NSFC.The most representative innovation team project in the field of basic research field is the CRG Project funded by the NSFC.
As one of the most important talent projects under NSFC, as well as the most representative innovation team project leading basic research, the CRG Project undertakes the important missions in breaking through the frontiers of basic research and training top talent teams.According to the situation notifications and project management measures of NSFC, the CRG project supports outstanding young and middle-aged scientists as academic leaders and the backbone of research to carry out innovative research around an important research direction, and cultivate and foster research groups that occupy a place at the forefront of international science [3].
In early 2016, at the annual performance evaluation meeting of the NSFC, which was co-hosted by the National Center for Science and Technology Evaluation (NCSTE) and the NSFC, experts from different departments of the NSFC who have participated in the CRG project and experts from relevant scientific research management departments peer-reviewed and evaluated the CRG project as whole.In addition to affirming the outstanding achievements of the CRG project in cultivating leading talents, enhancing teamwork and concentrating on solving major scientific problems, the experts also found and summarized some common problems: firstly, there is "collusion" and "patchwork" in the application and team structure of the CRG Project.The composition of some teams is not based on the combination of research needs, but rather based on seniority and titles, which subsequently restricts the development of potential youth teams.Secondly, the main participants of the CRG project are often at odds with each other, and an effective teamwork mechanism has not been formed.How to carry out in-depth cooperative research around common goals, genuinely solve one or two important frontier scientific problems, achieve influential results, cultivate outstanding talents, and establish a sustainable team all remain major challenges at present.
Although CRG is an innovation team with Chinese characteristics, there are many similarities between it and other innovative teams in terms of the team cooperation and further development.Existing classical team theories and models are exploited by some scholars for studying and explaining the cooperation and structural problems of the CRG [4].For the problems of "structure anomie" of CRG projects, there are some classical studies on team structure for team performance and team success [5][6][7].A good team structure is the prerequisite and important factor for team development.For the problems of "cooperation inefficiency" of CRG projects, few studies are directly related.However, the impacts of teamwork on team performance, personal success, project success, team success and sustainability have been widely studied in the field of organizational science and technological innovation [5][6][7][8].Since the 21st century, many studies have combined economics, sociology, organizational behavior and psychology to develop theories for explaining or predicting teamwork, team performance and team success and sustainability [9][10][11][12][13][14][15][16][17].
The theoretical model of teamwork quality (TWQ) is often used to study the relationship between teamwork factors and team performance, team members' personal success, team success and sustainability, including the research and evaluation of a research and development (R&D) team in enterprises [8,[12][13][14][15][16], medical care teams [18,19], learning groups in education [20], and agile team research in software development [21,22].However, although previous studies have used the TWQ model to explain the relationship between teamwork and different teams, few studies have examined high-level research and innovation teams, especially those working in the frontier areas of basic research and in innovative groups funded by government funds.At the same time, in the relevant literature, many studies have examined scientific research organizations and teamwork [22][23][24][25][26], but few have used the TWQ model to analyze the impact of team structure and teamwork on a team's success and sustainability.
Teamwork quality is directly related to "cooperation inefficiency" [8,[12][13][14][15][16], and team members' collocation and rationality are directly related to the "structure anomie" [23][24][25].In this study, the related research on team structure is included in terms of the extension of the TWQ model.Besides, the problems of "structure anomie" and "cooperation inefficiency" encountered by CRG in its development are explored and discussed.We use the extended model to analyze the empirical data, and to understand the teamwork factors related to the success and sustainability of the project team.Although the CRG is a high-level scientific research and innovation team with Chinese characteristics, this study is of great significance to developing related theories and the TWQ model, as applied to scientific research cooperation and basic research.The implications of this study could be taken into consideration during the development and evaluation of policies related to science and technology, as well as when considering how to improve the structure and teamwork quality of scientific research teams, thus contributing to the CRG to achieve high-level personnel training and major scientific breakthroughs.
Literature Review
Many studies have demonstrated the impacts of TWQ on team performance, project success, team success and sustainability [8,[12][13][14][15][16][17].What are the specific contents that affect TWQ, and what is its specific and extended theoretical model of TWQ when predicting the success and sustainability of the Creative Research Group (SSCRG)?Based on previous studies and the current state of the CRG, we propose the theoretical model of the extended version of teamwork quality, which takes TWQ, the rationality of team structure (ROTS), and SSCRG as its main constructs and impact factors.
Teamwork Quality (TWQ)
Hoegl and Gemuenden (2001) developed and constructed the theoretical model of TWQ, which is a comprehensive conceptual model of teamwork that has been highly recognized in academia [8].In order to better measure and analyze the nature of teamwork and open the "black box" of its dynamics, combined previous research and practical research, Hoegl and others have subdivided TWQ into six important dimensions or variables: communication, coordination, balance of member contributions, mutual support, effort and cohesion [8].Based on an empirical analysis of the survey data, it has been concluded that TWQ is positively related to team performance, the personal success of team members, and project success [8].For many years, Hoegl and his collaborators continued to publish research on teamwork and TWQ-related factors in important journals such as Organization Science, Research Policy and so on [12][13][14][15][16][17].By using the TWQ model, Hoegl and Proserpio discussed the relationship between team members' proximity and team performance [12]; Hoegl, Weinkauf, and Gemuenden's study provided support for the hypotheses predicting positive relationships between interteam coordination, project commitment, and teamwork quality [13].Hoegl and Parboteeah investigated how team autonomy is related to teamwork quality [14], analyzed the influence of the teamwork quality on creativity-relevant skills [15], and made a theoretical extension and hypothetical analysis of the team reflexivity [16].Hoegl, Ernst and Proserpio discussed the important role of high-quality team cooperation and team leaders in team performance [17].Based on the TWQ model, the above studies explore the relationships between teamwork-related factors of R&D and innovation team by adding a new variable or factor or expanding the model to a certain extent.This deserves to be learned and expanded by innovation teams in universities and research institutions, especially the CRG.
TWQ is an important factor or mediating variable for studying the rationality of team structure and team integration in forming cooperation mechanisms and their impact on team performance, team success and sustainability.It has been gaining more and more traction in the academic community [18][19][20][21][22]. Goebel, Guo and Wood explored how medical experts work together in interdisciplinary teams to achieve a good quality of collaboration [18].Thomas, Williams and Reichman discussed the communications and teamwork in neonatal resuscitation [19].Cureu and Pluut explored the mediating role of teamwork quality between learning group composition differences and cognitive complexity [20].Lindsjørn and co-workers studied the impact of team cooperation quality on many factors (learning and work satisfaction, team performance) in software development teams [21].Freire and co-workers, based on Bayesian network models and case studies, argued that teamwork quality is critical to the success of agile teams [22].These above research works are also representatives of TWQ's mutating and migrating to different fields.Meanwhile, research has begun to apply the TWQ model to the research and innovation team to explore the team cooperation mechanism of the CRG [4].As shown in Figure 1, Hoegl and co-workers developed the TWQ model to analyze the related factors of teamwork and team success [8,[12][13][14][15][16][17].Pluut explored the mediating role of teamwork quality between learning group composition differences and cognitive complexity [20].Lindsjørn and co-workers studied the impact of team cooperation quality on many factors (learning and work satisfaction, team performance) in software development teams [21].Freire and co-workers, based on Bayesian network models and case studies, argued that teamwork quality is critical to the success of agile teams [22].These above research works are also representatives of TWQ's mutating and migrating to different fields.
Meanwhile, research has begun to apply the TWQ model to the research and innovation team to explore the team cooperation mechanism of the CRG [4].As shown in Figure 1, Hoegl and co-workers developed the TWQ model to analyze the related factors of teamwork and team success [8,[12][13][14][15][16][17].Hoegl and Gemuenden's TWQ model mainly focuses on the teamwork of innovation teams [8].The CRG is a typical scientific research and innovation team in the basic research field with Chinese characteristics [4].It is of great relevance to use the TWQ model, which specializes in analyzing the innovation in team cooperation, to carry out this research.In recent years, the TWQ model has been gradually expanded to innovative team research in various fields.The theoretical model used in the research of the teamwork in innovation teams is either a simple causal hypothesis model [9,10], or a variant and extension based on the TWQ model [12][13][14][15][16][17].In this paper, we apply the TWQ model to the CRG in China, using the same six dimensions as well as other internal cooperation factors that are specific to the CRG.Our criteria and basis for judging the degree of TWQ is whether the internal cooperation in CRG teams has achieved communication, coordination, balance of member contributions, mutual support, effort and cohesion, and whether effective information (including communication, coordination, etc.), incentive compatibility (including contribution balance, mutual support, etc.) have also been achieved.
Success and Sustainability of the Creative Research Group (SSCRG)
The success and sustainability of project teams reflect many components, and are also the result of teamwork development.Previous scholars have distinguished and summarized two general kinds of results: task-related results (e.g., task completion quality, degree of compliance with budget and plan, etc.) and human-related results (e.g., team members' satisfaction, personal success, team feasibility, stability, continuity, team cooperation quality, mechanism, etc.) [8,[12][13][14][15][16][17].Based on the research of Hoegl and others, we believe that the CRG has the dual attributes of talent cultivation and frontier exploration, and so the success and sustainability of a project team should include team performance and personal success.
For sustainability, Cheng and Zhang believed that the sustainability of innovation teams included the ongoing operation of the team, guarantee of the team members' benefit, the sustainable Hoegl and Gemuenden's TWQ model mainly focuses on the teamwork of innovation teams [8].The CRG is a typical scientific research and innovation team in the basic research field with Chinese characteristics [4].It is of great relevance to use the TWQ model, which specializes in analyzing the innovation in team cooperation, to carry out this research.In recent years, the TWQ model has been gradually expanded to innovative team research in various fields.The theoretical model used in the research of the teamwork in innovation teams is either a simple causal hypothesis model [9,10], or a variant and extension based on the TWQ model [12][13][14][15][16][17].In this paper, we apply the TWQ model to the CRG in China, using the same six dimensions as well as other internal cooperation factors that are specific to the CRG.Our criteria and basis for judging the degree of TWQ is whether the internal cooperation in CRG teams has achieved communication, coordination, balance of member contributions, mutual support, effort and cohesion, and whether effective information (including communication, coordination, etc.), incentive compatibility (including contribution balance, mutual support, etc.) have also been achieved.
Success and Sustainability of the Creative Research Group (SSCRG)
The success and sustainability of project teams reflect many components, and are also the result of teamwork development.Previous scholars have distinguished and summarized two general kinds of results: task-related results (e.g., task completion quality, degree of compliance with budget and plan, etc.) and human-related results (e.g., team members' satisfaction, personal success, team feasibility, stability, continuity, team cooperation quality, mechanism, etc.) [8,[12][13][14][15][16][17].Based on the research of Hoegl and others, we believe that the CRG has the dual attributes of talent cultivation and frontier exploration, and so the success and sustainability of a project team should include team performance and personal success.
For sustainability, Cheng and Zhang believed that the sustainability of innovation teams included the ongoing operation of the team, guarantee of the team members' benefit, the sustainable innovation of the team and other factors [26].Some studies demonstrated that the sustainability of innovation teams is directly related to team composition, team member's communication, cooperation, integration and other factors directly related to the teamwork quality and team structure.In this context, teamwork quality and team structure are two important factors influencing sustainability [26][27][28].Compared to team success at the static level, team sustainability is more focused on dynamic level considerations.In this study, we combine team success with sustainability as an important construct (SSCRG) that is influenced by ROTS and TWQ.
The CRG's comprehensive impact on science, technology, economy, and society is also a core goal of the project as whole.These are important indices that have been emphasized by the NCSTE and NSFC, and have been applied in the evaluation of science and technology.While many scholars are developing new ways to measure the academic impact of scholars or teams [29,30], they are also constantly expanding their evaluation of the comprehensive impacts of scholars, teams and scientific research results [31,32].Wolf, Lindenthal and many scientists believed that the existing research evaluation focused on the scientific impact of research teams and research results, and should gradually extend the research evaluation to the comprehensive evaluation of scientists and their teams, taking into account the comprehensive impact of their activities and achievements on economic and social aspects [31].Therefore, combined with the existing requirements for the evaluation of the comprehensive impact of CRG, and the research of relevant scholars, the theoretical model of TWQ is expanded here to include relevant elements of the SSCRG, which includes three variables in this paper: team performance, personal success, and comprehensive impacts (mainly, scientific and technological, economic, and social impacts.The SSCRG means the staged completion of the innovative project, the excellent evaluation of the project, the continuous operation of the team, the guarantee and possibility of the interests and benefits of team members, and the potential value and likelihood of continuous innovation, continuous growth and transition of the team in the future [4,8,[12][13][14][15][16][17]23,26,31].The criteria for our judgment and evaluation are whether the CRG has successfully completed their research project; achieved their research goals and the cultivation of talents and teams; whether the individual members of the team have grown and attained greater motivation as researchers; and whether the team has made an important impact on academia and beyond.
As shown in Figure 2, we developed a first-stage extension of the TWQ model for analyzing the impact factors of the SSCRG.On the basis of the TWQ model developed by Hoegl and his collaborators, we define SSCRG and point out that it includes team performance, personal success, and team comprehensive impacts.
innovation of the team and other factors [26].Some studies demonstrated that the sustainability of innovation teams is directly related to team composition, team member's communication, cooperation, integration and other factors directly related to the teamwork quality and team structure.In this context, teamwork quality and team structure are two important factors influencing sustainability [26][27][28].Compared to team success at the static level, team sustainability is more focused on dynamic level considerations.In this study, we combine team success with sustainability as an important construct (SSCRG) that is influenced by ROTS and TWQ.
The CRG's comprehensive impact on science, technology, economy, and society is also a core goal of the project as whole.These are important indices that have been emphasized by the NCSTE and NSFC, and have been applied in the evaluation of science and technology.While many scholars are developing new ways to measure the academic impact of scholars or teams [29,30], they are also constantly expanding their evaluation of the comprehensive impacts of scholars, teams and scientific research results [31,32].Wolf, Lindenthal and many scientists believed that the existing research evaluation focused on the scientific impact of research teams and research results, and should gradually extend the research evaluation to the comprehensive evaluation of scientists and their teams, taking into account the comprehensive impact of their activities and achievements on economic and social aspects [31].Therefore, combined with the existing requirements for the evaluation of the comprehensive impact of CRG, and the research of relevant scholars, the theoretical model of TWQ is expanded here to include relevant elements of the SSCRG, which includes three variables in this paper: team performance, personal success, and comprehensive impacts (mainly, scientific and technological, economic, and social impacts.The SSCRG means the staged completion of the innovative project, the excellent evaluation of the project, the continuous operation of the team, the guarantee and possibility of the interests and benefits of team members, and the potential value and likelihood of continuous innovation, continuous growth and transition of the team in the future [4,8,[12][13][14][15][16][17]23,26,31].The criteria for our judgment and evaluation are whether the CRG has successfully completed their research project; achieved their research goals and the cultivation of talents and teams; whether the individual members of the team have grown and attained greater motivation as researchers; and whether the team has made an important impact on academia and beyond.
As shown in Figure 2, we developed a first-stage extension of the TWQ model for analyzing the impact factors of the SSCRG.On the basis of the TWQ model developed by Hoegl and his collaborators, we define SSCRG and point out that it includes team performance, personal success, and team comprehensive impacts.
Team Performance
Hoegl and his collaborators argued that team performance is reflected through effectiveness and efficiency.Team effectiveness refers to the degree to which the team meets the expectations of the quality of the results.In the case of innovation research projects, effective performance usually implies meeting the pre-defined objective attributes and degrees of project operation and
Team Performance
Hoegl and his collaborators argued that team performance is reflected through effectiveness and efficiency.Team effectiveness refers to the degree to which the team meets the expectations of the quality of the results.In the case of innovation research projects, effective performance usually implies meeting the pre-defined objective attributes and degrees of project operation and achievement, such as the functionality and robustness of results or products, the degree of final goal completion, and whether any major breakthroughs were made, etc. Team efficiency is achieved by adhering to the plan.For example, the degree to which tasks are completed according to budget and date is important; team effectiveness reflects the comparison between actual and expected results, while the efficiency evaluation is based on the comparison between actual and expected inputs [8].
The team performance of the CRG is evaluated based on the group's ability to complete the project on time, with products of good quality and of certain quantity; the ability to achieve the overall research goal; and whether it makes breakthroughs in the frontiers of certain basic research fields.
Personal Success
In addition to achieving performance goals, teams must work in ways that increase their ability to participate in future teamwork.Hoegl and his collaborators believe that satisfaction with team work will increase the enthusiasm of team members and their willingness to participate in projects now and in the future [8,[14][15][16].Denison et al. believe that the personal growth of members and the overall development of the team are mutually beneficial [33].
The personal success of the members of the CRG is a key component of the success of innovative group projects and team improvement.It includes not only the cultivation of graduate students, but also the growth and promotion of young and middle-aged scientists as the backbone of research, as well as the personal success of team leaders and other team members.Personal success, which is in line with the original intention of the establishment of the CRG Project, is also in line with the improvement and long-term development of the CRG from the team ability level down to the individual ability level.
Team Comprehensive Impacts (Scientific and Technological Impact and Economic, Social and Other Comprehensive Impacts)
Van Houten believes that academic influence is mainly reflected in how peers evaluate their academic research results, while the depth and breadth of academic influence can be reflected through research paper citations, which signifies how much the research is valued and recognized by others [34].Mainstream research judges the academic influence of scholars or teams mainly through bibliometric methods and scientometrics [35][36][37][38][39].Many studies use bibliometric methods to study the contribution and academic influence of scientists [35,36], while others use the methods of scientometrics and visual analysis to visualize the academic cooperation network and academic influence of scientists [37][38][39].With the development of information technology and network technology in this new era, the academic and social influence of a scholar or research team can be reflected not only by traditional academic platforms and co-citation networks constructed from publications and patents, but also by quantifiable and assessable data on the Internet and social media, for example, through the use of altmetrics which has emerged in recent years [29,30].
Some scholars have discussed the importance of understanding and evaluating the comprehensive influence of scientific research work and a research team in a certain field, and have made up for the shortcomings of pure objective data and measurement methods solely by means of questionnaires and interviews [31].Some scholars have also explored the economic and social impacts of teams and outcomes in terms of the transformation of scientific and technological achievements [32].
On 23 June 2016, the Ministry of Science and Technology of China officially abolished the appraisal method for scientific and technological achievements.The evaluation of scientific and technological achievements of related administrative departments at all levels was handed over to professional evaluation agencies by the entrusting party.This means that China is exploring and establishing a market-oriented evaluation mechanism for new scientific and technological achievements, and the evaluation of new scientific and technological achievements will be dominated by the market.The objective and fair evaluation of scientific, technological, economic and social value of scientific and technological achievements by third-party professional evaluation institutions will help investors, partners and government recognize such achievements more quickly, and facilitate the smooth progress of technological transactions.
The NCSTE was commissioned by the NSFC Planning Bureau from 2015 to 2016 for the annual project performance evaluation and conducted the research and evaluation for the CRG project.In the project evaluation forms, questionnaires and even field interviews, the academic and economic and social impacts of the CRG were taken as important factors and evaluation indicators.
Therefore, team comprehensive impacts are an important evaluation standard for SSCRG, especially sustainability.Whether the team has an important academic impact and comprehensive impacts; whether the results have important scientific and technological, economic, social and cultural impacts.All these are the requirements for the sustainable development of the CRG in the new era, consistent with changes in the identification standards and expected scope of China's scientific and technological achievements.
Rationality of Team Structure (ROTS)
Some studies have shown that the rationality of the team structure can have a positive impact on the quality of teamwork, project success, team success and sustainability [40][41][42].Stewart and Barrick believe that the rationality of the team structure has a positive impact on team performance, while the team structure also influences team performance through team self-leadership, teamwork mechanism, task complexity and other mediation factors [43].Yang and Tang used the social network method to explore the relationship between team structure and team performance.They found that team structure is a key factor for quality performance.Team cohesion is positively related to overall performance and the rationality of team structure.Then, through the benign characteristics and cohesion factors of the group, cooperative networks and mechanisms are formed at different stages, which have an impact on team performance [44].
Balkundi and Harrison believe that teams with close relationships among members can better achieve goals, while teamwork timelines, membership familiarity, and relationship ties can moderate the relationship between team structure and team performance [45].Some studies have shown that team structure does not necessarily have an absolute impact on the performance or success of the team.Team structure needs to be mediated or moderated by other variables or elements (such as a team cooperation mechanism, task complexity, team leadership, etc.) before it has an impact on team performance and overall project success [43,46].
Based on previous research, the rationality of team structure may have a direct or indirect positive impact on team performance, project success, team success and sustainability directly or indirectly.At the same time, it has also been found that on the basis of the rationality of team structure, the formation and development of cooperation and integration mechanism among team members also have an important influence on team performance and project success, team success and sustainability.Dynamic and process factors such as integration, communication, and mutual support among team members can be attributed to TWQ.
Therefore, the ROTS is closely related to the teamwork quality, team success and sustainability.The ROTS of the CRG means that the team structure and scale of the CRG is scientific and reasonable, and the distrubtion of age, background, education, title and posts range from prominent researchers to young researchers, graduate students and other members is reasonable.The group should have a strong collaborative foundation, strong likelihood of continuity and sustainability, and the potential to engage in high-level basic research and achieve cutting-edge breakthroughs.
Based on these descriptions of the role of TWQ and ROTS in SSCRG, combined with the previous research on TWQ and SSCRG by Hoegl and other scholars [8,[12][13][14][15][16][17], the previous research on ROTS by Hoegl and Weinkauf [42] and Balkundi and Harrison [45], and the latest research on the CRG teamwork mechanism by Gao and Ding [4], this study proposes that the second-stage extension of the TWQ model includes ROTS.Using the previous model as a foundation, we added ROTS, a key factor that has been shown to affect TWQ and SSCRG.As shown in Figure 3, TWQ and ROTS both may have a positive impact on SSCRG, and ROTS may have a positive impact on SSCRG.SSCRG consists of team performance, personal success and team comprehensive impacts.In this study, we analyze the possible positive impact of ROTS and TWQ on SSCRG and the possible impact of ROTS on TWQ.
Hypotheses Development
Based on the extended TWQ theoretical model, this study divides the impact factors of the SSCRG into TWQ and ROTS.
TWQ includes communication, coordination, balance of member contributions, mutual support, effort and cohesion; ROTS means that the team structure and scale of the CRG is scientific and reasonable, and distribution of age, background, education, titles and posts range from leading figures in the field to young researchers, graduate students and other members.
According to the TWQ theoretical model, the direct and main determinant of SSCRG is TWQ.Literature on the R&D teams of commercial organizations and on team cooperation mechanisms in high-level scientific and innovation teams show that TWQ has an important impact on SSCRG.Therefore, we present the following hypotheses (summarized in Table 1):
Hypothesis 1. TWQ has a positive impact on SSCRG.
According to Hoegl et al.'s research and the actual status of the CRG, this study expects the rationality of the team structure (ROTS) to have an positive impact on the communication and integration of team members, and that factors such as communication and integration can be attributed to TWQ.This leads to our second hypothesis.
Hypothesis 2. ROTS has a positive impact on TWQ.
According to the research of Hoegl and Weinkauf, ROTS has a positive impact on team performance, project success, team success and sustainability.At the same time, other research has also shown that ROTS is related to team performance, the personal success of team members, team success and sustainability.Therefore, we hypothesize that ROTS is positively correlated with SSCRG.
Hypothesis 3. ROTS has a positive impact on SSCRG.
We summarize the research hypotheses with theoretical structural constructs and variables in Table 1 as follows.
Hypothesis
Path Direction H1 TWQ has a positive impact on SSCRG TWQ → SSCRG H2 ROTS has a positive impact TWQ ROTS → TWQ H3 ROTS has a positive impact on SSCRG ROTS → SSCRG
Hypotheses Development
Based on the extended TWQ theoretical model, this study divides the impact factors of the SSCRG into TWQ and ROTS.
TWQ includes communication, coordination, balance of member contributions, mutual support, effort and cohesion; ROTS means that the team structure and scale of the CRG is scientific and reasonable, and distribution of age, background, education, titles and posts range from leading figures in the field to young researchers, graduate students and other members.
According to the TWQ theoretical model, the direct and main determinant of SSCRG is TWQ.Literature on the R&D teams of commercial organizations and on team cooperation mechanisms in high-level scientific and innovation teams show that TWQ has an important impact on SSCRG.Therefore, we present the following hypotheses (summarized in Table 1): Hypothesis 1. TWQ has a positive impact on SSCRG.
According to Hoegl et al.'s research and the actual status of the CRG, this study expects the rationality of the team structure (ROTS) to have an positive impact on the communication and integration of team members, and that factors such as communication and integration can be attributed to TWQ.This leads to our second hypothesis.
Hypothesis 2. ROTS has a positive impact on TWQ.
According to the research of Hoegl and Weinkauf, ROTS has a positive impact on team performance, project success, team success and sustainability.At the same time, other research has also shown that ROTS is related to team performance, the personal success of team members, team success and sustainability.Therefore, we hypothesize that ROTS is positively correlated with SSCRG.
Hypothesis 3. ROTS has a positive impact on SSCRG.
We summarize the research hypotheses with theoretical structural constructs and variables in Table 1 as follows.
Hypothesis Path Direction
H1 TWQ has a positive impact on SSCRG TWQ → SSCRG H2 ROTS has a positive impact TWQ ROTS → TWQ H3 ROTS has a positive impact on SSCRG ROTS → SSCRG Note: ROTS = Rationality of Team Structure; SSCRG = Success and Sustainability of Creative Research Group; TWQ = Teamwork Quality.
Research Method
This study adopts the two-step approach proposed by Anderson and Gerbing to test the reliability and validity of model factors and to perform hypothesis testing and model fitting [47].The number of valid samples for empirical analysis is 302, which are small and medium samples and mostly reflective the indicators we are interested in.Compared to the traditional structural equation model (SEM) analysis, our approach is more in line with partial least squares (PLS) analysis.PLS requirements for variable adaptability are more relaxed, and accuracy can also be maintained for a small sample size [48][49][50][51][52][53].In addition, PLS can conform to normality, and put forward more relaxed stochastic requirements in other predictive models [52,53].Therefore, we use PLS to carry out our main analysis, using the SmartPLS 2.0 software (www.smartpls.de)to test our research model.SmartPLS 2.0 is a very effective and widely-used analytical tool based on SEM [54].
Figure 4 below presents our research model, which is based on previous literature and the hypotheses we put forth in Table 1.This model is also constructed to examine the potential mediating role that TWQ might play between ROTS and SSCRG.
Research Method
This study adopts the two-step approach proposed by Anderson and Gerbing to test the reliability and validity of model factors and to perform hypothesis testing and model fitting [47].The number of valid samples for empirical analysis is 302, which are small and medium samples and mostly reflective the indicators we are interested in.Compared to the traditional structural equation model (SEM) analysis, our approach is more in line with partial least squares (PLS) analysis.PLS requirements for variable adaptability are more relaxed, and accuracy can also be maintained for a small sample size [48][49][50][51][52][53].In addition, PLS can conform to normality, and put forward more relaxed stochastic requirements in other predictive models [52,53].Therefore, we use PLS to carry out our main analysis, using the SmartPLS 2.0 software (www.smartpls.de)to test our research model.SmartPLS 2.0 is a very effective and widely-used analytical tool based on SEM [54].
Figure 4 below presents our research model, which is based on previous literature and the hypotheses we put forth in Table 1.This model is also constructed to examine the potential mediating role that TWQ might play between ROTS and SSCRG.To ensure reliability and validity, the multi-item measures for each construct were developed, adjusted and revised mainly by referring to the state of the existing research and maturity scale, combined with the development status and unique characteristics of the CRG, as shown in Table 2.
Most of the questionnaires developed by previous scholars were written in English.Therefore, we entrusted two university professors in charge of English-Chinese translation training at the School of English at Hebei Foreign Studies University to translate them into Chinese, and entrusted two other member teachers from Hebei Translation Society to conduct a reverse translation to ensure that the translated versions of the two language questionnaires were accurate and consistent.
Table 2. Operational definition of constructs and measurement items.
Teamwork Quality (TWQ) Source Definition: It includes communication, coordination, balance of member contributions, mutual support, effort cohesion and other internal cooperation factors in the Creative Research Group (CRG).Our criteria and basis for judging the degree of TWQ is whether the team internal cooperation of CRG has achieved communication, coordination, balance of member contributions, mutual support, effort and cohesion, along with effective information (including communication, coordination), incentive compatibility (including contribution balance, mutual support).[4]; Hoegl and Gemuenden [8]; Hoegl and Proserpio [12];
Gao and Ding
Hoegl et al. [13,17]; Hoegl and Parboteeah [14][15][16] TWQ 1. Whether there is frequent and efficient communication between team members and groups; whether tasks and work are reasonably and effectively allocated and completed; whether the team has a clear goal allocation, division and coordination of work, as well as abundant information and knowledge flow, to achieve effective information.To ensure reliability and validity, the multi-item measures for each construct were developed, adjusted and revised mainly by referring to the state of the existing research and maturity scale, combined with the development status and unique characteristics of the CRG, as shown in Table 2.
Most of the questionnaires developed by previous scholars were written in English.Therefore, we entrusted two university professors in charge of English-Chinese translation training at the School of English at Hebei Foreign Studies University to translate them into Chinese, and entrusted two other member teachers from Hebei Translation Society to conduct a reverse translation to ensure that the translated versions of the two language questionnaires were accurate and consistent.
Then, the Chinese version of the questionnaires were tested among team members at Tsinghua University and Harbin Institute of Technology who had been or were currently undertaking CRG projects.Data processing and analysis were carried out to evaluate the operability and applicability of the questionnaire and scale, and then adjustments and revisions were made to ensure its reliability and validity.The main contents of the questionnaire were related to the specific measurement of each variable, item and construct.In the end, we selected 9 items for 3 constructs.For each measurement item, the Likert scale was used with 1 being "complete inconsistency" to 7 being "complete consistency".Table 2. Operational definition of constructs and measurement items.
Teamwork Quality (TWQ) Source
Definition: It includes communication, coordination, balance of member contributions, mutual support, effort cohesion and other internal cooperation factors in the Creative Research Group (CRG).Our criteria and basis for judging the degree of TWQ is whether the team internal cooperation of CRG has achieved communication, coordination, balance of member contributions, mutual support, effort and cohesion, along with effective information (including communication, coordination), incentive compatibility (including contribution balance, mutual support).
Gao and Ding [4]; Hoegl and Gemuenden [8]; Hoegl and Proserpio [12]; Hoegl et al. [13,17]; Hoegl and Parboteeah [14][15][16] TWQ 1. Whether there is frequent and efficient communication between team members and groups; whether tasks and work are reasonably and effectively allocated and completed; whether the team has a clear goal allocation, division and coordination of work, as well as abundant information and knowledge flow, to achieve effective information.TWQ 2. Whether the team is aware of the potential and contribution value of different members, so that they can effectively play their role and receive the appropriate incentives; whether team members support, cooperate and help each other as much as possible to make sure their incentives are compatible.TWQ 3. Whether the team members have done their best to complete the project; whether they have achieved an mental and action-oriented coordination under the agreed-upon concepts and paradigms; whether the team has cohesion, a healthy cooperation atmosphere, strong cohesion in practical work; whether scientific and technological resources represented by human resources have been effectively allocated and integrated.
Rationality of Team Structure (ROTS)
Definition: It means the team structure and scale of CRG is scientific and reasonable, and the distribution of age, background, education, titles and seniority ranging from leading figures to young researchers, graduate students and other members is scientific and reasonable.The team has a good foundation in research and cooperation, good continuity and sustainability, and the potential to engage in high-level basic research and achieve cutting-edge breakthroughs.
Success and Sustainability of CRG (SSCRG)
Definition: It means the staged completion of the innovative project, the excellent evaluation of the project, the continuous operation of the team, the guarantee and possibility of the interests and benefits of team members, and the potential value and likelihood of continuous innovation, continuous growth and transition of the team in the future.It includes team performance, personal success, and the team's comprehensive impacts (scientific and technological, economic, social, and other comprehensive impacts).
Gao & Ding [4]; Hoegl & Gemuenden [8]; Hoegl & Proserpio [12]; Hoegl et al. [13,17]; Hoegl & Parboteeah [14][15][16]; Contractor [23]; Cheng & Zhang [26]; Wolf et al. [31] SSCRG 1. Whether the CRG has successfully completed the project and achieved stipulated research goals and the cultivation of talent teams.Does it have the potential value and possibility of continuous innovation, continuous growth and transition of the team?What is the likelihood and evaluability of the team's future development and transition?SSCRG 2. In addition to the success and sustainability of the team, whether the individual members of the team have obtained corresponding satisfactory growth and motivation.Compared with other projects and teams, are the team's researchers satisfied with their personal growth, new honors and projects?Are team members given opportunities for future development, the guarantee and possibility of the interests and benefits?SSCRG 3. Whether the team has made an important academic impact and other comprehensive impact; whether the results have important scientific and technological value, as well as economic, social and cultural value.Whether the comprehensive impacts have been evaluated and recognized by scientific and technological evaluation institutions and third-party assessment institutions or other organizations.
The release of the questionnaire and the development of interviews in this study benefited from the NCSTE being a third party that evaluated the annual performance of NSFC projects in 2016.Along with NCSTE staff, we conducted a performance evaluation and questionnaire survey on CRG projects, and sent questionnaires to CRG team leaders and members over a four-month period; 376 questionnaires were distributed and 302 valid questionnaires were collected.A preliminary descriptive statistical analysis of the respondents show that 212 were males, accounting for 70.20% of the total; 90 were females, accounting for 29.80%; there were 196 team leaders and senior researchers, accounting for 64.90% of the total; 99 young researchers and graduate students, accounting for 32.78%; 7 other members, accounting for 2.32%; 10 members who were undergraduates and below, accounting for 3.31% of the total; 292 students at the master's level and above, accounting for 96.69%.Lastly, 258 members (85.43%) had worked in research for three years or less, while 44 members (14.57%) had worked for more than three years.
To reiterate, this study analyzed and expanded three dimensions of the theoretical model of TWQ: 3 dimensions of TWQ, 3 dimensions of ROTS, and 3 dimensions of SSCRG, which were mainly revised by Hoegl and Gemuenden [8] and Gao and Ding [4].Table 2 summarizes the operational definitions of our measurement constructs, the items measured by the questionnaires, and the sources of our constructs and variables.
Results
According to Hair and other scholars, the factor loading in the research model we implement here should be at least above 0.5 [55].Our results show that the standardized factor loading of variables such as TWQ, ROTS and SSCRG are at least above 0.8, which meet the criteria of structural validity.Composite reliability (CR) is used to measure internal consistency.According to research by Nunnally and Bernstein, CR values greater than 0.7 meet the requirements of model fidelity [56][57][58].Table 3 shows that all CR values are greater than 0.8, which is high enough to meet the criteria of internal consistency.At the same time, internal consistency can also be tested by Cronbach's Alpha (α), which is generally greater than 0.6 [58,59].In contrast, the structural measurements in this study were based primarily on research at home and abroad, and were adopted and modified for our research purposes.The measures adopted are in line with the content validity standards, and thus, there is sufficient validity in the study.
To satisfy the test of discriminant validity, Fornell and Larcker suggested that the AVE (Average Variance Extracted) of each variable should be larger than the square of the correlation coefficient [58].Table 4 shows that the square root of AVE (the diagonal numbers in the table which are bold and italicized) is larger than the correlation coefficient, that is, AVE values are larger than the square of correlation coefficients, which is in line with the test of discriminant validity in Fornell and Larcker's research.
Research Findings
Firstly, it is clear that good TWQ and ROTS are important factors that have a positive impact on SSCRG.The validation of H1 and H2 imply that good TWQ and ROTS are positively correlated with SSCRG, and the impacts are significant.Meanwhile, the standardized path coefficients and T-Value of H1 (TWQ→SSCRG) are 0.331 and 5819 respectively, which are larger than 0.296 and 5497 of H3 (ROTS→SSCRG), indicating that compared with ROTS, TWQ plays a greater role in SSCRG.While the standardized path coefficients and T-Value of H2 (ROTS→TWQ) are 0.526 and 12.234, ROTS turns out to have the most strong and significant impact on TWQ.This suggests that more attention should be paid to team formation and structural distributions.
Secondly, it is clear that TWQ is the key mediation construct between ROTS and SSCRG.After the verification of H2 (ROTS→TWQ) and H1 (TWQ→SSCRG), we found that ROTS is positively correlated with TWQ, and also that TWQ is positively correlated with SSCRG.At the same time, through the analysis of the mediation effect, it was found that the mediation effect of TWQ is highly significant in the "ROTS-TWQ-SSCRG" path, which emphasizes the key role of TWQ in the progress of the entire development of teamwork in the CRG project.
Combined with the development progress of the CRG project, the analysis above shows that having a reasonable team structure is not equivalent to the success and sustainability of the CRG; rather, there is a positive correlation.On the basis of a reasonable structure, the development and establishment of good teamwork quality and teamwork mechanism is the key to promote the success and sustainability of the CRG.In contrast, TWQ is particularly important and key, which is not only the direct influencing factor of SSCRG, but also an important mediation construct of ROTS and SSCRG.After TWQ's transition and mediation, the positive correlation between ROTS and SSCRG is more significant.
In addition, there is a need to focus on SSCRG, which means the staged completion of the innovative project, the excellent evaluation of the project, the continuous operation of the team, the guarantee and possibility of the interests and benefits of team members, the potential value and likelihood of continuous innovation, continuous growth and transition of the team in the future.It includes three variables: team performance, personal success, and comprehensive impacts (mainly, scientific and technological, economic, and social impacts).At this stage, we mostly evaluate and measure them through experts' scores and peer reviews.In the future, we hope to build more databases and evaluation index systems, and even use big data and artificial intelligence to evaluate their success and sustainability.
Implications for Project Management, Theoretical Contributions and Limitations of the Study
In recent years, many scholars have discussed the relationship between teamwork and factors related to team sustainability [11,26,[66][67][68].However, few studies have explored this relationship from the perspective of integrated teamwork-related factors, such as the relationship between static factors (rationality of team structure, ROTS), dynamic process factors (teamwork quality, TWQ), and team success and sustainability.Our paper attempted to fill this gap.
The main contribution of this study is (i) to extent the TWQ model by including ROTS, TWQ, and SSCRG, and (ii) to present its applications in analyzing and predicting the teamwork mechanism and team sustainability.We extend the analysis based on the TWQ model to the field of research and innovation teams, providing the possibility to further deepen and extend such analyses in the future.
Our research findings show that it is clear that ROTS and TWQ both have a positive impact on SSCRG, and TWQ's impact seems to be stronger, while ROTS has the strongest positive impact on TWQ.TWQ is the mediation factor between ROTS and SSCRG.This implies that team leaders and members alike should pay greater attention to ROTS and TWQ, especially in terms of team formation and structural distributions and how these factors may affect the growth and cohesion of the entire team.Only under the premise of a reasonable structure and strong cooperation qualities can a good team cooperation mechanism be established.Only then can the CRG project be successful, achieve research objectives, achieve scientific breakthroughs and innovations, cultivate high-level talent and teams, and achieve sustainable development.In fact, many factors related to teamwork quality, team structure, team success and sustainability in this study are not limited to CRG, but they share common characteristics with many scientific research and innovation teams.In future, the applicability and promotion of such factors in similar teams would require further research and confirmation.
In general, the discovery and realization of high-level innovation require a better technology governance system and the cultivation of innovative talent teams.From the perspective of basic research projects and talent teams' ability to produce innovative research results, the NSFC needs to better play the role of a governing institution with "one body and two wings".That is to say, it should not only be the promoter of basic research and subversive innovation, but also play a role as the scientific and technological evaluator and discoverer.From the perspective of innovative talent teams, especially the CRG, these innovative talents and high-level teams are the implementers of innovation.The summary of teamwork rules and the construction and analysis of teamwork mechanisms can help them better achieve true team cooperation and successfully complete the project.It may also make it easier for managers from fund committees and other institutions to discover the rules of teamwork, the structure and the evolution mechanisms of a teamwork network, and the mutual relationships and action paths of various elements within teamwork dynamics.A well-defined teamwork mechanism design, as well as scientific and reasonable evaluation methods and systems, play an important role in promoting the development of innovation groups and their corresponding research.
Finally, the NSFC and experts in science and technology management should be reminded that the CRG is a unique frontier team undertaking basic research within China's science and technology system.Its future development and research should be combined with the reality of Chinese society in order to make greater progress in theoretical approaches and practical applications.
Although we have carried out some theoretical explorations and empirical studies on the teamwork of CRG and given some policy recommendations, our research still has many limitations, and further research is needed in the future.First, based on the existing extended model of teamwork quality and team success and sustainability relationship, we discuss teamwork in innovation groups.The theoretical framework and impact factors of our research have certain limitations; for example, we do not discuss the teamwork mechanisms of the CRG thoroughly enough.The second limitation is that the research methods are mostly based on questionnaires and semi-structured interviews, with certain subjectivity and limitations, and the sample size needs to be further expanded.This study is limited to a certain group of research projects.There is a need to examine teamwork in other types of high-level scientific research projects beyond those only in China, in order to conduct comparative studies in the future for scientists and research managers to have a reference for how to achieve scientific breakthroughs by adjusting teamwork dynmics.Future studies should combine more objective data with social network analysis methods, scientific measurements, structural equation models, peer reviews and other methods to conduct in-depth research on the cooperation mechanisms and sustainability of scientific research and innovation teams.
Figure 2 .
Figure 2. First-stage extended TWQ model for the analysis of impact factors of the Success and Sustainability of Creative Research Group (SSCRG).
Figure 2 .
Figure 2. First-stage extended TWQ model for the analysis of impact factors of the Success and Sustainability of Creative Research Group (SSCRG).
Sustainability 2019 ,
11, x FOR PEER REVIEW 8 of 17 study, we analyze the possible positive impact of ROTS and TWQ on SSCRG and the possible impact of ROTS on TWQ.
Figure 3 .
Figure 3. Second-stage extended TWQ model for the analysis of impact factors of SSCRG.
Figure 3 .
Figure 3. Second-stage extended TWQ model for the analysis of impact factors of SSCRG.
Sustainability 2019 ,
11, x FOR PEER REVIEW 9 of 17Note: ROTS = Rationality of Team Structure; SSCRG = Success and Sustainability of Creative Research Group; TWQ = Teamwork Quality.
Figure 4 .
Figure 4. Schema of research model and hypotheses.
Figure 4 .
Figure 4. Schema of research model and hypotheses.
ROTS 2 .
Whether the age, background, education, professional titles, and seniority of the team members are reasonably distributed and fairly matched.ROTS 3. Whether the members represented by leading figures have a strong foundation in research and cooperation, good continuity and sustainability, high-level research cooperation ability and the potential to execute cutting-edge research and realize specific research goal.
Table 3 .
Reliability and convergent validity.
Table 4 .
Correlation coefficients matrix and square roots of AVE.
Note: ROTS = Rationality of Team Structure; SOCRG = Success and Sustainability of Creative Research Group; TWQ = Teamwork Quality.The diagonal bold and italic values are square roots of AVE, and the other values are correlation coefficients.It can be seen that the square roots of AVE are larger than the correlation coefficients, that is, AVE is larger than the square of correlation coefficients, so the discriminant validity exists.
|
2019-03-04T13:12:20.735Z
|
2019-02-24T00:00:00.000
|
{
"year": 2019,
"sha1": "99603b6165bac5852d488a83f0e3ad798b367feb",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2071-1050/11/4/1195/pdf?version=1550996437",
"oa_status": "GOLD",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "99603b6165bac5852d488a83f0e3ad798b367feb",
"s2fieldsofstudy": [
"Business"
],
"extfieldsofstudy": [
"Sociology"
]
}
|
137376655
|
pes2o/s2orc
|
v3-fos-license
|
On the structural-optical correlations in radiation-modified chalcogenide glasses
In this work, we report our recent results on the y-irradiation-induced structural transformations in the Ge-Sb-S glasses as observed from the structural studies using high-energy synchrotron x-ray diffraction and extended x-ray absorption fine structure spectroscopy in comparison with the optical measurements using VIS/IR spectroscopy techniques. The structural-optical correlations in the radiation-induced effects are established. The structural changes upon irradiation are explained in the frames of the concept of coordination topological defects formation.
Introduction
Chalcogenide glasses are known as perspective optoelectronic materials with unique physical and chemical properties, induced phenomena, etc. [1]. Their attractive application in optoelectronics is tightly related to the externally induced structural defects.
The concept of coordination topological defects (CTDs) was applied to interpret both photo-and radiation-induced phenomena in amorphous chalcogenides [2]. Infrared (IR) Fast Fourier Transform spectroscopy [3] is a sensitive method for CTDs' detection due to direct relation between intensity of vibrational mode and concentration of oscillating centres/chemical bonds. Recently, it has been shown [4] that Raman scattering in depolarized configuration is also an informative tool for CTD identification. However, application of vibrational methods is limited because they provide information on the statistics of chemical bonds only and help to construct possible topological CTD-related schemes, but no information on the local atomic structure in the vicinity of CTD can be obtained. Besides, the sensitivity of IR and Raman spectroscopy is rather low if oscillating complexes are present in the glass matrix in a small amount, while the role of these complexes in the externally induced defect formation processes could be significant.
The goal of the present work is to apply direct structural techniques such as high-energy synchrotron xray diffraction (XRD) and extended x-ray absorption fine structure (EXAFS) spectroscopy in combination with optical VIS/IR measurements in the visible (VIS) and infrared (IR) spectral ranges for the study of radiation-modified chalcogenide glasses on the example of Ge-Sb-S ternary system.
Experimental
The bulk Ge x Sb 40-x S 60 samples of chalcogenide glasses with x = 5, 25, 27 and 35 were prepared by a standard melt-quenching method. Radiation treatment of the samples with accumulated dose of 2.41 and 7.72 MGy was performed at the normal conditions of stationary radiation field, created in a closed cylindrical cavity by a number of concentrically established 60 Co radioisotope capsules as shown schematically in Fig. 1. No special measures were taken to prevent uncontrolled thermal annealing of the samples, but maximum temperature in the irradiating camera did not exceed 320-330 K during prolonged -irradiation (more than 30 days), providing absorbed dose power P < 5 Gy/s. Principal advantages of irradiation as a structure modification factor for amorphous chalcogenides among other types of ionizing irradiation are discussed elsewhere [5]. High-energy XRD experiments were performed at the synchrotron experimental station BW5 at HASYLAB, DESY in Hamburg (Germany). EXAFS measurements at Ge K-edge were carried out at the synchrotron beam line X1 (HASYLAB) in transmission mode using Si (111) double-crystal monochromator. The measuring time was k-weighted during collection of the signal. The experimental data were treated using VIPER program [6] and FEFF8.4 code [7]. VIS optical transmission measurements were performed with "Specord M40" spectrophotometer. Far IR optical reflectance measurements were conducted with IR spectrometer "KSDI-82". IR optical reflectance measurements [9] support the changes observed on g(r)'s, but only for the GeS correlations (Fig. 3). Indeed, only the strong peak at 370 cm -1 corresponding to the bond stretching modes in GeS 4 tetrahedra (vibrations of GeS bonds) [10] is modified upon irradiation. The peak at 290 cm -1 (well resolved for the glass with 25 at. % Ge) corresponds to the bond stretching modes in SbS 3 pyramids (vibrations of SbS bonds) [11]. A shoulder at 330 cm -1 (better seen at higher Ge content) is assigned to the bond stretching modes in SbS 3 pyramids [12]. The SbS correlations are found to be stable to radiation impact. The GeSb vibrations are not resolved in the IR spectral region studied probably due to a small amount of these groups in the glass matrix of GeSbS ternaries as it was predicted earlier by Feltz [13] and observed experimentally in the XRD [8] and EXAFS [14] studies. In the frames of the concept of coordination topological defects [2], the radiation-induced defective structural transformations detected in the XRD and IR studies of the Ge 25 Sb 15 S 60 and Ge 35 Sb 5 S 60 glasses can be explained by the formation of CTDs as a result of chemical bond distortion and switching upon irradiation. The changes in GeS correlations can be an indicative of the distortion of GeS bonds with switching and creation of negatively charged under-coordinated Ge 3 and positively charged overcoordinated S 3 + defects. Schematically this mechanism is illustrated in Fig. 4 (left) [8]. The radiationinduced non-defective structural transformations detected only from XRD study are connected with formation of GeSb bonds with normal atomic coordination (Ge 4 0 , Sb 3 0 , S 2 0 ) in their vicinity with intermediate configuration accompanied by appearance of unstable Ge 3 and S 3 + CTDs as shown in Fig. 4 (right) [8]. The nature of Sb atoms (property of Sb to annihilate a wrong coordination defect in its vicinity) plays a key role for occurrence of the non-defective mechanism. It is understandable that the main radiation-induced structural transformations (defective and nondefective) take place in GeS sub-system, whereas no changes in SbS correlations demonstrate radiation stability of SbS sub-system.
Results and discussion
The EXAFS (k)k 3 Ge K-edge spectra and their Fourier transforms for Ge 27 Sb 13 S 60 glass in the unmodified and -radiation modified states (Fig. 5) give an evidence for the conclusion on the changes in the GeS sub-system. Indeed, the first nearest neighbor distance r = 2.24 Å detected in the first coordination shell is attributed to the GeS bonding [15]. Upon irradiation, the coordination number of Ge atoms N GeS decreases by 1.3 % and Debye-Waller factor 2 GeS or MSRD increases by 4.4 % that support the appearance of CTDs or atomic pairs with wrong coordination and static disorder. The radiation stability or non-sensitivity of SbS sub-system is evidenced also from optical transmission measurements as illustrated in Fig. 6. It is clearly seen that the red shift of optical transmittance (or radiation-induced darkening effect) is detected for the Ge-rich composition with 27 at.% Ge and no effect is observed for the Sb-rich alloy with 35 at.% Sb.
Conclusion
We report our recent results on the -irradiation-induced structural transformations in Ge-Sb-S glasses as observed from the structural studies using high-energy synchrotron XRD and EXAFS in comparison with the optical investigations using VIS/IR spectroscopy techniques. The structural-optical correlations in the radiation-induced effects are established. In contrast to the IR optical measurements, XRD study reveals the role of heteronuclear metal-metal chemical bonding (Ge-Sb), which cannot be negligible in the radiation-induced structural transformations in the glass system investigated. It is concluded that both defective and non-defective mechanisms of the radiation-induced structural changes might be responsible for the effects observed in the XRD and IR experiments. The radiation-induced structural changes can be plausibly explained with the coordination topological defects formation concept.
|
2019-04-28T13:13:06.375Z
|
2011-04-01T00:00:00.000
|
{
"year": 2011,
"sha1": "edd99a4ec6ac08e10418b0935fc0de412aa2cc97",
"oa_license": null,
"oa_url": "https://doi.org/10.1088/1742-6596/289/1/012007",
"oa_status": "GOLD",
"pdf_src": "IOP",
"pdf_hash": "dccbe36570cb097c616465b5dd2ccfa8879522f1",
"s2fieldsofstudy": [
"Physics",
"Materials Science"
],
"extfieldsofstudy": [
"Materials Science"
]
}
|
211064130
|
pes2o/s2orc
|
v3-fos-license
|
Visible Light-Driven p-Type Semiconductor Gas Sensors Based on CaFe2O4 Nanoparticles
In this work, we present conductometric gas sensors based on p-type calcium iron oxide (CaFe2O4) nanoparticles. CaFe2O4 is a metal oxide (MOx) with a bandgap around 1.9 eV making it a suitable candidate for visible light-activated gas sensors. Our gas sensors were tested under a reducing gas (i.e., ethanol) by illuminating them with different light-emitting diode (LED) wavelengths (i.e., 465–640 nm). Regardless of their inferior response compared to the thermally activated counterparts, the developed sensors have shown their ability to detect ethanol down to 100 ppm in a reversible way and solely with the energy provided by an LED. The highest response was reached using a blue LED (465 nm) activation. Despite some responses found even in dark conditions, it was demonstrated that upon illumination the recovery after the ethanol exposure was improved, showing that the energy provided by the LEDs is sufficient to activate the desorption process between the ethanol and the CaFe2O4 surface.
Introduction
Metal oxide semiconductors have shown the best characteristics in term of sensitivity, selectivity, and stability in gas sensor technology. The development of new materials based on different methods, techniques and working principles has been carried out to achieve the best performance. However, the operating temperatures of metal oxide gas sensors are usually above 150 • C to activate the absorption and desorption processes between the targeted gases and the surfaces of the materials [1][2][3][4][5][6][7][8]. The high operating temperature is one of the main drawbacks of metal oxide-based gas sensors because it ultimately results in high power consumption and undesirable long-term drift problems caused by sintering effects in the metal oxide grain boundaries, yielding poor selectivity and stability [9,10]. Another disadvantage of the metal oxide-based gas sensors with the high operating temperature is
Sensor Preparation
Gold-interdigitated electrodes (Au-IDE) on glass (MicruX Technology, Asturias, Spain) were used as an electronic platform for measuring the electrical characteristics of the CaFe 2 O 4 . The IDE size is (10 × 6 × 0.75 mm), with 90 pairs of electrodes having a pitch of 10 µm and a line width 10 µm. CaFe 2 O 4 nanoparticles were deposited on the surface of the Au-IDEs by drop casting, followed by a spin-coating process. After setting the substrate on the sample holder of the spin coater, 5 µL of CaFe 2 O 4 nanoparticle 10 mg/l suspensions in ethylene glycol were deposited onto Au-IDE by a single layer spin coater and spin-coated at 2000 rpm for 40 s in the air and dried on the hot plate at 90 • C for a few minutes in order to evaporate the solvent. To attain the desired thickness of CaFe 2 O 4 nanoparticles film, the above procedure was repeated 3 times. Afterwards, an annealing process at 450 • C for 1 h with a ramping level 5 • C/min was applied to fix the material onto the substrate and achieve good electrical contact with the Au-IDEs.
Sensor Measurement
Gas-sensing experiments were conducted in a customized chamber of 200 mL in volume. The gas flow was maintained stably at 200 mL/min during all the measurements. Reference gaseous atmospheres were provided by independent mass flow controllers blending synthetic air (SA) and ethanol (100 ppm in SA). To investigate CaFe 2 O 4 optoelectronic properties by activating visible light, resistance measurements were conducted under synthetic air flow with different LED wavelengths (i.e., blue (465 nm), green (520 nm), yellow (590 nm), and red (640 nm)) and in dark condition (without illumination). To determine the sensitivity of the CaFe 2 O 4 sensors towards reducing gases, different concentrations of ethanol vapors from the low to the high concentration (i.e., 10,20,30,50, and 100 ppm) were then applied to the chamber under LED irradiation. The response was defined as [(R g − R a )/R a ] × 100%, where R a and R g are the electrical resistances of the sensor in the air and when exposed with ethanol, respectively. The response and recovery times were defined as the times needed by a sensor to achieve 90% of the total resistance change during the adsorption and desorption process, respectively. All experiments were performed at room temperature.
Sensor Characterization Results
The sample as-prepared powder was tested using XRD as shown in Figure 1a. The mixture of various compounds (e.g., CaCO 3 (ICDD 04-007-4989), Fe 2 O 3 (ICDD 00-002-1047), γ-Fe 2 O 3 (ICDD 00-004-0755), and CaFe 2 O 4 (ICDD 04-007-4989)) was formed during the auto-combustion reaction. Nevertheless, the admixture of different compounds crystallizes to pure CaFe 2 O 4 compound (Figure 1b A scanning electron microscope (SEM) image of an as-prepared powder cross-section of gas sensor pellet annealed at 850 °C for 3 h shown in Figure 2a. The as-prepared sample powders are composed of amorphous-like anisotropically shaped and closely packed grains. Figure 2b shows a surface sensor on the Au-IDE after being prepared by the spin-coating process and annealed at 450 °C for 1 h. The porous structures of interconnected grains were observed, while the grains keep their anisotropic shape. The size of individual grains of smaller dimensions varies from 70 to 300 nm, while the length of anisotropic nanoparticles is up to 650 nm. Particles are very well interconnected and fused together, at the same time maintaining open structures for gas diffusion. Gas-accessible microstructures are preferred for the high gas response. Due to the particle size and form, the coating result was non-uniform on the IDEs. That was not the best result to make a uniform layer by a spin-coating process. However, knowing the response to the ethanol vapors will make a good opportunity to introduce this material as a potential candidate for p-type semiconductor which has a suitable bandgap for visible light-driven gas sensors at room temperature. Another possible method to make the uniform layer on the IDE is by the screen-printing process which is a widespread method in the industry of metal oxide gas sensors. Figure 3a shows the optical adsorption UV-vis diffuse reflectance spectra (DRS) of CaFe2O4 nanoparticles and Tauc's plot approach to determine the bandgap. It has confirmed the adsorption spectra of CaFe2O4 in the visible light range (400-700 nm). The absorption peak at 425-455 nm (dashed green line/pattern) is due to the maximum adsorption spectra of CaFe2O4 nanoparticles. The dashed red line is the linear fit absorption spectra which are leading to the optical bandgap value. Its optical bandgap was determined to be ~1.9 eV according to the energy dependence relation of A scanning electron microscope (SEM) image of an as-prepared powder cross-section of gas sensor pellet annealed at 850 • C for 3 h shown in Figure 2a. The as-prepared sample powders are composed of amorphous-like anisotropically shaped and closely packed grains. Figure 2b shows a surface sensor on the Au-IDE after being prepared by the spin-coating process and annealed at 450 • C for 1 h. The porous structures of interconnected grains were observed, while the grains keep their anisotropic shape. The size of individual grains of smaller dimensions varies from 70 to 300 nm, while the length of anisotropic nanoparticles is up to 650 nm. Particles are very well interconnected and fused together, at the same time maintaining open structures for gas diffusion. Gas-accessible microstructures are preferred for the high gas response. Due to the particle size and form, the coating result was non-uniform on the IDEs. That was not the best result to make a uniform layer by a spin-coating process. However, knowing the response to the ethanol vapors will make a good opportunity to introduce this material as a potential candidate for p-type semiconductor which has a suitable bandgap for visible light-driven gas sensors at room temperature. Another possible method to make the uniform layer on the IDE is by the screen-printing process which is a widespread method in the industry of metal oxide gas sensors. A scanning electron microscope (SEM) image of an as-prepared powder cross-section of gas sensor pellet annealed at 850 °C for 3 h shown in Figure 2a. The as-prepared sample powders are composed of amorphous-like anisotropically shaped and closely packed grains. Figure 2b shows a surface sensor on the Au-IDE after being prepared by the spin-coating process and annealed at 450 °C for 1 h. The porous structures of interconnected grains were observed, while the grains keep their anisotropic shape. The size of individual grains of smaller dimensions varies from 70 to 300 nm, while the length of anisotropic nanoparticles is up to 650 nm. Particles are very well interconnected and fused together, at the same time maintaining open structures for gas diffusion. Gas-accessible microstructures are preferred for the high gas response. Due to the particle size and form, the coating result was non-uniform on the IDEs. That was not the best result to make a uniform layer by a spin-coating process. However, knowing the response to the ethanol vapors will make a good opportunity to introduce this material as a potential candidate for p-type semiconductor which has a suitable bandgap for visible light-driven gas sensors at room temperature. Another possible method to make the uniform layer on the IDE is by the screen-printing process which is a widespread method in the industry of metal oxide gas sensors. Figure 3a shows the optical adsorption UV-vis diffuse reflectance spectra (DRS) of CaFe2O4 nanoparticles and Tauc's plot approach to determine the bandgap. It has confirmed the adsorption spectra of CaFe2O4 in the visible light range (400-700 nm). The absorption peak at 425-455 nm (dashed green line/pattern) is due to the maximum adsorption spectra of CaFe2O4 nanoparticles. The dashed red line is the linear fit absorption spectra which are leading to the optical bandgap value. Its optical bandgap was determined to be ~1.9 eV according to the energy dependence relation of The dashed red line is the linear fit absorption spectra which are leading to the optical bandgap value. Its optical bandgap was determined to be~1.9 eV according to the energy dependence relation of (αhν) 2 = A hν − E g , where α and E g are the absorption coefficient and the bandgap of CaFe 2 O 4 , respectively. In addition, it can be seen that the material shows efficient visible light absorption spectra and the bandgap of CaFe 2 O 4 . Figure 3b shows the optoelectronic properties of CaFe 2 O 4 under LED illumination with different wavelengths and intensities of visible light without gases. The responses indicated that the visible light is suitable for this material because its energy is equal or larger than the bandgap of CaFe 2 O 4 . It confirms that the response and recovery times depend on the energy and intensity of LEDs. When the light is ON (photo-activated), electron-hole (e-h) pairs are generated in CaFe 2 O 4 and will interact with an oxygen molecule and pre-chemisorb oxygen ion in the surface, thus facilitating their chemisorption and increasing the majority charge in CaFe 2 O 4 . This reaction will form a hole-accumulation layer, leading to decrease in electrical resistance. On the contrary, when the light is OFF the recombination process leads to an increase of the resistance to the initial value. According to the results, visible light activation works properly in this material. Moreover, the density of majority charge carriers is not only related to the intensity of visible light, but also to the absorption at this particular energy. In thin nanoparticle films, higher absorption leads to a larger generation rate of e-h pairs, and hence to a stronger impact on desorption.
Sensors 2020, 20, x 5 of 12 ( ℎ ) = ℎ − , where α and Eg are the absorption coefficient and the bandgap of CaFe2O4, respectively. In addition, it can be seen that the material shows efficient visible light absorption spectra and the bandgap of CaFe2O4. Figure 3b shows the optoelectronic properties of CaFe2O4 under LED illumination with different wavelengths and intensities of visible light without gases. The responses indicated that the visible light is suitable for this material because its energy is equal or larger than the bandgap of CaFe2O4. It confirms that the response and recovery times depend on the energy and intensity of LEDs. When the light is ON (photo-activated), electron-hole (e-h) pairs are generated in CaFe2O4 and will interact with an oxygen molecule and pre-chemisorb oxygen ion in the surface, thus facilitating their chemisorption and increasing the majority charge in CaFe2O4. This reaction will form a hole-accumulation layer, leading to decrease in electrical resistance. On the contrary, when the light is OFF the recombination process leads to an increase of the resistance to the initial value. According to the results, visible light activation works properly in this material. Moreover, the density of majority charge carriers is not only related to the intensity of visible light, but also to the absorption at this particular energy. In thin nanoparticle films, higher absorption leads to a larger generation rate of e-h pairs, and hence to a stronger impact on desorption. To investigate the sensitivity of the CaFe2O4 sensors towards reducing gases (i.e., ethanol), different concentrations of ethanol vapors (i.e., from 10 to 100 ppm) were then applied to the chamber under LED illumination. The first phenomenon to be noticed was that the resistance increased in the presence of ethanol (reducing gas), confirming that CaFe2O4 is a p-type material. For comparison, the experiments were also performed by introducing another reducing gas (i.e. NH3), as shown in Figure S1 and the oxidizing gas (i.e., NO2) with different concentrations ( Figure S2). Figure 4a-e shows the dynamic response to different concentrations of ethanol under blue, green, yellow and red LED illumination and also in dark condition (without illumination). The response comparison toward different conditions is shown in Figure 4f. The blue LED had better sensitivity than the others because of its energy. The higher energy and intensity are illuminated as the material increases the energetic state and density of charge carrier on the surface, which influences the sensitivity while being exposed to the target gas. The maximum response corresponding to adsorption spectra is 3.6% at 100 ppm of ethanol for blue LED. In this case, the response and recovery times were ~18 min and ~41 min, respectively. The results obtained provide evidence that CaFe2O4 is a good candidate for visible light-driven gas sensor because of its suitable bandgap (energy of visible light spectra is 1.9-2.7 eV). The sensor responses from other LEDs are too slow compared to that from blue LED, which can be due to their insufficient energy to detect gas.
Some responses were, however, also found in dark conditions. The sensitivity in dark conditions (Figure 4f) indicates that p-type material has the ability to chemisorb the higher To investigate the sensitivity of the CaFe 2 O 4 sensors towards reducing gases (i.e., ethanol), different concentrations of ethanol vapors (i.e., from 10 to 100 ppm) were then applied to the chamber under LED illumination. The first phenomenon to be noticed was that the resistance increased in the presence of ethanol (reducing gas), confirming that CaFe 2 O 4 is a p-type material. For comparison, the experiments were also performed by introducing another reducing gas (i.e., NH 3 ), as shown in Figure S1 and the oxidizing gas (i.e., NO 2 ) with different concentrations ( Figure S2). Figure 4a-e shows the dynamic response to different concentrations of ethanol under blue, green, yellow and red LED illumination and also in dark condition (without illumination). The response comparison toward different conditions is shown in Figure 4f. The blue LED had better sensitivity than the others because of its energy. The higher energy and intensity are illuminated as the material increases the energetic state and density of charge carrier on the surface, which influences the sensitivity while being exposed to the target gas. The maximum response corresponding to adsorption spectra is 3.6% at 100 ppm of ethanol for blue LED. In this case, the response and recovery times were~18 min and~41 min, Sensors 2020, 20, 850 6 of 12 respectively. The results obtained provide evidence that CaFe 2 O 4 is a good candidate for visible light-driven gas sensor because of its suitable bandgap (energy of visible light spectra is 1.9-2.7 eV). The sensor responses from other LEDs are too slow compared to that from blue LED, which can be due to their insufficient energy to detect gas. Some responses were, however, also found in dark conditions. The sensitivity in dark conditions (Figure 4f) indicates that p-type material has the ability to chemisorb the higher concentrations of oxygen which react with gases since the formation of a hole-accumulation space charge layer is not limited by concentrations of free charge carriers [50] despite no illumination of light. The light irradiation has not only provided visible light activation on CaFe 2 O 4 , but also contributed to the desorption process when the ethanol was removed from the chamber. The energy provided by LEDs is sufficient to activate the desorption process between ethanol and the surface of CaFe 2 O 4 . In a dark condition, there was no external energy to break the bonding of target gases on the surface sensing. Thus, the sensor signal was not able to be well recovered. The maximum recovery ability in the dark condition is 35% at 100 ppm of ethanol and the average recovery ability is less than 25% for all concentrations of ethanol.
Sensors 2020, 20, x 6 of 12 concentrations of oxygen which react with gases since the formation of a hole-accumulation space charge layer is not limited by concentrations of free charge carriers [50] despite no illumination of light. The light irradiation has not only provided visible light activation on CaFe2O4, but also contributed to the desorption process when the ethanol was removed from the chamber. The energy provided by LEDs is sufficient to activate the desorption process between ethanol and the surface of CaFe2O4. In a dark condition, there was no external energy to break the bonding of target gases on the surface sensing. Thus, the sensor signal was not able to be well recovered. The maximum recovery ability in the dark condition is 35% at 100 ppm of ethanol and the average recovery ability is less than 25% for all concentrations of ethanol.
Sensing Mechanism
The working principle of gas sensors based on metal oxide depends on chemisorbed oxygen molecules on the surface (i.e., adsorption and desorption), which ionize into species such as O , O and O by taking electrons near the surface of the metal oxides. Generally, ionosorption species of O , O and O are known to be dominant at <150 °C, between 150 and 400 °C, and at >400 °C, respectively [57]. In case of a p-type metal oxide formed by an aggregate of nanoparticles the conduction mechanism is governed by the grain boundaries. However, unlike in the case of n-type metal oxide materials where the outer shell of the nanoparticles is "insulating" because of the ionosorption of oxygen species, in p-type metal oxides the outer shell develops a hole accumulation layer (i.e., "conducting" layer). Figure 5a shows the condition of CaFe2O4 when it is exposed to air in the dark, in which the adsorbed oxygen molecules trap electrons from the valence band of CaFe2O4 and form pre-chemisorbed oxygen ion (O ) on the surface at room temperature [57,58]. Pre-chemisorbed O on the surface results in the presence of a high-conductivity hole-accumulation region in the surface layer of CaFe2O4. Consequently, the energy bands bend upward near the surface of CaFe2O4 (Figure 5d) in comparison with the flat band situation before any surface reaction (Figure 5c) [59]. In the dark condition, the pre-chemisorbed oxygen ion is thermally stable and difficult to remove from the surface of CaFe2O4 at room temperature due to the large absorption
Sensing Mechanism
The working principle of gas sensors based on metal oxide depends on chemisorbed oxygen molecules on the surface (i.e., adsorption and desorption), which ionize into species such as O − 2 , O − and O 2− by taking electrons near the surface of the metal oxides. Generally, ionosorption species of O − 2 , O − and O 2− are known to be dominant at <150 • C, between 150 and 400 • C, and at >400 • C, respectively [57]. In case of a p-type metal oxide formed by an aggregate of nanoparticles the conduction mechanism is governed by the grain boundaries. However, unlike in the case of n-type metal oxide materials where the outer shell of the nanoparticles is "insulating" because of the ionosorption of oxygen species, in p-type metal oxides the outer shell develops a hole accumulation layer (i.e., "conducting" layer). Figure 5a shows the condition of CaFe 2 O 4 when it is exposed to air in the dark, in which the adsorbed oxygen molecules trap electrons from the valence band of CaFe 2 O 4 and form pre-chemisorbed oxygen ion O − 2 on the surface at room temperature [57,58]. Pre-chemisorbed O − 2 on the surface results in the presence of a high-conductivity hole-accumulation region in the surface layer of CaFe 2 O 4 . Consequently, the energy bands bend upward near the surface of CaFe 2 O 4 (Figure 5d) in comparison with the flat band situation before any surface reaction (Figure 5c) [59]. In the dark condition, the pre-chemisorbed oxygen ion is thermally stable and difficult to remove from the surface of CaFe 2 O 4 at room temperature due to the large absorption energy [60]. The kinetic reaction can be explained as follows [6]: When the light illuminates the materials (Figure 5b), electrons are excited from the valence band to conduction band and electron-hole pairs are generated. The holes react with the pre-chemisorbed O − 2(ads) to form oxygen molecules which will be desorbed from the surface of CaFe 2 O 4 ( h + + O − 2(ads) ↔ O 2(g) ). At the same time, new oxygen molecules will be adsorbed and capture the photo-electrons to form photo-induced oxygen ions: ). The net result of these adsorbtion and desorbtion processes of oxygen molecues is that the photoinduced holes acumulate into the surface increasing the width of the hole-accumulation layer (Figure 5e). Consequently, the resistance of CaFe 2 O 4 decreases in this reaction. The reaction can be explained as in the following equation: Sensors 2020, 20, x 7 of 12 When the light illuminates the materials (Figure 5b), electrons are excited from the valence band to conduction band and electron-hole pairs are generated. The holes react with the pre-chemisorbed O ( ) to form oxygen molecules which will be desorbed from the surface of CaFe2O4 Figure 6 shows a scheme of the sensing mechanism when the material is exposed to the target gas (ethanol vapors) under illumination together with the dynamic response of the sensor (green line) and the corresponding energy band diagram for each situation. When the sensor is exposed to the ethanol vapors, the ethanol molecules are absorbed on the surface and react with photo-induced Figure 6 shows a scheme of the sensing mechanism when the material is exposed to the target gas (ethanol vapors) under illumination together with the dynamic response of the sensor (green line) and the corresponding energy band diagram for each situation. When the sensor is exposed to the ethanol vapors, the ethanol molecules are absorbed on the surface and react with photo-induced oxygen ions to form water vapor (H 2 O) and CO 2 consuming photo-induced oxygen ions from the surface by releasing electrons (Figure 6b). The reaction can be described as follows: oxygen ions to form water vapor (H2O) and CO2 consuming photo-induced oxygen ions from the surface by releasing electrons (Figure 6b). The reaction can be described as follows: The released electrons will return to the valence band and cause a decreasing concentration of oxygen ions in the surface, resulting in electron-hole compensation and eventually narrowing the hole-accumulation layer. This narrowing process results in an increased resistance when a reducing gas is introduced [50], as shown in Figure 6e. When ethanol vapors are removed from the chamber, the remaining ethanol molecules adsorbed in the surface of the material will eventually desorb through reactions (5) and (6) and be replaced again by adsorbed oxygen molecules returning to the original situation (increase the concentration of hole and the width of the HAL and also resulting in a decrease the electrical resistance of CaFe2O4 (Figure 6c,f). However, the energy needed to either desorb ethanol from the surface or induce reactions (5) and (6) could be higher than that provided by the incident photons, and therefore some of the adsorbed ethanol molecules (or acetaldehyde from reaction (5)) will remain attached to the surface resulting in a state slightly different from the original with a different resistance. Figure 6. The dynamic response (the green line ) toward reducing gas (i.e., ethanol vapors) under visible light irradiation that corresponded to: (a) photo-activation before target gas exposure has been introduced, (b) target gas interacts with photo-induced oxygen ion and attaches on the surface (photo-adsorption process), (c) photo-desorption process when target gas is detached from the surface, (d) energy band diagram photo-activation before interaction with target gas, (e) photo-adsorption process causes a decreasing the width of HAL and increasing of the electrical resistance, and (f) photo-desorption process causes an increasing the width of HAL and resulting in a decrease the electrical resistance of CaFe2O4. Figure 6. The dynamic response (the green line) toward reducing gas (i.e., ethanol vapors) under visible light irradiation that corresponded to: (a) photo-activation before target gas exposure has been introduced, (b) target gas interacts with photo-induced oxygen ion and attaches on the surface (photo-adsorption process), (c) photo-desorption process when target gas is detached from the surface, (d) energy band diagram photo-activation before interaction with target gas, (e) photo-adsorption process causes a decreasing the width of HAL and increasing of the electrical resistance, and (f) photo-desorption process causes an increasing the width of HAL and resulting in a decrease the electrical resistance of CaFe 2 O 4 .
The released electrons will return to the valence band and cause a decreasing concentration of oxygen ions in the surface, resulting in electron-hole compensation and eventually narrowing the hole-accumulation layer. This narrowing process results in an increased resistance when a reducing gas is introduced [50], as shown in Figure 6e. When ethanol vapors are removed from the chamber, the remaining ethanol molecules adsorbed in the surface of the material will eventually desorb through reactions (5) and (6) and be replaced again by adsorbed oxygen molecules returning to the original Sensors 2020, 20, 850 9 of 12 situation (increase the concentration of hole and the width of the HAL and also resulting in a decrease the electrical resistance of CaFe 2 O 4 (Figure 6c,f). However, the energy needed to either desorb ethanol from the surface or induce reactions (5) and (6) could be higher than that provided by the incident photons, and therefore some of the adsorbed ethanol molecules (or acetaldehyde from reaction (5)) will remain attached to the surface resulting in a state slightly different from the original with a different resistance.
Conclusions
CaFe 2 O 4 nanoparticles have been synthesized by a sol-gel auto-combustion method resulting in unconventional metal oxides with bandgap of around 1.9 eV, which is suitable for visible light spectra. Light-activated room-temperature gas sensors based on this material has been tested and validated. The maximum responses toward ethanol vapor and recovery time were 3.6% at 100 ppm,~18 min and~41 min, respectively. The maximum response corresponds to the maximum absorption spectra (425-455 nm) of CaFe 2 O 4 based on results of optical absorption UV-vis diffuse reflectance spectra. The energy provided by the LEDs is sufficient to activate the desorption process between the ethanol and the surface of CaFe 2 O 4 . In addition, it has been confirmed that visible light activation contributed to breaking the bonding of target gases from the surface sensing.
|
2020-02-09T14:07:05.821Z
|
2020-02-01T00:00:00.000
|
{
"year": 2020,
"sha1": "b0973119534e82bd4ba0d2ecd77987c239800932",
"oa_license": "CCBY",
"oa_url": "https://res.mdpi.com/d_attachment/sensors/sensors-20-00850/article_deploy/sensors-20-00850-v2.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "1f21d2ba2b6aaebfdc1b9f67db7c45d7c51dbe40",
"s2fieldsofstudy": [
"Materials Science",
"Chemistry",
"Engineering"
],
"extfieldsofstudy": [
"Computer Science",
"Materials Science",
"Medicine"
]
}
|
236716404
|
pes2o/s2orc
|
v3-fos-license
|
Readiness for action in emergency circumstances in professional activities of penal system employees
. Correctional facilities are created to implement penalties for persons who have committed criminal offences. Consequently, persons held in them are often socially dangerous, criminally infected and inclined to deviant behavior. At the same time, the isolation, restriction of discretion and movement, the need to stay for a long time in a certain group of people are triggering factors and cause such conditions as boredom, irritation, apathy, depression, emotional instability. These circumstances do not exclude the occurrence of emergencies caused by a violation of standards and rules of conduct by persons sentenced to imprisonment, as evidenced by the facts recorded in penitentiary practice. Working in such conditions imposes special requirements to employees. In respect thereof, particular attention is paid to the need for penal system employees to be ready for emergency circumstances. This type of readiness is complex and includes motivationally interdependent will, cognitive and activity components. The formation of the given components depends on a number of external and internal factors. While preparing staff for professional tasks in emergency circumstances, it is rather important to pay attention not only to personal, but also to the collective readiness of staff units. The staff readiness to emergencies contributes to the effective service tasks execution, increases the sense of self-confidence in staff and their authority among colleagues, and affects the nature of professional communication.
Introduction
correctional system (hereinafter referred to as the PCS), represented by operational, regime and educational services, are obliged to prevent unlawful actions by citizens in places of deprivation of liberty. Apart from the usual tasks established by law, due to the international implications, increased terrorist threats, attempts to discredit the Federal Penal Service of the Russian Federation in the society and on television, optimization of the number of employees of the penal system and other phenomena, in the functioning process of the PCS facilities may arise situations that require special attention and some provisional measures. Therefore, the increased attention should be paid to the tactics of the territorial and correctional personnel in emergency circumstances and to the methods of its improvement [1]. According to the Federal Constitutional Law of May 30, 2001, No. 3-FKZ "On state of emergency" [2], emergency circumstances refer to circumstances that constitute an immediate threat to the life and security of citizens or to the constitutional order of the Russian Federation and whose elimination is impossible without the application of emergency measures. Criminal emergencies include riots, hostage-taking, terrorist acts and other social processes and phenomena requiring special organizational, legal and other measures from the federal executive bodies. [3].
The work in non-standard conditions that pose a threat to the life and health of employees implies special requirements for their professional training and psychological qualities. Therefore, the concept of readiness for emergency circumstances and its development among PCS staff becomes relevant.
Results and discussion
A number of penitentiary researchers believe that the definition of "emergency circumstances" combines two terms: "emergency situations" and "emergency incidents," where the first term is understood as an unfavourable situation caused by circumstances of a natural and man-made character, and the second as short-term and single events, usually of a criminal nature, disrupting the correct work of penal facilities [4].
Turning to the domestic penal system, it should be mentioned that the most common emergency circumstances are the following: escapes of convicted, suspected and accused persons from custody or convoying; riots and group disobedience, expressed through the active hooliganism on the territory of the correctional facility; hostage-taking by convicts; armed attacks against the protected site, service facilities, employees of the institution; another natural, man-made and biological and social emergencies.
According to media reports, in 2019 alone, 13 group disobediences were committed by convicts in correctional facilities (mainly related to a hunger strike or personal injury (selfinjury)) [5].
Official legislation pays special attention to this problem. Mainly concerning not only the suppression of the similar situations already arisen, but also the prevention and avoidance of the above-mentioned emergency circumstances.
Among them are factors that can lead to the occurrence of emergency incidents and circumstances in the facilities: external (natural disaster), as well as internal (riots, group disobedience, escapes of convicts from custody, etc.) [6]. Special risks are often created by purposely planned actions aimed at destabilizing the activities of correctional facilities. That was demonstrated at the Main Directorate of the Federal Penal Service of Russia in the Republic of Bashkortostan in 2015, and at the Main Directorate of the Federal Penal Service of Russia in the Irkutsk Region in 2020. Based on an analysis of the service activities of the territorial bodies of the Federal Penal Service of Russia, a number of group disobediences and riots committed in penal institutions can be noted. With a decrease in the number of persons in detention, the number of crimes committed by them does not only stabilize, least of all decrease, but even tends to increase. In 2014-2019, the number of such crimes respectively amounted to 861, 940, 960, 977, 1025, 1151 [7]. It should be noted that this dynamics is also characteristic of crimes with an increased degree of public danger. Thus, the number of crimes under article 321 of the Criminal Code of the Russian Federation "Disorganization of the activities of facilities providing isolation from society" committed in correctional colonies during the specified period respectively amounted to 158, 166, 194, 180, 192, 238 [8].
For employees of facilities of the Federal Penal Service of Russia, it is necessary to distinguish the concept of "actions in case of emergency circumstances". In science and practice, it refers to the activities of the PCS personnel in the event of riots, the release of hostages, the escapes, the elimination of the explosions hazard, the attack defense, causing the necessity to use weapons, special means, apply special tactics and other force methods against the people in detention [9].
The PCS personnel must act confidently in a difficult operational environment. To provide this, it is important to have an effective staff training system [10]. In the author's opinion, preparation for action in case of emergency circumstances, shall be carried out in classes of tactical and special training. Tactical and special training involves the focused process of training employees, management bodies, integrated units in actions in case of emergency circumstances.
In the case of high-quality training within the framework of tactical and special training, the readiness of personnel to emergencies increases.
It is believed that readiness for action in the event of emergency circumstances is necessary for everyone to not be taken by surprise, not to become their victim, to correctly fulfill their functional duties, and especially for those who work in difficult conditions [11]. The fundamental condition for the success of the tasks realizations in the PCS is the readiness of the staff to fulfill them. The occurrence of a readiness state begins with the understanding of the task received, determining the motives and necessity for its implementation. Then there is an assessment of the situation, an assessment of the own forces, an assessment of the criminal, the determination of the nearest task, a further task, the development of a task implementation plan, models for its implementation, arrangement of future actions.
It is believed that emergency readiness is a dynamically variable structure with the complex formation and functional relationships among its components. When performing service tasks, the employee on the basis of internal activity (biological, physiological and mental) achieves certain goals [12, p. 204].
The content and structure of readiness are determined by the task performed, the operational situation, the type of equipment, weapons and special equipment used, the degree of readiness of the opposing side for active actions and other factors.
The readiness of the PCS employee to act in an emergency is expressed in his/her desire and motivation to fulfill a significant goal through willpower, to influence such components of his/her psyche as fears, instinct for self-preservation, selfishness, established behaviors. A staff member ready for the task should have sufficient willpower, awareness of the need to carry out the task and increased responsibility [13].
The increased level of readiness is determined by the employees' deep awareness of the goals of the upcoming actions, personal readiness to perform tasks and inflexible will to win, deep knowledge of special tactics, high physical condition, psychological stability, the effectiveness of the management system, confidence in the correctness of their superiors, a high level of moral and will qualities, comprehensive resource provision, corresponding to the needs and demands of employees, carrying out a set of measures to restore the combat effectiveness of forces and means, including psychological rehabilitation and restoration of spiritual forces.
The most significant indicators of readiness for emergencies are targeted mobilization of mental processes, properties, experience, their concentration on fulfillment of the obtained task and ways of its execution, availability of attitudes for maximum use of forces and means in order to overcome difficulties and achieve the set goal [14].
Works of Kugno [15, p. 49] have shown that the following factors are necessary for the creation of a state of readiness among employees for complex activities. As a result, the employee is set up to perform the received task and achieve the goal.
It is logical to consider the readiness of employees to act in emergencies as the variable degree of stable prerequisites for various personality conditions, including mental and physiological ones. Practice shows that the main and important indicators of the level of staff readiness for service are the existing internal affirmation of the staff member to perform the service tasks; conditions of preparation for the performance of service duties together with the organizational activity of the head; the mood in which the employee arrived at the service [16].
In our opinion, the essential definition of the readiness of employees for actions in case of emergency includes not only the personal readiness of the PCS employee for actions in an extreme situation, but also collective readiness. This could represent in performing a task as a part of a unit, for example, in case of eliminating riots, suppressing convicts group disobedience. In such cases, the success of the task completed depends on the quality of each person's actions. If one of the participants shows cowardice, irritability, depressed mood, signs of distress, either psychological infection or a disagreement with the necessary interaction can occur. In both cases, the risk of unsuccessful performance of the task assigned to the unit increases.
The team readiness is not a direct consequence or the sum of readiness of individual members. Its readiness is also influenced by public opinion, collective traditions, general mood and attitudes, the orientation of clear and hidden leaders. The readiness of the team is revealed in the general purpose, orientation, general will, collective skills, special coherence of actions, setting for the unquestionable and unfailing implementation of the superior's orders.
Conclusion
Thus, readiness for action in case of emergency circumstances in relation to the staff of the Federal Penal Service should be understood as a state formed on the basis of acquired knowledge, skills, qualities and ensuring the performance of service duties in case of aggravation of the operational situation at the facilities of the penal system. The readiness of employees of institutions of the Federal Penal Service of Russia in the current conditions of aggravation of criminal situation becomes especially relevant. Often, it is the high level of readiness for action in case of an emergency that allows employees to stop or eliminate it. The readiness of employees to act in case of emergency also contributes to the establishment of professionally competent relations in the team and is a regulator of the effectiveness of their further activity.
|
2021-08-03T00:06:23.135Z
|
2021-01-01T00:00:00.000
|
{
"year": 2021,
"sha1": "de35c83a3747ec54cd4760f51d501eaadd71c964",
"oa_license": "CCBY",
"oa_url": "https://www.shs-conferences.org/articles/shsconf/pdf/2021/19/shsconf_blf2021_05006.pdf",
"oa_status": "GOLD",
"pdf_src": "Adhoc",
"pdf_hash": "17a6a76bc95ea946fa0d83c63045cfe42b2a6952",
"s2fieldsofstudy": [
"Psychology"
],
"extfieldsofstudy": [
"Psychology"
]
}
|
13245069
|
pes2o/s2orc
|
v3-fos-license
|
Clinical profile, species‐specific severity grading, and outcome determinants of snake envenomation: An Indian tertiary care hospital‐based prospective study
A bs tr ac t Objective: We undertook this study to assess the clinical profile and outcome determinants of different snake envenomation as well as to assign species‐specific severity grade to different cases based on clinico – laboratory evidence scale. Materials and Methods: A prospective clinico – epidemiologic evaluation for outcome determinants of snakebite envenomation was carried out based on a clinico – laboratory severity grading scale, among 76 patients over a period of 2 years, in a tertiary care hospital in southern India. Results: Majority of patients were male agricultural workers (53.9%) followed by housewives (19.7%), and students (9.2%). Occurrence of viper snake envenomation with hemotoxic syndrome (73.68%) was highest followed by cobra and krait envenomation with neurotoxic (19.73%) and hemo – neurotoxic (5.3%) syndrome, respectively. On the contrary, maximum mortality and severity was seen in krait (60%) followed by cobra (13.33%) and viper (8.9%) envenomation. The average dose of anti‐snake venom (ASV) administered varied from 9.83 (±7.22) to 20.25 (±4.92) vials throughout grade I to IV in all snake species envenomation. An increase in severity grade, ASV dose, and mortality were observed with the corresponding delay in ‘bite to needle time.’ Also, initial traditional treatments and krait species envenomation were significantly associated with higher grades of severity and mortality. Conclusion: There is an urgent need to spread awareness among the community for avoidance of traditional treatment and any delay in medical intervention in snakebite incidents.
Introduction
Snake envenomation is a serious medical crisis, wherein the spectrum of injury can vary from local tissue damage to involvement of almost all vital organs of the body. Clinico -toxicologically, nature of snake envenomation is categorized into hemotoxic, neurotoxic, and myotoxic syndromes. [1] Most snakebites are harmless and are caused by non-poisonous species. Nonetheless, of the 3,000 different species of snakes, about 450 are found to be dangerous for humans worldwide. [2] Out of 216 Indian snake species, 52 are poisonous. [3,4] Among these, there are 4 major poisonous species viz. cobras or Naja naja, krait or Bungarus caeruleus, Russell's viper or Daboia russelli, and saw-scaled viper or Echis carinatus. [5] World health organization (WHO) has recognized snakebite as neglected and important public health problem in rural areas of tropical and subtropical countries situated in Asia, Africa, Oceania, and Latin America. [6] According to the same WHO report, the global annual incidence of envenoming and resulting deaths ranges from a minimum of 421,000 to a maximum of 1,841,000 and 20,000 to 94,000, respectively. Also, it is mentioned that the highest burden of snakebites is in South Asia, Southeast Asia, and sub-Saharan Africa. Among these, India has the highest incidence of snakebite-resulted mortality, ranging from 13,000 to 50,000 cases annually. [7,8] Attributes for such a high mortality due to snakebite are scarcity of anti -snake venoms (ASV), difficulties with rapid access to health centers, poor health services, and traditional treatments. [9,10] Furthermore, wrong/ erroneous identification of the snake species also leads to inappropriate/inadequate treatment and outcomes. [8] At presentation, a snake-bitten victim can be promptly diagnosed and treated if the clinical syndrome of snakebites are well-defined and pre -distinguished based on analysis of a series of reliably identified bites. [8] Thus, we undertook this priority area of research to study the clinical profile and outcome determinants of different snake envenomation as well as attempted to assign species-specific severity grade to different cases based on clinico -laboratory evidence scale.
Materials and Methods
We conducted a hospital-based prospective cohort study in patients admitted to a 2032-bedded tertiary referral center in South India during June 2008 to May 2010. A written informed consent was obtained from each subject included in the study. Ethical approval for the study was obtained from the hospital ethics committee prior to the commencement of the study. The catchment area for the study comprised of whole Udupi district in Karnataka, India which spreads over 929 square km, with a population of 529,225.
It is the hospital policy to admit all snakebite patients for observation, and patients with "dry bites" were discharged after 24 hours. Snakebite and species identification were confirmed by a reliable history from patients, eye witnesses, from the patient's attendants, and presence of fang marks and signs of local and systemic envenomation. Snakes brought by the patient to the hospital were identified by forensic medicine experts. All patients were interviewed using a standardized questionnaire to maintain a record of patient's history relevant to snakebite envenomation. Information about the victim, its management (first-aid/traditional treatment), time between bite, and administration of ASV was obtained in each case. Only the snakebite cases with signs of envenomation were included for the study. Non-poisonous snakebites and poisonous snakebite cases without envenomation were excluded. Patients were followed for 24 hours for "dry bite" confirmation. Patients with signs of envenomation were administered ASV manufactured by ViNS bioproducts limited, Kothur (Mandal), Mahaboobnagar (Dist.), Andhra Pradesh, India. This ASV is a sterile preparation containing equine immunoglobulin fragments F (ab') 2 freeze-dried powder when reconstituted to 10 ml of sterile water for injections I.P. supplied along with the vial, each 1 ml has capacity of specifically neutralizing the venom of these species of snake: −0.60 mg of dried Indian Cobra (Naja naja) venom, 0.45 mg of dried Common Krait (Bungarus Caeruleus) venom, 0.60 mg of dried Russell's viper (Dabioa russelli) venom, and 0.45 mg of dried Saw -scaled Viper (Echis carinatus) venom. These anti-snake venom equine immunoglobulin and their derivatives are obtained from serum of healthy equines immunized against venoms of the above species of snakes. Further, patients were managed as per the clinical judgment of consultant based on the nature of envenomation and the severity.
An attempt was made to identify the species of the snake involved and assign the degree of envenomation on a grading scale dependent on clinical symptoms/ signs and laboratory parameters [Appendix 1]. [11] Two viperidae species were combined for analysis purpose because of inability to differentiate from history or laboratory data. The average quantity of ASV required in each grade of severity was estimated. We also tried to determine the risk factors associated with poor outcomes in these patients. Patients who had grade III and IV severity and who succumbed to the complications were considered to have poor outcomes. The primary outcome was the severity of the snakebite as per grading and death. The secondary outcomes assessed were the dose of ASV administered and the duration of the hospital stay. All data were analyzed using the SPSS 15.0 statistical software package for Windows.
Results
A total of 130 cases of snakebites had presented to the hospital emergency department during the study period. Seventy-six patients [46 (60.52%) male, 30 (39.47%) female] satisfying the inclusion criteria were included in the present study. The rest of 54 cases had no signs of envenomation and were discharged after 24 hours. All the patients were in the age range of 16 (15.78%)]. Most of the cases occurred during the month of June to December. Seventy-five percent of patients were bitten between 6 PM to 12 AM, whereas 25% cases between 12 AM to 6 PM. Most of the bites were on the lower limb (77.63%); upper limbs were bitten in 21.95% of victims. Only one patient was bitten by krait on the trunk while sleeping over the floor.
Based on the symptoms, signs, and laboratory parameters, as described by Kumar et al. 2006, [11] we attempted to identify the species of the snake and found viper as the most common species identified in 73.68% of the cases [ Table 1]. Hemotoxic (defined as prolongation of PT, INR, and APTT, reversal was considered with normalization of PT, INR, and APTT) and neurotoxic nature of envenomation were observed in 56 (73.68%) and 15 (19.73%) cases respectively, whereas 4 (5.26%) cases had both hemotoxic and neurotoxic manifestations.
Clinical characteristics, complications, and severity grading of viper envenomation
Viper bite was characterized by severe local symptoms. Swelling and pain at the bite site were the commonest symptoms seen in 51 (91.07%) and 49 (87.5%) patients, respectively. Persistent bleeding from fang wounds was seen in 23 (41.07%) patients.
Other bleeding manifestations such as bleeding gums in 4 (7.14%), hematemesis in 2 (3.57%), hematuria in 1 (1.78%) cases were less common. Cellulitis (Presence of localized pain, erythema, and swelling) and coagulopathy were the most common complications seen in 44 (78.57%) and 35 (62.50%) patients, respectively. Oliguria and renal failure occurred in 16 (28.57%) and 12 (21.42%) patients, respectively. Other complications were sepsis (Systemic inflammatory response syndrome with proven or suspected microbial etiology) due to wound infection in 8 (14.28%), necrotizing fasciitis in 6 (10.71%), compartmental syndrome (presence of local neurovascular compromise as judged clinically; compartmental pressure measurements were not performed) with surgical intervention requirement in 3 (5.35%), and acute respiratory distress syndrome (ARDS) in 1 (1.78%) cases. The time between bite and ASV administration varied from 15 minutes to 36 hours, with majority of patients [32 (57.14%)] presenting within 6 hours. The ASV dose, duration of hospital stay, mortality, and severity grades were found to be increased with corresponding increase in 'bite to needle time.' ASV dose administered Table 2].
Clinical characteristics, complications, and severity grading of cobra envenomation
Pain followed by swelling and ptosis were seen in 11 (73.33%) and 13 (86.66%) patients with cobra bite, respectively. Difficulty in breathing and weakness of the limbs were seen in 11 (73.33%) patients. Diplopia, dysphagia, and dysarthria were seen in 4 (26.66%), 3 (20%), and 1 (6.66%) patients, respectively. Cellulitis seen in 12 (80%) and respiratory paralysis in 11 (73.33%) patients were the most common complications followed by sepsis and necrotizing fasciitis, which were seen in 3 (20%) patients. Renal failure, persistent vegetative state following cardiac arrest, refractory hypotension, and gangrene were the other complications occurred in 1 (6.66%) patient. As seen with viper bites, in cobra envenomation also, the grade of severity, duration of Table 2].
Clinical characteristics, complications, and severity grading of krait envenomation
Abdominal pain was one of the common symptom in 4 (80%) patients followed by neurological symptoms like dyspnea in 4 (80%), ptosis in 3 (60%), dysphagia in 2 (40%), and oliguria in 1 (20%) patient. Respiratory paralysis was seen in 4 (80%) patients. Hypokalemia was seen in 3 (60%) patients. Two (40%) patients presented with coma, and 1 (20%) patient had renal failure. Of the 5 patients with krait bite, 4 (80%) presented with grade IV severity whereas, 1 (20%) with grade II severity. The average time between bite to administration of ASV was 11.25 (±2.21) hours in patients with grade IV severity. The average dose of ASV used was 20.25 (±4.92) vials in patients in grade IV severity, whereas in only 1 patient with grade II severity, the ASV dose administered was 24 vials. Three (60%) patients' required ventilator support and all of them expired [ Table 2].
Determinants of outcome
An initial visit to a traditional healer was statistically associated with an increased risk of death with an odds ratio of 4.22. Of the 13 (17.1%) patients who received traditional forms of treatment, 11 (14.47%) had grade IV severity and 4 (5.26%) expired. Delay in presentation to hospital was significantly longer for victims with higher grades (III and/or IV) of severity and mortality, which makes it a risk factor associated with poor outcome [ Table 3]. Severity grade was found to be increased with corresponding increase in 'bite to needle time.' Mortality proportion was higher in patients with late presentation (>24 hours) than early presentation (<6 hours); P value was significant (<0.01). Krait bite itself was found to be risk factor with 60% mortality. The mortality and corresponding proportion was calculated for each species [ Table 4]. Chi-square test for trend was used to find the association between the severity grading and the species. Since P = 0.009, there is a significant difference in distribution of grading between the species. Chi-square test was used to compare the mortality rates between the species. Since P = 0.013, there is a significant difference in the mortality rates between the species.
Discussion
The present study was carried out on 76 cases of snakebite envenomation. Male preponderance seen in our study is in close agreement with earlier studies [12][13][14] and may be attributed to their life styles involving outdoor activities and occupational exposure as farmers or herdsmen. Young male agricultural workers were the most common affected group in our study, making snakebite an occupational hazard. We observed the occurrence of most of the cases during the month of June to December, which is the monsoon season with agricultural activity in our study region, and also the time of increased activity of snakes as they come out of their shelters. [15] Maximum events of envenomation were seen with viper snakebite followed by cobra and krait species, whereas the proportion of mortality was highest in krait followed by cobra and viper species. This indicates that our study region i.e., Udupi district of Karnataka state in southern India is more prevalent for viper snakes envenomation than other snakes. Distribution of different snake species in a given geographic region is dependent on various environmental and climatic conditions viz. rainfall, altitude, vegetation, and abundance of preferred prey etc. [16,17] Hemotoxic syndrome was the commonest snakebite syndrome, attributable to the prevalence of viper snake species (hallmark for hemotoxic syndrome), followed by neurotoxic syndrome of cobra bites (hallmark for neurotoxic syndrome). [8,18] Simultaneous neurotoxic and hemotoxic manifestations in patients were attributable to krait envenomation. [10] These results are very similar to other reports from the neighboring states of Karnataka and other states in India. [10,19,20] Severe local symptoms observed among viper envenomation were also in accordance with other studies. [19,[21][22][23] Swelling and pain at the site of the bite were the most common symptoms among the patients with viper envenomation. Cellulitis and coagulopathy were found to be the most common complications with viper envenomation in accordance with other studies. [22][23][24] Renal failure was attributable to tubular damage by venom, hemoglobinuria, rhabdomyolysis, hypotension, and renal microthrombi formation causing acute tubular necrosis. [23,25,26] The rationale for the use of ASV is well-defined; doses required in different envenomation situation vary greatly and are subject to the severity grade and the snake species associated. [3,18,27] In the present study, the severity grade increased as the time delay between the occurrence of bite and ASV administration i.e., 'bite to needle time' increased. Severity grade was also found to be in direct proportional relation with the duration of hospital stay, the mean effective neutralizing dose of ASV, the time taken for normalization of coagulation abnormalities as well as the mortality. Possibility of venom's direct toxicity on organ system is pointed by the direct proportionality between incidences of complication with venom neutralization time and delay in ASV administration. [9,28,29] Hence, an early institution of ASV is beneficial in preventing complications.
As observed with viper, in cobra envenomation also, the severity grade increased as the bite to needle time increased. Patients, who presented late, resulted in delayed administration of ASV and required more aggressive therapies like mechanical ventilator support. Of the 5 patients, identified to have bitten by Krait, 4 had grade IV severity and 3 expired. Since Krait bites are usually painless with mild local symptoms, [30] victims might not have recognized the bite and/or have not been sufficiently compelled to seek treatment immediately, resulted in delayed ASV administration and higher mortality. Patients with traditional treatments were at a higher risk of complications and mortality with an odds ratio of 4.22. Consultation with traditional healers is a classic cause of delay, which exposes the patient to useless or dangerous interventions. [7,30] Thus, there is an urgent need to spread awareness among the community for avoidance of any delay and traditional treatment in snakebite incidences. Coagulopathy could not be extensively evaluated in this study, and 2 viperidae species were studied together as it was not possible to differentiate by history and/or laboratory details, which are the limitations of the present study.
To summarize, initial consultation with traditional healers, delayed ASV administration, and krait bites were the determinants of poor outcome in our study. Majority of snake envenomation resulted from hemotoxic viper snakebites and presented with grade IV severity. Viper envenomation was also associated with significant local reaction and systemic manifestations due to coagulation abnormalities. The presence of neurological symptoms and signs with absence of local reaction, presence of abdominal pain, and hypokalemia favors a diagnosis of krait envenomation. Avoidance of consultation with traditional healers as well as prompt medical intervention can reduce both the morbidity and mortality in snakebite envenomation cases.
Ethical Approval
The study was approved by the Kasturba hospital, Manipal ethics committee.
|
2018-04-03T00:34:13.805Z
|
2012-10-01T00:00:00.000
|
{
"year": 2012,
"sha1": "6b99c44cf6c257fe50d2b1cbf5f9a3238fd82f21",
"oa_license": "CCBYNCSA",
"oa_url": "https://europepmc.org/articles/pmc3610449",
"oa_status": "GREEN",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "95f7fba8c831297586b391061aa3f4cf859b2547",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
85567184
|
pes2o/s2orc
|
v3-fos-license
|
Prediction of drug-disease associations based on ensemble meta paths and singular value decomposition
Background In the field of drug repositioning, it is assumed that similar drugs may treat similar diseases, therefore many existing computational methods need to compute the similarities of drugs and diseases. However, the calculation of similarity depends on the adopted measure and the available features, which may lead that the similarity scores vary dramatically from one to another, and it will not work when facing the incomplete data. Besides, supervised learning based methods usually need both positive and negative samples to train the prediction models, whereas in drug-disease pairs data there are only some verified interactions (positive samples) and a lot of unlabeled pairs. To train the models, many methods simply treat the unlabeled samples as negative ones, which may introduce artificial noises. Herein, we propose a method to predict drug-disease associations without the need of similarity information, and select more likely negative samples. Results In the proposed EMP-SVD (Ensemble Meta Paths and Singular Value Decomposition), we introduce five meta paths corresponding to different kinds of interaction data, and for each meta path we generate a commuting matrix. Every matrix is factorized into two low rank matrices by SVD which are used for the latent features of drugs and diseases respectively. The features are combined to represent drug-disease pairs. We build a base classifier via Random Forest for each meta path and five base classifiers are combined as the final ensemble classifier. In order to train out a more reliable prediction model, we select more likely negative ones from unlabeled samples under the assumption that non-associated drug and disease pair have no common interacted proteins. The experiments have shown that the proposed EMP-SVD method outperforms several state-of-the-art approaches. Case studies by literature investigation have found that the proposed EMP-SVD can mine out many drug-disease associations, which implies the practicality of EMP-SVD. Conclusions The proposed EMP-SVD can integrate the interaction data among drugs, proteins and diseases, and predict the drug-disease associations without the need of similarity information. At the same time, the strategy of selecting more reliable negative samples will benefit the prediction.
Background
De novo drug discovery is a complex systematic project which is expensive, time-consuming and with high failure risks. As reported, it will take 0.8-1.5 billion dollars and about 10-17 years to bring a small molecule drug into market, and during the development stage, almost 90% of the small molecules can not pass the Phase I clinical trial and finally be eliminated [1,2]. For the approved drugs in market, their pharmacological and toxicological properties are clear and the drug safeties are often guaranteed, but only some of their indications are found. For example, there are 2589 approved small molecule drugs in DrugBank [3], and more than 25000 diseases in UMLS medical database [4], resulting in over 60 millions of drug-disease pairs. However, only less than 5% of the drug-disease pairs were identified to have therapeutic relationships, and most of the drug-disease relationships are unknown [5]. Therefore, to discover the new indications of approved drugs, known as drug repositioning, can greatly save money and time, especially can improve the success rate, has become a promising alternative for de novo drug development.
Historically, finding a new indiction of a drug is likely to be an accidental event with a bit of luck. For example, Minoxidil, originally for the treatment of hypertension, was found by chance to have the treatment efficacy for hair loss [6]; Sildenafil (trade name: Viagra), originally for the treatment of angina, was occasionally found to have the potential to treat erectile dysfunction [7]. Such occasional findings of the drugs' new indictions suggest a new methodology of drug development. However, the "pot-luck" approach can not promise drug repositioning effectively and efficiently. It is necessary to develop a computational method that helps to redirect approved drugs. Fortunately, with the accumulation of multiple omics data and the development of machine learning methods, it is possible to mine the drugs' potential indications in silico. Up to now, many computational methods have been proposed to find new indictions of drugs by predicting potential treatment relationships of drug-disease pairs.
Based on the hypothesis that the gene expression signature of a particular drug is opposite to the gene expression signature of a disease, some gene expression based methods [8,9] have been proposed. Noticing that such kind of methods may fail to consider the different roles of genes and their dependencies at the system level, system-level based approach that integrates the gene expressions and related network has recently been proposed [10].
Recently, along with the increase of drugs and diseases related multi-omics data, many methods have been proposed to integrate multiple sources of data to predict the drug-disease interactions based on machine learning techniques. Gottlieb et al. proposed a method (PRE-DICT) to predict new associations between drugs and diseases by integrating five drug-drug similarities and two disease-disease similarities data [11]. Wang et al. proposed a computational framework based on a three-layer heterogeneous network model (TL-HGBI) by integrating similarities and interactions among diseases, drugs and drug targets [12]. Luo et al. utilized some comprehensive similarities about drugs and diseases, and proposed a Bi-Random walk algorithm (MBiRW) to predict potential drug-disease interactions [13]. Martinez et al. developed a method named DrugNet for drug-disease and disease-drug priorization by integrating heterogeneous data [14]. Wu et al. integrated comprehensive drug-drug and disease-disease similarities from chemical/phenotype layer, gene layer and treatment network layer, and proposed a semi-supervised graph cut method (SSGC) to predict the drug-disease associations [15]. Moghadam et al. adopted the kernel fusion technique to combine different drug features and disease features, and then built SVM models to predict novel drug indications [16]. Liang et al. integrated drug chemical information, target domain information and gene ontology annotation information, and proposed a Laplacian regularized sparse subspace learning method (LRSSL) to predict drugdisease associations [17]. Zhang et al. introduced a linear neighborhood similarity [18] and a network topological similarity [19], then proposed a similarity constrained matrix factorization method (SCMFDD) to predict drugdisease associations by making use of known drugdisease associations, drug features and disease semantic information [20].
However, most of the existed methods are facing two main problems: one is that most of them are based on the hypothesis that similar drugs treat similar diseases, thus they need the similarity information between drugs, proteins, diseases, and so on. However, the similarity data can be not easily obtained. People often need to customize a program to collect data and to calculate the similarities so as to satisfy their own needs. Moreover, the calculation of similarity scores depends on the adopted measures, which may lead that the similarity score of a pair varies dramatically from one method to another. For example, two proteins are similar according to their structures, while they may be dissimilar according to their sequences. Even worse, some features required for calculating the similarities may be unknown or unavailable, resulting that these methods fail to work [21]. The other problem is that supervised learning based methods usually need both positive and negative samples to train the prediction models, whereas the drugdisease pair data, like other biological data, is lack of experimental validated negative samples. To train the models, most of the existing methods randomly select some unlabeled samples as the negative ones. Obviously, such strategy is very rough, for we are not sure whether there are some positive samples uncovered in the unlabeled data.
In this paper, we propose a method, called EMP-SVD (Ensemble Meta Paths and Singular Value Decomposition), to detect drug-disease treatment relations by using drug-disease, drug-protein and disease-protein interaction data. Unlike other methods, EMP-SVD needs no similarity information at all. In order to integrate different kinds of interaction data and consider different dependencies, we introduce five meta paths. For each meta path, we first generate a commuting matrix based on the corresponding interaction data, and then get latent features of drugs and diseases by using SVD (Singular Value Decomposition). All drug-disease pairs can be represented by the features. Finally, we train a base classifier by using the Random Forest algorithm. Five base classifiers are combined as an ensemble model to predict the drug-disease interactions. The framework of our method is shown in Fig. 1. In order to train out a more reliable prediction model, we select more likely negative ones from unlabeled samples under the assumption that non-associated drug and disease pair have no common interacted proteins, which is different from other methods. To evaluate our proposed method, we will compare it with the stateof-the-art methods, and also do case studies by literature investigation.
Data sets
In this paper, we mainly made use of the interaction data of drug-disease, drug-protein and disease-protein to build the prediction model. We collected such data from DrugBank [3,22,23], OMIM [24] and Gottlieb's data set [11]. Concretely, we collected 4642 drug-protein interaction data from DrugBank, involving 1186 drugs and 1147 proteins; 1365 disease-protein interactions from OMIM, involving 449 diseases and 1147 proteins; and 1827 drugdisease interactions from Gottlieb's data set, involving 302 disease, 551 drugs. Obviously, the heterogenous network composed of drugs, proteins, diseases and the known interactions is sparse. The statistic of the data is shown in Table 1.
Although our method does not need the similarity information, most of other machine learning based methods do need. For the convenience of comparison, we still collected the chemical structure of drugs and the sequence data of proteins from DrugBank. We computed the drugdrug chemical similarities according to their SMILES strings [25] via Openbabel tool [26], and the proteinprotein similarities according to the sequence data by Smith-Waterman algorithm [27]. Moreover, we directly downloaded the disease-disease similarities from Mim-Miner [28].
Definitions and notations
In this section, we will give the formal definitions and notations used in this paper. The network schema M G severs as a template of a network G. For a drug-protein-disease heterogenous network, the network schema is shown in Fig. 2.
For simplicity, we also omit the link types in denoting the meta path if there is no multiple links between the two types, for examples, The length of P is the number of links in P.
where A T i T j is the adjacency (interaction) matrix between type T i and type T j . X(i, j) represents the number of path instances between object u i ∈ T 1 and object v j ∈ T k under meta path P.
Since we want to detect the interactions between the drugs and the diseases, we only consider the cases of T 1 = Drug and T k = Disease.
Now that there are only three kinds of nodes (drug, protein and disease) in the heterogenous network, we think the meta path with length greater than three may be too long to contribute to the prediction. Sun's work also has shown that short meta paths are good enough, and long meta paths may even reduce the quality [29]. Therefore, in this work, we only selected meta paths with length no longer than three. As a result, we select five meta paths described below.
Let A ds be the drug-disease interaction matrix, A dp be the drug-protein interaction matrix, and A sp be the disease-protein interaction matrix, we can get the commuting matrices of the five meta paths as follows: The commuting matrix of it, denoted as X1 , can be obtained by: The commuting matrix of it, denoted as X2 , can be obtained by : By using meta-path-2, we can integrate the drug-protein interaction information and the disease-protein interaction information, that is to say, we easily take the protein related information into account.
Meta-path-3: Drug By using meta-path-3, we can integrate drug-protein interaction and drug-disease interaction information.
What's more, meta-path-3 also indicates that if two drugs share some common proteins, they may have similar indications.
Meta-path-4: Drug
The commuting matrix of it, denoted as X4 , can be obtained by : By using meta-path-4, we can integrate the drug-disease interaction information. Besides, meta-path-4 also indicates that if two drugs share some common indications, then the indication of one drug may also be the potential indication of another drug.
Meta-path-5: Drug
The commuting matrix of it, denoted as X5 , can be obtained by : By using meta-path-5, we can integrate the drug-disease interaction and the disease-protein interaction information. What's more, meta-path-5 also indicates that if two disease share some common proteins, the drug for treating one disease may also be the potential therapeutical drug for another disease. As the definition, the element X(i, j) of the commuting matrix X denotes the number of path instances from drug d i to disease s j under the corresponding meta path. We show an example in Fig. 3. There are two path instances from drug d 3
Feature extraction with singular value decomposition
Now that element X(i, j) in a commuting matrix X denotes the number of path instances from the drug d i to disease s j , then row i in the commuting matrix can be used as features of drug d i , and column j can be used as features of disease s j . And we can use the concatenation of them to represent the drug-disease pair. Suppose there are m drugs and n diseases, we will have m + n (In this work, m = 1186, n = 449) features to represent the drugdisease pair. By contrast, the number of drug-disease pairs is small (We only have 1827 known interactions in this work). Obviously, the feature dimension is relatively high, which is not proper to construct a robust prediction model. Now that the singular value decomposition (SVD) has been successfully used to reduce the dimension in Fig. 3 An example of the meaning of commuting matrix many researches, we also employed SVD to extract small number of features in our work.
By using SVD, the commuting matrix X ∈ R m×n can be factorized into U, and V such that where U ∈ R m×m , ∈ R m×n and V ∈ R n×n . The diagonal entries of are equal to the singular values of X (Other elements in other than diagonal entries are 0 ). The columns of U and V are, respectively, left-and rightsingular vectors for the corresponding singular values.
As is known to all, the magnitude of the singular values represents the importance of the corresponding vectors; and in , the singular values are ordered in descending order. Moreover, in most cases, the sum of the first 10% or even 1% of the singular values is over 99% of the total sum of all singular values. Specifically in this drug-disease associations prediction problem, in the biomedical meaning, the most useful information about drug and disease features will be included in the first 10% even less singular values. In the process of dimensionality reduction, the useful data will not be lost, but the redundant information will be discarded. That is to say, we can use the top r singular values to approximate the matrix X: where r min(m, n). Row i in U can be used as latent features of drug d i , and row j in V can be used as latent features of disease s j . As a result, the dimension of the latent feature vector of each drug-disease pair can be reduced to 2 * r. In this work, we will introduce a parameter latent_feature_percent far less than 1 (say 1%, 2%,...) to control the value of r such that r = latent_feature_percent × min(m, n).
Selection of likely negative samples from unlabeled drug-disease pairs
To build a prediction model by using supervised learning, we need both positive and negative samples. The known drug-disease treatment relations are positive samples. Being lack of validated negative samples, most methods simply select some of unlabeled samples as negative ones by random. However, the unlabeled samples are not necessarily negative, some of them may be positive samples that still remain uncovered by experiments [30]. Different with other methods, we try to find more reliable negative samples from the unlabeled ones in this work.
If a drug shares some proteins with a disease, then the drug may have potential to treat the disease. Intuitively, if a drug and a disease have no common related proteins, we can think the disease is not the indication of the drug, and thus the drug-disease pair is more likely a negative sample. By this means, we can select out more reliable negative samples from the unlabeled pairs based on the drugprotein and disease-protein interactions information. The procedure is listed in Algorithm 1.
Construction and ensemble of classifiers
The five meta paths we have selected to integrate heterogeneous data reflect different aspects of the drug-disease treatment relationship, such as two drugs with common proteins having similar indications, two drugs sharing one common indication also sharing another indication, and so on. Thus we can build five base classifiers for the prediction of drug-disease treatment relations from different sides. In our work, the base classifiers are built based on the Random Forest algorithm which was implemented by using the RandomForestClassifier function in the scikit-learn package [31], we set the number of trees as 256.
Since ensemble learning can often help to improve the performances [32,33], after the five base classifiers are constructed, we can obtain an ensemble classifier. For an input of drug-disease pair, each base classifier outputs two probabilities indicating that the pair being negative and positive respectively. Since we want to know whether the pair has treatment relation, we only take the positive probability as considered in the ensemble model.
For a drug-disease pair x with unknown label, suppose the predicted score (probability) of each base classifier be h i (x), i = 1, 2, ...5, we used average strategy to get the final score of the ensemble model: If H(x) is greater than a predetermined threshold, then the sample x is predicted as the positive. Because F 1measure is a comprehensive metric, in this work, we let the program automatically determine the threshold value when F 1 -measure reaches the maximum value, which is the same strategy as the other researchers used.
Experiments and results
We perform 5-fold cross validation to evaluate our method. Since the filtered negative samples are more than the positive ones, we randomly select a subset from them that with size equal to the positives, and use the balanced data to train the models. We first select the appropriate number of features according to the relationship of the model performance and the feature number. Then we did three kinds of evaluation experiments: (1) We investigate whether our negative samples filtering strategy can help to improve the prediction performance; (2) We compare EMP-SVD with other state-of-the-art methods by using the same data; (3) We check the practicality of our method by doing case studies.
Evaluation metrics
Just as most other work, we where TP, FP, TN and FN denote the number of true positive samples, false positive samples, true negative samples and false negative samples, respectively. Since Precision(PRE) and Recall(REC) have some conflicts, in general, a classifier gets a higher PRE will have a lower REC, and vise versa. To get a comprehensive performance, Area Under Precison-Recall Curve(AUPR) and Area Under Receiver Operating Characteristic Curve(AUC) are often used. AUPR takes both PRE and REC into account, AUC takes both the true positive rate(TPR, the same as REC) and the false positive rate (FPR) into account, so they are comprehensive metrics. At the same time, with the help of the curves we can intuitively find which classifier is better. Therefore, in this work, we adopted AUPR and AUC as the main metrics.
Determination of appropriate number of features
Parameters are often used in existing computational methods, which limits the generalization of a model. So, it will be better to use fewer parameters or to get an analytical solution.
In this work, we just need to determine the number of singular values (corresponding to the feature number that is controlled by the parameter latent_feature_percent) during the model construction, which is very different with most state-of-the-art methods. Just mentioned above r min(m, n), so we set latent_feature_percent as 1%, 2%, 3%, ......, 20% respectively, and the performance curves of five base classifiers and the ensemble one with different latent_feature_percent are shown in Fig. 4. The results have shown that the performances of the ensemble classifier are better than other five base classifiers, illustrating that our ensemble rule is effective. Moreover, the performances of the six classifiers are robust across different parameter settings. Anyway, we set latent_feature_percent as 3% according to the curves in this work.
We also find that the performances of classifiers based on meta-path-1 and meta-path-4 are the worst. Noticing that both meta-path-1 and meta-path-4 just take drug-disease interactions into consideration, while the other three meta paths contain more information on drug-protein or protein-disease interactions, we think integrating more interaction information into the meta path can help to improve the performance of the classifier.
Investigation of the filtering strategy of negative samples
Being lack of validated negative samples, most of the other methods randomly select unlabeled samples to be negative ones. However, the unlabeled samples are not necessarily negative, some of them may be positive samples still uncovered by experiments. So in this work we selected out more likely negative samples from unlabeled ones according to the common protein information (as described in Algorithm 1). As shown in Table 2, all the classifiers achieve better performances in most metrics when using our negative samples filtering strategy. We also noted that the improvement is little, which may due to the fact that the known drugprotein interactions and disease-protein interactions are too few (with density of 0.0034 and 0.0027, as shown in Table 1), resulting that very few proteins could be used in the filtering process. Anyway, our strategy for selecting more reliable negative samples is useful, feasible and interpretable. We believe that along with the increase of interactions data, we will get more reliable negative a b Fig. 4 Influence of different latent_feature_percent on the a AUPR b AUC samples and thus achieve more great performance improvements.
Comparison with other methods
In this section, we compare EMP-SVD with state-of-theart methods to demonstrate the superior performance of our method. PREDICT [11] and TL-HGBI method [12] are classical methods used to predict the drug-target and drug-disease interactions. MBiRW [13], LRSSL [17] and SCMFDD [20] are the methods proposed in these two years, and achieved high performance in the prediction of drug-disease interaction. So we choose these state-of-theart methods to compare. PREDICT calculates the score of a given drug-disease pair (d r , d i ) according to all the known drug-disease pairs d r , d i associated with that given pair by equation TL-HGBI is a three layer heterogenous network model, which makes use of the similarities and interactions of drugs, diseases and targets by iterative update. MBiRW adjusts the similarities of drugs and diseases by correlation analysis and known drug-disease associations, then uses Bi-random walk algorithm to predict the potential drug-disease associations. LRSSL is a Laplacian regularized sparse subspace learning method used to predict the drug-disease associations which integrates drug chemical information, drug target domain information and target annotation information. SCMFDD is a similarity constrained matrix factorization method for the prediction of drug-disease associations by using known drug-disease interactions, drug features and disease semantic information. We obtained the source code of PREDICT, TL-HGBI and SCMFDD from the authors, the code of MBiRW, LRSSL are publicly available, and the parameters were set according to their papers. The parameter latent_feature_percent in EMP-SVD was set 3%. To be fair, the five parts data were kept the same division in all methods when conducting 5-fold cross validation.
As shown in Table 3, compared with other five stateof-the-art methods which make use of several kinds of similarities as well as the interaction data, the proposed Table 2 Performances comparison with different negative samples selecting strategies (random strategy is denoted "random", our strategy is "reliable") classifier EMP-SVD only uses the known interaction data but achieves better performances in most metrics, especially the comprehensive metrics (AUPR and AUC). To make it more intuitively, we plotted the Precison-Recall Curve and ROC curve, which are shown in Fig. 5a and b, respectively. The AUPR and AUC of the proposed EMP-SVD are 0.956 and 0.951, respectively, better than the compared methods. Hence, it shows the simplicity and effectiveness of our method.
Case studies
Here, we test the practicality of EMP-SVD for predicting unknown associations. Except for training set composing of the known 1827 drug-disease associations and randomly selected 1827 negative samples by using our strategy, we used the trained EMP-SVD model to predict the associations for other unknown drug-disease pairs, and validate the results by literature investigation.
The new predicted top 20 drug-disease associations are shown in Table 4. We checked them carefully by literature validation and found that 13 of the top 20 predicted associations have been reported in the literatures. And these predicted associations were not originally in our data set, but we could find it out by our method, thus showing the practicality of our proposed EMP-SVD.
It should be noted that Triamcinolone (DrugBank ID: DB00620) and Betamethasone (DrugBank ID: DB00443), as glucocorticoid, are commonly used in the treatment of various skin diseases such as "Eczema" [34][35][36], and we find that their predicted associations include the disease "Growth Retardation, Small And Puffy Hands And Feet, And Eczema" (OMIM ID:233810). During the process of literature validation, we also find a case of growth retardation and Cushing's syndrome due to excessive application of betamethasone-17-valerate ointment [37]. In a responsible attitude, we think that whether they can be used to treat the disease "Growth Retardation, Small And Puffy Hands And Feet, And Eczema", or the usage and dosage should be further carefully studied by the chemists and doctors, especially should be with caution when used on children and pregnant women.
In more details, we checked the predicted potential indications of drug "Amitriptyline" (DrugBank ID: DB00321). Amitriptyline is a tricyclic antidepressant which is often used to treat symptoms of depression with the brand name: Vanatrip, Elavil, Endep. As shown in Table 5 Breast cancer is a relatively common malignant tumor for female, which seriously endangers women's health and life safety. To discover the potential drugs is of great value. So we also checked the drug list that have been predicted to treat the disease "Breast Cancer" (OMIM ID: 114480). In the top 10 drugs, as shown in Table 6, we found that 8 have been reported to be used in the clinical treatment.
Therefore, the case studies have further shown the practicality of the proposed method EMP-SVD.
Conclusions and discussions
To uncover the potential drug-disease associations is an important step in drug development, but it is timeconsuming and costly to uncover them by wet experiments. Along with the accumulation of drug and disease related multi-omics data, as well as the development of machine learning techniques, more and more computational methods have been proposed to predict the potential drug-disease associations. To help the prediction, many methods integrate multiple source of data, including drugs, diseases, targets, side effects, and so on. They achieved good performances and could provide a helpful reference to the drug development. Most of them need the similarities of drug and disease related data. However, the similarity data can not be easily obtained, and people often need to customize a program to crawl data and to compute the similarities to satisfy their own need. Even worse, some features needed to calculate the similarity are unknown or unavailable. These methods will not work facing the incomplete data. Besides, being lack of validated negative samples in the prediction of drugdisease associations, most of the machine learning based methods assume the unlabeled samples to be negative ones in the training of the model. Such strategy may input errors because there may be positive samples uncovered in the unlabeled samples. What's more, most of the existing methods use many parameters in the data integration and the model construction. The parameters are difficult to tune, which limits the generalization ability of the method. In this work, we proposed a method named EMP-SVD to predict drug-disease interactions based on ensemble meta paths and singular value decomposition. Five meta paths from source node (drug) to end node (disease) were selected to integrate the interaction information of drugs, proteins and diseases. Then the commuting matrices of these meta paths were calculated out, each element indicates the number of path instances between the corresponding drug and disease pair. By using singular value decomposition on the commuting matrices, we can extract small number of latent features of drugs and diseases. In order to get reliable negative samples, we selected those unlabeled samples as negative under the assumption that if a drug and a disease have no common proteins, then there is smaller probability for them to be treatment relationship. Based on each meta path we first built a base classifier, and then combined them to get an ensemble classifier. The experiments results have shown that our proposed EMP-SVD method outperformed several state-of-the-art methods. Better than other methods, EMP-SVD has few parameters and very easy to set. Further more, case studies have shown the predicted new associations could be useful for further biomedical research, which demonstrate the practicality of our method. Although there are meta path based methods in social network and some other networks, to the best of our knowledge, it is the first work in the prediction of drugdisease associations by using ensemble meta paths and singular value decomposition. Different with many existing methods, we do not need the similarity data which are not easily obtained or sometimes unavailable or unknown. Instead, we just use the interaction data which can be easily accessed in many databases to build the prediction model. The other advantage of method is that there is only one parameter that can easily set. Though we use ensemble strategy to improve the performance, each of the five base classifiers can independently act as the model as well to predict the drug-disease interactions. Since there are many computational methods to predict the target proteins for a new drug such as docking methods. For a new drug which has no known interactions with any diseases, we still can predict its interacted diseases by building classifier using meta-path-2 by making use of drug-protein and protein-disease interactions.
Though the results of our methods are promising, there are still some limitations. Firstly, we only use the information of drugs, proteins and diseases, there are many other information could also be integrated in the further work, such as the information of side effects, pathways, tissues, and so on. Secondly, we only make use of common proteins to select out the negative samples, some other information such as gene expression data can also be used for this purpose. Or we can directly build the model by positive and unlabeled samples based learning method. We will address these issues in the future study.
|
2019-03-30T13:37:40.733Z
|
2019-03-01T00:00:00.000
|
{
"year": 2019,
"sha1": "1723551ad2e1d666ded8bb477ed7f331eb46f5d0",
"oa_license": "CCBY",
"oa_url": "https://bmcbioinformatics.biomedcentral.com/track/pdf/10.1186/s12859-019-2644-5",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "1723551ad2e1d666ded8bb477ed7f331eb46f5d0",
"s2fieldsofstudy": [
"Computer Science"
],
"extfieldsofstudy": [
"Medicine",
"Computer Science"
]
}
|
244269559
|
pes2o/s2orc
|
v3-fos-license
|
Online public discourse on artificial intelligence and ethics in China: context, content, and implications
The societal and ethical implications of artificial intelligence (AI) have sparked discussions among academics, policymakers and the public around the world. What has gone unnoticed so far are the likewise vibrant discussions in China. We analyzed a large sample of discussions about AI ethics on two Chinese social media platforms. Findings suggest that participants were diverse, and included scholars, IT industry actors, journalists, and members of the general public. They addressed a broad range of concerns associated with the application of AI in various fields. Some even gave recommendations on how to tackle these issues. We argue that these discussions are a valuable source for understanding the future trajectory of AI development in China as well as implications for global dialogue on AI governance.
Introduction
"When the young people of Generation Z start discussing AI ethics, constructing a set of perfect rules for AI governance is no longer out of reach." So claimed Alter, the author of a widely read and reposted article published on Chinese social media WeChat (2020). The post referred to a popular vlog produced by a journalism student from Tsinghua University on the video sharing site BiliBili who interviewed two young researchers from China's top AI start-up Megvii about the ethical issues around AI (Xiaosu 2020). Although Alter's expectations for "perfect" AI governance rules may be overly optimistic, it is worth noting that the rapid development and deployment of AI in China has been accompanied by growing societal discussions about its social and ethical implications. Social media platforms have become important fora for multi-stakeholder exchanges on fundamental questions, hopes, concerns and recommendations on how AI should be developed, used and regulated.
How AI is represented, communicated and perceived in cultural narratives can profoundly influence research, reception and regulation (Cave et al. 2018). Existing analyses of public communication about AI, by means of English-language media content analysis, have found that media coverage tends to focus on its economic impact, despite the recently growing attention to related ethical issues (Chuan et al. 2019;Ouchchy et al. 2020). Moreover, surveys among the general public found skewed perceptions that are either utopian or dystopian. Lack of awareness regarding the realistic implications of AI is a significant hurdle in its uptake for social benefit ).
China's emergence as a global leader in the field of AI raises the importance of understanding its development trajectory in this specific cultural context. The government's ambitious AI strategy, regulatory approaches, the vast resources invested, and the capabilities of Chinese AI players are important factors shaping this trajectory (Allen 2019; Sheehan 2019; Roberts et al. 2021;Colvin et al. 2020;Ding 2018). However, despite the party-state's control over the public sphere, societal discourse in China can shed light on what sociotechnical future the population of over one billion can imagine and is currently negotiating (Jasanoff and Kim 2015). With the aim of examining online public discourse The Chinese government is aiming to establish an ethical framework and a system of laws and regulations to govern AI by 2030 (State Council 2017). Several high-level ethical guidelines such as the Beijing AI Principles and the Governance Principles for a New Generation of AI have been released as collaborative efforts between government-affiliated research institutions, universities, and leading companies. However, translating lofty principles into meaningful moral axioms that industry and social actors can relate to requires public engagement. AI-related science popularization activities, as well as public opinion guidance, were also listed as key tasks to be completed in the 2017 New Generation AI Development Plan (State Council 2017).
Social media platforms play an increasingly important role in mediating China's public sphere as the result of rapid adoption of the Internet and smartphones. However, the role these platforms play is by no means simply as channels for unrestricted public opinion. Besides censorship, social media content is often subjected to overt propaganda or covert efforts to guide opinion (Tai and Fu 2020;Creemers 2017). The latter are found to be done by both hired "internet commentators" or "50 cent party members" 1 and public intellectuals (King et al. 2017;Abb 2021). In the case of discourse relating to AI in general, and not any specific ethical aspects, researchers have found that social media content largely conforms to the party-state media's framing of AI's economic benefits, with little evidence of critical debates (Zeng et al. 2020).
Nevertheless, social media platforms can facilitate online public opinion and remain as places where diverse ideologies and voices relating to current social and political affairs can be found (Stockmann and Luo 2017;Shi-Kupfer et al. 2017;Shi-Kupfer and Mao 2020). In some cases, these platforms help to form important networks and spaces for citizens to collectively voice their interests and concerns, and to facilitate activism (Shan and Tang 2017;Mao 2020;Gleiss 2015;Lei 2018). There is ample evidence that opinions and concerns voiced on social media can set the agenda for traditional (although not party-state) media and can gain the recognition and response from the government (Wang 2018b;Luo 2014). Moreover, online criticism has also frequently exposed failings of commercial actors and exerted pressure on the government to hold them accountable. For example, in 2018 the Chinese AI startup Megvii was widely criticized online for a demo video of its facial and emotion surveillance system designed for schools, which was posted on the social media platform Weibo. In the wake of this backlash, the Ministry of Education issued "Opinions on Guiding and Standardizing the Orderly and Healthy Development of Mobile Internet Applications in Education" and stated the intention of regulating AI applications for educational use (Ministry of Education of the People's Republic of China 2019; Wu 2019).
Social media platforms are also becoming increasingly important for science communication in China. The government encouraged the scientific community to use social media as channels for disseminating scientific knowledge among the public and to encourage public engagement with science (Xinhua 2019; Dijkstra and Yin 2019). Partly validating this strategy, researchers found that information on social media can influence people's attitudes toward and support for controversial technologies such as genetically modified organisms (GMO) and AI (You 2019;Cui and Wu 2021).
The role of social media platforms in China's political and science communication make them a valuable source for analyzing public discourse on AI. However, there are multiple platforms with different technical attributes and user bases which perform these roles in different ways and to a different extent. Our study focuses on two platforms-WeChat and Zhihu-because of their shared popularity and yet distinctively different communication styles, user bases and content foci.
WeChat, a "super app" that combines the function of Whatsapp and Facebook and even integrates other commercial and public services is the most popular social media platform in China. In 2020, WeChat's active monthly users in China numbered over 1.2 billion. It is equally popular among men and women, and its users cover all age groups (58.5% of the users are 30 or below. 41.5% of the users are above 30 (Ho 2021). The app's public account function, which we focused on in our research, allows individuals and organizations to publish articles to their subscribers with text, pictures, audios and videos. These articles may be short but can also be the length of an entire academic paper. There are around 8.5 million public accounts, many of them having several million subscribers and they cover a variety of topics (Li 2019c). Some of these account owners are prominent figures and organizations in various fields offline. The communication style and users of WeChat public accounts mostly represent the cultural elites who use the platform to promote new concepts and to institutionalize idea.
The second platform, Zhihu, is a Chinese Q&A online forum similar to Quora. As of 2019 it had 220 million registered users and more than 130 million user-generated answers (Smith 2021). Established with the slogan "Let people better share knowledge, experience and insights and find their own answers," the market strength of the platform is its peer-to-peer knowledge network. It is deemed as the best Chinese social media platform to get professional expert insights on various topics (Graziani 2018). Although most Zhihu users do not give details of their personal identity, professions and fields of expertise are commonly used as identity markers. Analysis of Zhihu's user base has identified that most users are professionals in the tech industry, students of computer sciences, and other highly educated groups (Cxb168 2019;Xueshenkeji 2020). The gender imbalance in these fields also seems to be reflected by the dominant presence of male users (53% male, 33% female and 14% unknown) (Zhang 2019). Content from Zhihu mainly represents individual opinions from the frontier of technological development in China.
To sum up, content on social media can provide valuable material for analyzing online public discourse on the social and ethical implications of AI, whether it be part of the official opinion-guiding efforts, Chinese cultural elites' genuine intellectual investigations, or societal discussions. Our choices of two distinctive social media platforms, WeChat and Zhihu, largely cover the potentially different dimensions of public discourse.
Methods
To understand the discussions about AI ethics on Chinese social media in a structured and contextualized way, we adopted qualitative content analysis methods in our study. As a social studies research technique commonly used in communication studies, content analysis applies systematic and rule-guided procedures to the analysis of large amounts of texts such as media content. This includes developing a category system instrumental in understanding the texts, using the system to encode texts, and generating various analyses such as quantitative category occurrences and qualitative contexts. The main goal of this technique is to reduce text materials in such a way that the essential contents remain and prominent themes emerge (Mayring 2014). This methodology has been frequently used to understand the meanings a group of people or a culture attribute to an issue within a specific context, taking that meaning can be derived from content of communication, which supports inference about its social context (Krippendorff 1989). It is therefore the most suitable for our study which seeks to navigate through emerging online discussions in China on AI ethics.
Data collection
WeChat's public application programming interface (API) is notoriously difficult to access. To collect relevant posts from its public accounts, we used the proven method of webscraping through the third-party search engine Sogou, which provides access to historical data of WeChat public accounts (You et al. 2018;Zeng et al. 2020). When we started our research, the search engine Sogou did not support site search for Zhihu posts, so we used another Chinese search engine, Baidu, to collect Zhihu data. However, in a later phase when we were updating our database, Sogou added this function. Once it was technically possible, we used Sogou to doublecheck the results and added the new results that had not been yielded by Baidu.
The keyword combinations "AI Ethics" and "AI Morality" 2 were used separately to retrieve data from the time range set from the earliest post up to and including December 31, 2020. The results from WeChat included a total of 1173 posts with the first post on record from November 24, 2014. After removing duplicates, event/book promotions, articles that only contain audio, video, and image content and, for the purpose of our analysis, excluding non-opinion pieces such as purely factual news reports, 395 posts were qualified for further analysis. The results from Zhihu included 1123 3 threads (one question with three answers) with the first on record from November 28, 2014. 4 After removing duplicates, threads with less than three answers, question posts composed of only one sentence, and answers which are too short or irrelevant, 372 posts (124 questions with 3 answers each) were included in the final analysis.
Data analysis
To perform content analysis on the dataset we collected, we first developed a coding scheme based on a literature review of global discussions on AI ethics. We also built upon the concept of framing in media and communication studies, which is defined as selecting "some aspects of a perceived reality" to enhance their salience "in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation" (Entman 1993, p. 53). According to the coding scheme, we treated an article on WeChat and a question or answer on Zhihu as a unit of analysis and coded the following variables: author type, 5 foreign sources(s) of reference, context(s) of discussion, risk/opportunity/neutral assessment/not applicable, reasons given for above assessment, 6 and recommendation(s). 7 The two authors independently conducted manual coding on the data from each platform. To ensure intercoder reliability, we used 20 random listings from each platform as a sample to code, compared the results with each other, and reached an agreement. Then the two authors independently conducted manual coding on data collected from each of the two platforms, respectively.
Limitations
We acknowledge several limitations of our research. First, our data collection was highly reliant on the search engines we used, and we cannot assess how the underlying algorithms could have influenced the search results. To avoid personalized results, we conducted a search for WeChat articles on Sogou with a newly registered account without any previous data usage. The searches for Zhihu posts were conducted on Baidu and later on Sogou without a personal Zhihu login. However, this does not ensure access to the complete and unbiased pool of relevant data. Second, although the user demographics of WeChat and Zhihu are diverse, it should be noted that the data from two specific social media platforms can only include the discourse and views of those who have access to digital technologies and use them. Third, our use of the keyword "AI" rather than specific terms such as "facial recognition," or "content recommendation algorithms" inevitably missed some of the discussions on specific AI applications, which is worthy of further research.
Types of authors
On WeChat, 47.1% of the analyzed articles' authors were researchers in the field of AI ethics and governance. The most prolific include Duan Weiwen, director for the Department of Philosophy of Science and Technology in the Institute of Philosophy at the Chinese Academy of Social Sciences, Zeng Yi, director for the Research Center on AI Ethics and Safety, Beijing Academy of Artificial Intelligence, and Chen Xiaoping, director of Robotics Lab at University of Science and Technology of China. The public accounts of media outlets, especially those focusing on the tech industry such as Leiphone.com and Jiqizhixin.com, published a further 18.2% of the articles. Individuals and 5 We defined the categories for the type of author based on the availability of information, and the categories based on professional background. 'Academia', 'Media', 'Industry', 'Government/Affiliated organizations' proved to be the most workable criteria. If the author's identity was given in a post, we coded the author category according to the author's profession; if not, we coded it according to the type of public account. For example, if a scholar's text was published in a media account, we coded it under the category of 'Academia.' Users classified as 'General public' are either anonymous, unverified authors, or authors with either no description or a prosaic one. 6 Due to limited capacity, we coded a maximum of three reasons, both concerns and hopes, for each unit of analysis (a post in the case of WeChat and an answer in the case of Zhihu). 7 Due to limited capacity, we coded a maximum of three recommendations. 3 760 entries from Baidu and an additional 363 entries from Sogou were later added to the database. 4 The Zhihu platform puts the answer with the most replies or likes at the top of the page, ranking them first below the question. The date of the question is not displayed. The date of this first reply post is then captured via Baidu and Sogou as the first entry (consisting of question and answers) in the search results, while the same question might appear with the date of another answer further below in the search results. So it may be that the initial question related to the topic was posted on another date within the applied time frame. Then this post would be filed under the earliest date and duplicates would be removed from the overall sample. organizations in the tech industry accounted for 18% of the articles. Among them, Tencent Research Institute, the social research arm of the Chinese internet giant, is the most prolific account among all types of authors. Since 2017 it has published numerous articles about potential ethical issues associated with AI, introducing global and especially European ethics principles and regulations aimed at the Chinese public. Among the remaining articles, 6.3% were published by members of the general public, who were identified as such by the account type 'individual' on the accounts' 'about' pages. 6.1% were published by government affiliated organizations and party-state media. The remaining 4.3% were published by organizations categorized as 'Others.' These included local associations promoting science and technology, religious associations and non-governmental organizations (NGOs). Notably, only one Chinese environmental NGO and no NGO with specific focus on AI governance was identified in the discussions on WeChat (Fig. 1).
On Zhihu, authors from the general public accounted for 65.3% of the posts. The second largest group of contributors (16.1%) were authors with a self-stated industry background. They were mostly individuals, but featured were also institutional accounts of companies broadly related to the field of AI such as Microsoft Asia, Amazon, iFlytek. Like on WeChat, Tencent Research Institute was one of the most active Chinese institutional accounts here, offering substantial analysis and links to further studies. Users with academic backgrounds accounted for 14% of the posts. Media accounts (3.2%) and authors with government or government-affiliated background (0.8%) were nearly absentat least based on the publicly displayed information.
Analysis showed that a wide range of stakeholders participated in the AI ethics discussions on both platforms. As WeChat public accounts and Zhihu were chosen for this study due to the different characteristics of their user base, our analysis of the type of authors proved that the prominent voices on the two platforms were indeed quite different. The majority of authors on WeChat were from academia, the media or industry, representing cultural elites, while on Zhihu most were members of the general public and individual IT professionals. This difference provided the background for analyzing how and why opinions expressed on the two platforms differ, which we elaborate on in the subsequent sections.
Foreign references
To discover how public discourse on AI ethics in China relates to the global discourse, we coded the sources of foreign references when they were used. On WeChat, 30.1% of the authors cited one or more sources from the US, 8.4% from Europe, and 28.9% from both Europe and the USA, while 18% cited sources from other parts of the world. The most frequently cited set of principles is the "Three Laws of Robotics" by the science fiction writer Isaac Asimov. The EU commission's "Ethics guidelines for trustworthy AI," Microsoft's "Responsible AI principles," and the IEEE's Ethically Aligned Design were also frequently cited. Notably, in multiple academic papers posted on WeChat, there were lengthy interpretations of the EU's "Ethics guidelines for trustworthy AI" and even discussions of its potential implementation in specific fields such as education and healthcare in China (Deng and Li 2020;Hu et al. 2020). Other international references included AI principles from Canada, Japan, Korea, Australia and the G20. Individual scholars' works such as Nick Bostrom's Superintelligence: Paths, Dangers, Strategies, Yuval Harari's Homo Deus: A Brief History of Tomorrow, Herbert Marcuse's One-Dimensional Man were also frequently cited. Some scholars were referred to for their general theories or entire body of work, for example, Immanuel Kant, Bruno Latour, and Luciano Floridi. Interviews with scholars such as Wendell Wallach and Alan Winfield also appeared in some articles (Fig. 2).
Unlike authors on WeChat, Zhihu users seldom refer to foreign sources. Academic users counted as exceptions for providing graphics or links to articles in English. However, few cited specific sources or provided bibliographic references. The few exceptions included John Brockman's 25 possible ways to look at AI, Lewis Mumford's The Myth of the Machine, and The Early History of Data Networks by Gerard J. Holzmann and Bjoern Pehrson. Github content, scientific magazines such as Nature or Science and foreign news websites were also sometimes cited. More frequently, users made reference to science fiction such as Matrix, Upgrade and the series Black Mirror. These references were mostly used to address concerns related to AI, especially concerns over AI's potential threat to humanity, which we elaborate on in Sect. 4.4.4.
To sum up, the discussions on AI ethics in China have been to some extent shaped by international deliberations. In line with our findings that the discussions were mainly driven by cultural elites on WeChat and the general public and individual tech professionals on Zhihu, the authors on WeChat also displayed more interests in and familiarity with international developments and debates, when compared to those on Zhihu. Moreover, the reference analysis demonstrated that deliberations in the USA and Europe have exerted great influence on the research of Chinese scholars and even tech companies, at least at the discursive level. Chinese researchers have extensively explored Western philosophies concerning science, technology and their social implications. They have, in turn, informed the Chinese public about global initiatives concerning the governance of AI. Although Zhihu users appeared to be less receptive to the international high-level deliberations on AI ethics and governance, they were tuned in to the imaginaries of AI created in science fiction produced in the US. These parallel engagements at the two levels -cultural elites and general publicwith other parts of the world, although predominantly the Western world, provide a positive outlook for future societal dialogues on this topic at the international level.
Context of discussion
We coded the context(s) in which the WeChat articles and Zhihu posts discussed the ethical and social implications of AI. 45.1% of the articles on WeChat and 65.3% of the answers on Zhihu discussed the topic at an abstract level, approaching AI as a broad field and interrogating some fundamental philosophical questions. These discussions, which were coded under the 'General/Philosophy' category, can be summarized as covering, but not limited to the following aspects ( Fig. 3): 1. The nature of technology itself (e.g., whether it can possess or develop morality and the ability to distinguish right from wrong). 2. The nature of humanity (e.g., consciousness, intelligence, creativity, senses, emotion) and what the use of AI means for this nature. 3. The relationship between humans and machines (e.g., love, competition), how they should interact and their responsibilities toward each other (e.g. no harm, mutual respect).
4. The changes in human societies caused by AI technologies (e.g., intelligence revolution or survival of the fittest), and the way in which people live and associate with each other.
In the discussions about AI applications in specific fields, on WeChat most were focused on healthcare (11.6%). Research on this topic seemed to have gained momentum in 2020, with a series of academic papers analyzing the ethical issues around the use of AI in assisted or automated diagnosis or prescription, AI medical monitoring wearables, AIbased health counseling apps, etc. (e.g., , Zhou 2020, Chen and Zhang 2020Zhou et al. 2020). The second most frequently addressed field of application was autonomous vehicles, which accounted for 9.1% of the discussions. Part of the reason was the use of the 'trolley problem' as a classical example to introduce the topic of AI ethics. 8.7% of the discussions on WeChat focused on AI in the media and other entertainment applications. Deepfake apps, Cambridge Analytica, and chat bots such as Microsoft's Tay have triggered much media coverage on the potential harmful impacts of AI technologies. There were also a few articles investigating the ethical implications of using AI in journalism (e.g., Li 2019a; Zhao 2019b). The remaining discussions addressed the use of AI in the judicial system (5.2%), autonomous weapons (5%), education (4.4%), the impact of AI on labor (4%), and miscellaneous other applications such as in financial and social services (6.9%).
On Zhihu, most posts discussed the ethical issues around AI in relation to labor (16.4%), such as whether or not AI Fig. 3 Context of discussion on WeChat and Zhihu will lead to more unemployment and/or human idleness. Negative consequences for individuals or for a specific group of people were frequently weighted against the overall "progress of society", as many authors called it, which was generally seen more positively, and the "liberation of the people", mentioned in several posts, from tedious labor to focus on more creative activities or even leisure in the long term. All other categories were only mentioned by a minority of users. 'Autonomous driving' (5.4%) was mostly addressed in relation to the 'trolley problem' or the latest developments by Chinese companies like Baidu in this technology. AI in the context of "media/entertainment" (4%) was mostly mentioned in relation to science fiction, chat bots such as Microsoft's Tay (with regard to racist comments) and Little Ice (pornographic conversations).
Comparison of discussions on WeChat and Zhihu, despite the shared focus on general philosophical questions regarding human-machine relations, uncovered a clear difference with regards to AI applications across specific sectors. On WeChat, a recent surge of academic papers about AI in healthcare led this field to being the most discussed. However, the impact of AI on labor was paid the least attention to. Conversely, labor issues around AI attracted the most attention on Zhihu, and AI in healthcare the least. Zhihu general public users' lack of interest in fields seemingly less relevant for them was demonstrated by the near absence of discussion on the use of AI in military scenarios.
General assessment in terms of opportunity, risk and neutrality
The dataset for this research consisted of an analysis of opinion-based writing, meaning that much of the data contained assessments relating to the risks and/or opportunities that AI technologies posed to individuals as well as broader society. We coded the articles and posts mentioning risks and opportunities as 'Neutral' and those without such an assessment as 'Not applicable.' This section gives a quantitative overview of the presence of these assessments on both platforms. The detailed reasons given for these assessments are further analyzed in the Sects. 4.4.2-4.4.4. On WeChat, most (37.7%) of the articles assessed AI technologies' impact as neutral, acknowledging both the negatives and the positives. A significant number of the articles (28.1%) made no assessment at all, which were often academic papers that directly launched into an analysis of philosophical questions. 27.1% of the articles made the assessment that AI will bring risks, stressing AI technologies' uncertain or unintended consequences. For example, Duan Weiwen, a researcher from the Chinese Academy of Social Sciences and the top expert on AI ethics in China, argued in an article that "The increasing application of AI and automated decision making systems including robots is not only an open-ended technological innovation with uncertain consequences, but also a social ethical experiment with a far-reaching impact in the history of human civilization" (Wen 2018). Only 7.1% of the articles assessed AI as generally beneficial.
On Zhihu, a slightly higher percentage of users (38.7%) focused on the risks of AI when compared to those who offered neutral assessments (35.5%). Risks were linked both to unintended consequences and non-foreseeable technological developments leading to AI acting in a "hostile" and "evil" way, or AI "overtaking" humans (He 2017). When compared to WeChat users, more (19.4%) of the Zhihu users made positive assessments of AI's potential impacts. However, on Zhihu there were significantly fewer users (6.5%) who did not make any assessment at all.
Neutral perceptions
On both platforms, about one-third of the authors had neutral perceptions of AI's impact on society at large or individuals. Authors on WeChat typically held the narrative that AI, similarly to other technologies, is a double-edged sword and its impacts would depend on how it would be used. This was particularly common among the academic, industry and media accounts, who often made a neutral assessment of AI in general at the beginning of the article, then discussed the risks and opportunities associated with the technology at length later on. For example, Tan Tieniu, a professor at the Institute of Automation of the Chinese Academy of Sciences, gave a keynote titled "Artificial Intelligence: Angels or Demons" which was posted by the WeChat public account "Towards Intelligence Forum". In the keynote, he argued that, High-tech itself makes no distinction between angels and demons, and so is AI. So, is the double-edged sword of AI an angel or a devil? It depends on human beings. We should plan ahead and form a joint force to ensure the full effect of AI in benefiting mankind! (Tan 2018).
In a WeChat article about its AI principles issued in 2018, Tencent Research Institute also claimed that "The birth of a new technology itself is not good or bad, but it is our responsibility to ensure that these technologies can become 'good technologies' through ethical norms, laws and various institutional designs" (Si 2018).
The neutral perceptions on Zhihu were mostly backed by the belief that humankind can control the direction in which AI will develop. This was often based on a distinction between weak and strong/super AI, which was commonly emphasized by Zhihu users who were familiar with the technologies. Some Zhihu users also argued that AI's benefits and risks cannot be easily distinguished: The first thing to be sure is that AI is absolutely beneficial to humans, but beneficial to humans as a whole does not mean that it is beneficial to every individual. The increase in automation in various industries will concentrate wealth in the hands of fewer and fewer people. Robots take jobs away from people, and although it improves overall productivity, do those people who have their jobs taken away have their real income increased? (Ma 2020) Notably, some Zhihu users also saw this technological development as inevitable, expressing deterministic and transhumanist arguments. For example, one Zhihu user claimed that he was not afraid of AI because "what should come is always coming…most likely, if not the only, way for human beings to survive on the light-year scale and 100,000-year scale as civilization is to move from carbon-based to silicon-based. This [flesh-and-blood] body is still too fragile when facing certain situations" (Zhaotangmixiang 2014).
However, most WeChat articles and Zhihu posts' neutral perceptions were derived from a combination of concerns and hopes associated with AI; a detailed analysis substantiating this assessment is in Sects. 4.4.3, which focuses on hopes and 4.4.4, which focuses on concerns.
Opportunities
On both platforms, only a small number of authors expressed their hopes for the opportunities that AI can bring to society or to individuals. Most of the hopes that were expressed fell into the following three categories: AI provides social benefits, humans and AI complement each other, and AI helps human society to revise and evolve.
In 7% of the WeChat articles and 11.6% of Zhihu posts authors expressed hopes about AI's potential to provide social benefits. These included increasing productivity and efficiency, eradicating poverty, improving education, providing medical care, promoting sustainable development, relieving humans of tedious tasks, alleviating the shortage of labor in aging societies and increasing safety in autonomous cars and even weapons. Some also used specific AI applications to illustrate their points: for example, a system developed to prevent suicide by screening the Chinese social media platform Weibo for posts that show suicidal tendencies and dispatching rapid intervention rescue teams (Li 2020). Zhihu users also mentioned the benefits of algorithms in finding a suitable partner through assessing and matching various personality traits, and AI robots that cater to sexual needs (Winterhouse 2016).
2% of the articles on WeChat and 11.3% posts on Zhihu claimed that AI is beneficial because humans and AI complement each other. This argument typically holds that AI and humans can perform different tasks to different levels: humans are better at tasks involving creativity or emotions, and AI are better at more mechanical and repetitive work. One user used the field of education as an example: "I think AI can help improve students' learning efficiency, while teachers' repetitive teaching work will be gradually replaced, but there is no way for AI to really replace teach-ers…Education is education precisely because it is the awakening of personality and mind…And these things can only be achieved in the process of human interaction. Robots are certainly not able to transmit this kind of knowledge" (Yunduokecheng 2018).
1.4% of the articles on WeChat and 3.5% posts on Zhihu argued that AI can help human societies evolve by revising past rules and norms. This argument holds that the ethical norms in human society are not constant but need to be adapted to new environments as technology develops. For example, in an article published by Tencent Research Institute's public account on WeChat, the author argued: Smart assistants may become a powerful boost to our efforts to promote gender equality. We hope that the new technology will lead the whole society in a more equal and pluralistic direction, rather than cementing our existing biases through more preference algorithms (Wang 2018b).
Many authors who held this type of opinion expressed belief in some kind of "new civilization" which would replace the older, inferior one (Fengqichanglin 2017). Lan Jiang, a philosophy professor at Nanjing University, analyzed brain-computer interface (BCI) technology and claimed in an article published by the party-state media outlet Guangming Theory that, "BCI will constitute a kind of transhumanism. Perhaps what we will see is not necessarily an ethical disaster, but a new kind of hope for ethics" (Lan 2019).
Concerns
A significant share of the discussions on both platforms was devoted to concerns associated with the use of AI technologies. They largely fell into the following nine categories shown in Fig. 4. The following analysis focuses on the most frequent mentioned topics on the respective platforms.
9.1% of the concerns addressed on Zhihu and 10.3% on WeChat fell into the category 'concerns for humanity.' On the one hand, this included concerns inspired by science fiction scenarios about the threats AI could pose to the human race if AI were to surpass humans in terms of intelligence and capability (e.g., Luo 2015). This was particularly common among Zhihu users in the discussions prior to 2018. On the other hand, it included concerns regarding AI's impacts on human nature, for example, that AI could become the "breakpoint" of human relations, leading to isolation, that it could replace or "technify" human activities, senses, body parts, and eventually humanity (Representative of Ethics Course 2020; Yang 2020). This type of concern was mostly addressed in the discussions on the general/philosophical level but sometimes in association with the use of AI in specific fields such as healthcare and education. Many authors expressed concern over the loss of the "human touch," compassion and empathy in these scenarios, qualities which are deemed essential in caring and learning.
'Responsibility concerns' were the most frequently addressed concerns (18.1%) on WeChat. Discussions that fell under this category included those about AI systems' moral standing (if AI can be regarded as having moral agency), their legal statuses/personalities, the social responsibility and liability when AI systems fail, and the intellectual property rights around AI. 8 However, this concern was addressed by a small minority (4.1%) of Zhihu users, mostly linked to questions of liability with regard to autonomous driving.
16.1% of the concerns on WeChat addressed the potential bias embedded in AI systems or discriminative uses. However, most of the examples used to illustrate the harmful impacts of bias in AI systems-passed on from humans or as a result of a poor training dataset-are from abroad. Examples included Amazon's recruiting algorithms favoring male applicants over females, and Google's search engine that labels people with dark skin tones as apes (Informatization Collaborative Innovation Committee 2017). Examples from the Chinese local context for discriminative use of AI were only mentioned in the case of some commercial platforms' use of algorithms for discriminatory pricing. On Zhihu, these concerns were rarely addressed (4.7%), and mostly in relation to Microsoft's chatbots Tay and Little Ice. Users shared screenshots of conversations they had with those bots and their racist or sexist comments. 7.4% of the concerns on WeChat and 9.4% on Zhihu addressed issues relating to human autonomy and agency. Many authors on WeChat were worried about over-dependence on technologies, for example, if medical workers were to increasingly rely on AI for diagnosis and prescription it could lead to them losing relevant skills in the long run. Many authors also discussed the manipulative power of algorithms based on their access to users' behavior data and personal data. Cambridge Analytica was often cited as an example of this risk. On Zhihu, these concerns were Fig. 4 Opportunity/risk/neutral assessment of AI on WeChat and Zhihu frequently based on the possible emergence of strong AI with self-consciousness, which could manipulate humans.
Privacy was the second most frequently addressed concern on WeChat (16.5%), as the development of AI depends on large-scale collection of individuals' personal and behavioral data. Notably, some scholars expressed pessimism towards the protection of privacy in relation to the introduction of AI. Liu Yibo, a scholar in the field of social governance, acknowledges that the boundaries of privacy will continue to shrink with the development of AI, arguing that in the future privacy will only exist at a conceptual level and humankind will enter the "post-privacy era" (Liu 2018). By contrast, only 2.9% of concerns expressed on Zhihu were about privacy. 8% of the articles on WeChat addressed concerns over AI's impacts on employment. Despite being one of the least mentioned concerns on WeChat, this was the second most frequently mentioned concern on Zhihu (11.7%). While the view that the development of AI will inevitably lead to the elimination of jobs was prevalent, many authors disagreed as to how many new jobs might be created because of AI, and whether people should aim for less wage-based work and more space for personal development.
8.7% of the concerns on WeChat and 9.4% on Zhihu addressed issues regarding the increase of the digital divide and inequality. On WeChat, users mentioned inequality between social classes, regions, and industries arising from the uneven distribution of AI technologies. Du Junfei, a sociology professor, used two examples to illustrate his concerns: first, many senior citizens were excluded from social service because they were unable to use their health code during the COVID-19 pandemic; and second, the family of a 94-year-old woman had to carry her to the bank and maneuver her into an awkward position to use the facial recognition technology required by the banking services (Du 2020). Users on Zhihu stressed that AI is likely to increase societal inequality and even exploitation if AI remains in the hands of a few powerful actors for them to "squeeze the interests of the majority of people" (Yao 2019). Notably, Marxist theory appeared frequently in discussions on this platform. One user analyzed AI's power in eliminating human labor using Marxist ideology, arguing that this would lead to the loss of peoples' consumption power: "If capitalists still exist, then they will certainly prevent this from happening, specifically by demanding that…artificial intelligence will be limited to a level that is not too high or too low." Two possible outcomes of AI development therefore depend on the existence of "capitalists": But if capitalists do not exist, or are eliminated, this would result in the means of production being transferred to universal ownership, and the manufactured products would be sufficient to satisfy the needs of so many people. The management body could be elected or be run directly by artificial intelligence...This does not mean that people would have completely given up work, but only that it would be the end of wage labor. Similarly, as at this point work is no longer wage labor, and you are no longer working for someone, you are doing it because you really want to (Yuexiashouwangzhe 2019).
AI systems' technical complexity and perceived incomprehensibility-the 'black box' phenomenon-were the least mentioned concerns both on WeChat (5.6%) and on Zhihu (2.3%).
To summarize, the variety of concerns associated with the use of AI across social domains that were addressed on both platforms, and the different emphasis given to different issues, is noteworthy. On WeChat, where most authors were from academia, the media and industry, the majority of the concerns were related to individuals, such as responsibility, privacy, and bias. On Zhihu, where most of the authors were members of the general public, more emphasis was given to concerns at a society level, especially concerns for the future of humanity. This can be partly attributed to the general public's fascination with science fiction; however, their concerns over employment, inequality and autonomy, which were not as prominently featured in the discussions on WeChat, showed how different social groups have different priorities when considering the ethical issues around AI. Together with the opportunities and neutral perceptions that were also expressed on the two platforms, this analysis provides a more complex picture of AI's perception and development in China. The fervent pursuit of technological advancement by the state and companies, usually fueled on hopes for alleviating social issues, increasing efficiency and productivity, is accompanied by growing and diversifying concerns. One thing has become clear from these discussions: ensuing opportunities and risks associated with AI are not equally shared by the Chinese society, and adequate AI governance needs to be based on the negotiations of these hopes, concerns, and different interests.
Recommendations for AI governance
63.5% of the articles on WeChat did not stop short at assessing the risk and benefits that AI can bring, but also gave recommendations on how to mitigate the risks and foster the opportunities associated with it. A significantly smaller number of posts on Zhihu did this (14.2%). The different recommendations largely fell into the eight categories shown in Fig. 5. Like in Sect. 4.4., the most frequently mentioned categories will be elaborated on further.
Most of the recommendations (31.2%) on WeChat proposed the 'legal/standard approach', arguing that the state should be responsible for ensuring beneficial AI, and 1 3
Fig. 6
Types of recommendation offered on WeChat and Zhihu advocated for the establishment of regulations, laws, and standards. It also accounted for 15.1% of the recommendations from Zhihu authors (Fig. 6).
The 'personal responsibility approach,' advocating that individuals or humankind as a whole should be responsible for coping with the impacts of AI by improving themselves and learning to adapt to the future with AI, was the most recommended on Zhihu (28.3%) but least on WeChat (4.8%). A quote from a Zhihu author illustrated this type of argument well: [Humans need to] maintain a tolerant mindset of continuous learning and updated cognition...Learn a bit of programming language properly, know yourself and your enemy, so you can never lose a battle. 9 If you need to better understand AI, then learning how to manipulate it may be a necessary skill for the future (Bafangdejuzeng 2020).
Another Zhihu author attributed this responsibility to the whole of humankind: Accepting the coming of the age of AI means adjusting how we ourselves evolve to new needs that will arise from new changes. Change is eternal; what doesn't change is the moment (Qi 2019).
It is noteworthy that both the authors who advocated this approach echoed Chinese philosophical traditions. The first author cited Sun Tzu and the second reflected Daoist belief in eternal change. This approach, most often recommended by members of the general public on Zhihu, demonstrates that perceptions and responses towards AI can be culturally situated.
16.4% of recommendations on WeChat and 15.1% on Zhihu took the 'multi-stakeholder approach' which distributed the responsibility for the ethical development and use of AI across the shoulders of many. For example, scientists and engineers should be transparent and explain to the public the potential risks and uncertainties associated with the technologies. Managers of tech companies should be responsible for working with scientists, engineers and humanities scholars to conduct risk and ethical assessments. Policymakers should be responsible for science and technology policies, technological standards, regulations and laws that ensure the correct development trajectory. Humanities and social sciences scholars should research the social impacts of AI and formulate policy advice. The media should engage in promoting public understanding of AI. The public should also take responsibility for active participation in the social governance of AI. In addition, many authors stressed the importance of multidisciplinary research collaborations.
9.9% of recommendations on WeChat and 5.7% on Zhihu advocated for the 'self-regulation approach' which held that tech companies should be responsible for developing ethical AI. This was primarily because of the need to find a balance between ethics and innovation-to ensure technological development that brings societal benefits and economic growth while not stifling the innovation capabilities of enterprises. Suggested concrete measures for implementing self-regulation included the establishment of AI ethics committees composed of scientists, legal experts and project leaders, and engineers from within the companies. Besides committees, industry associations were sometimes advocated as the "external ethical gateway to control AI" (Chuangshi International Asset Management Group 2019). The Partnership on AI established by Google, Amazon. com, Microsoft Corporation, Facebook, IBM was also cited as a positive example of this. In addition, some authors on WeChat suggested the involvement of third-party review agencies to ensure compliance. Notably, but not surprisingly, this approach is mostly recommended by authors from the tech companies themselves (Tencent Research Institute 2018; Chuangshi International Asset Management Group 2019). It is the second least favored approach by the authors on Zhihu.
11% of the recommendations on WeChat and 17% on Zhihu supported the 'technical approach' which argued that ethical considerations should be part of the system design process. Ethical standards should be implemented by tangible techniques. For example, to tackle the 'black box' problem, technologies such as those employed by IBM Watson OpenScale, a tool to track and measure outcomes from AI models that helps ensure they are explicable and accurate, was cited for using technologies to solve technological shortcomings. Likewise, Zhihu users stressed the importance of 'trial and error' and a constant monitoring of mistakes/accidents in technology, either due to the technology itself or human behavior. In particular, one Zhihu user argued that ethical considerations should be involved in the process of designing AI, based on a specific order of ethical priority of a particular society (Brain 2015).
To summarize, among those who gave concrete recommendations for AI governance, there was again a clearly different emphasis between the cultural elites on WeChat and members of the general public on Zhihu. The former put more emphasis on the role of governments and the latter put more emphasis on people's own responsibility. While authors on both platforms shared opinions in recommending the multi-stakeholder approach and technical approach, Zhihu authors demonstrated less trust in letting companies regulate themselves and their interest in international collaborations.
Conclusion and implications
Our analysis has demonstrated that online discussions about the social and ethical implications of AI have been extremely lively and diverse in China. The discussions analyzed in this paper took place on different fora within a range of social groups, and had clearly varied foci and took different approaches. Despite similar assessments regarding the risks and opportunities associated with AI technologies, scholars, journalists, and leading tech corporations on the social media platform WeChat mainly addressed concerns over algorithmic bias, discriminative uses, responsibility and legal issues. They also demonstrated more familiarity with international scholarly and policy explorations. Members of the general public on the Zhihu platform, including those working in the tech industries, approached the issues around AI from an abstract level concerning the human collective in the long run. This was partly influenced by science fiction narratives produced in the US, but also concerns over job displacements caused by AI, rising inequality and the loss of autonomy. The cultural elites on WeChat were more vocal in terms of policy recommendations, stressing the responsibility of governments in regulations and leading multi-stakeholder participation. The recommendations given on Zhihu, however, emphasized the importance for individuals and human society to continue learning and adapting in order to remain competitive in the face of AI.
Above findings offer valuable ground for understanding the future trajectory of AI development in China. The diverse perceptions of AI in general and the wide range of concerns as identified in online discourse mean that the Chinese state's and Chinese companies' development of AI may continue, but not without addressing concerns raised by Chinese society. In fact, since 2020 multiple developments in China have been related to the concerns addressed in the above discussions. For example, the Chinese state started a crackdown on its domestic tech industry, launching investigations into how companies such as Alibaba and Tencent handled their users' data and engaged in monopolistic and discriminatory business practices. In August 2021, China's legislature passed the Personal Information Protection Law, called by some "the world's strictest data-privacy law" (Xiao 2021). Although these developments put no constraints on how AI can be used by the government, the salience of concerns addressed to issues such as job displacement and increasing inequality among the members of the general public as identified on Zhihu will put increasing pressure on the government to manage the social implications of technological innovation.
Several implications for global dialogue on AI governance and directions for further research can be drawn from the findings in this paper which demonstrated both the influence of international developments in AI governance on and cultural specificities within the Chinese domestic discourse. First, although scholars, journalists and industry researchers in China appeared to be familiar with academic literature, especially in Europe and the US, regarding AI ethics, as well as policy developments globally, the 'multinational approach' did not top the approaches recommended on WeChat, even less on Zhihu. While China has shown a tepid attitude towards participating in international AI governance initiatives, such as signing onto G20's non-binding AI principles and yet stayed away from Partnership in AI due to the fact that it consists of mostly Western actors, the lack of driving force in the Chinese public sphere is noteworthy and warrants further research. However, this does not mean that there is no prospect for international collaboration. Although observers outside China have anticipated that AI principles from China prioritize social responsibility and community relations over individual rights (Roberts et al. 2021), our research has demonstrated the importance of looking beyond official AI principles to the public discourse, which focused strongly on issues relevant to individuals such as algorithmic bias and privacy. Despite differences in philosophical traditions as well as political and economic priorities, the possibility of agreements on the practical implications of values such as security and privacy need to be examined through empirical evidence (Whittlestone et al. 2019). In addition, as demonstrated in this paper, the general public has expressed a sense of anxiety towards a future permeated with AI, in which their jobs and humanity could be threatened. This contradicts the widely held view that Chinese people demonstrate more positive attitudes toward digital technologies (e.g., The Digital Society Index 2018; Kostka 2019) and demonstrated similarities with the attitudes found in Western societies such as the UK ). The reasons for Chinese society's observed acceptance towards technologies may lie deeper in beliefs such as how to cope with changes and competition, as evident in some of the Zhihu authors' recommendations. More research is needed to understand the thinking that underpins Chinese people's attitudes towards AI technologies, especially in comparative perspectives. At a time of global competition and rivalry, especially between the Chinese party-state and liberal democracies, this research provides impetus for ongoing societal engagement, despite disagreements at government level.
Funding Open Access funding enabled and organized by Projekt DEAL. Max Planck Institute for the History of Science provided partial funding.
Data availability Available upon request.
Code availability Available upon request.
Conflict of interest
The author declares that they have no conflict of interest.
Ethical approval Not applicable.
Consent for publication Not applicable.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
|
2021-11-18T05:15:15.419Z
|
2021-11-16T00:00:00.000
|
{
"year": 2021,
"sha1": "e23fb4a268ece6fac82b49b3806040b81e0bd649",
"oa_license": "CCBY",
"oa_url": "https://link.springer.com/content/pdf/10.1007/s00146-021-01309-7.pdf",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "e23fb4a268ece6fac82b49b3806040b81e0bd649",
"s2fieldsofstudy": [
"Political Science"
],
"extfieldsofstudy": []
}
|
253387220
|
pes2o/s2orc
|
v3-fos-license
|
Assessment of extraneous water inflow in separate sewerage system by different quantitative methods
Extraneous water that inflow to the sewage system is basically divided into two streams—accidental water (mainly rainwater) and infiltration water. The aim of the research was to assess the amount of extraneous water inflow to the considered system. Five different quantitative approaches were applied. Three well-known methods were used: the triangle method, the minimum night flow method, and the moving minimum method. The annual balance of water consumption and sewage supply to the wastewater treatment plant were calculated. Also, some analysis of sewage discharge during wet and dry weather was carried out. The study covered data from 6 years from 2014 to 2019. It was established that the main source of extraneous water was infiltration, because three methods which concern both streams (triangle method, minimum night flow, variability in wet and dry weather) confirm the conclusion. Merely the moving minimum method results differ from the others. In this investigation, accidental water (basically rainwater inflow) poses a significantly less share in the total volume of sewage compared to infiltration water. The total amount of extraneous water was estimated as in the range from 38 to 53% of annual sewage supply to wastewater treatment plant, depending on the year. Share of infiltration and accidental water is changing in different methods. Share of infiltration was in a range between 18 and 68%, depending on the year and the method used. Share of accidental water was in a range between 7 and 22%.
Introduction
Besides water used by inhabitants, sewage inflow to the wastewater treatment plant (WWTP) consists of some extraneous water (or external). This water input does not come directly from water users. External water in separated sewerage systems included two main fractions. Basically, these are infiltration inflow, through leaky pipes and rainwater inflow (Karpf and Krebs 2005). Rainwater-induced flows could penetrate the sewage system by the illicit connection of drains from private proprieties, misconnection of drains from gullies, misconnection of storm sewers, and entry of surface water through manhole covers (Machado et al. 2007). That is why in literature, rainwater input in the sanitary sewage system is calling accidental water. This term is also used in this paper. Regardless of its origin, extraneous water causes increasing hydraulic loads and consequently costs for pumping and treatment (Dimova et al. 2015).
Permeated groundwater in the sewerage system is nonpolluted or a little polluted (Schultz et al. 2005). Its input is essentially related to the periodicity of the groundwater level (Wittenberg and Brombach 2002). In German sewers, wastewater flow contains up to 40% of groundwater (Wittenberg and Brombach 2002). A case study of the system in Brussels shows a considerable seasonal variation in infiltration inflow (de Ville et al. 2017). It ranged from 15% (in summer) to 45% (in winter) of dry weather flow. In Berkhamsted (UK), chemical oxygen demand (COD) analysis results in 37.4% infiltration inflow (Brian and Bertrand-Krajewski 2010). The same method was used in Mueva and Lockwitz (Germany) and results in 46% and 57% of infiltration inflow. Hydraulic overloading due to infiltrated volumes can reach up to and even exceed 100% of wastewater volume (Brian and Bertrand-Krajewski 2010). According to Peters et al. (2002), approximately 50% of stormwater is discharged into the sanitary sewers and transported to WWTP. In another 1 3 278 Page 2 of 12 study (Weiss et al. 2002) during 4 years of investigation in 34 WWTP, 70% of inflow was not domestic sewage. An investigation carried out in Trondheim indicated that during dry weather 46% of sewage discharge to WWTP was extraneous water (Beheshti and Saegrov 2018). A review of infiltration and inflow (I/I) in Norway, Sweden, Denmark and Finland showed also large amounts of extraneous water volumes (Jenssen Sola et al. 2018). In big Norwegian systems, the average value of I/I in 2017 was 66% and a small decrease in comparison to the 2009 year was noticed. For Danish districts, the amount of I/I was 30%, in Finlandabout 40% and in Sweden-46% (in 2016). Calculations made for the Suzhou district in China (for years [2014][2015][2016][2017] showed about 30% of external water share annually (Wang et al. 2019). All mentioned studies show a considerable fraction of external water addition in WWTP supply.
There are several methods to assess the infiltration and inflow into sewer systems. These methods can be divided into two groups: quantitative methods (for assessing the volume) and qualitative methods (for detecting the sources) (Beheshti et al. 2015). In the work (De Benedittis and Bertrand-Krajewski 2005), 15 traditional quantitative approaches were mentioned. They base on general data like water consumption and wastewater supply to WWTP. These data are analysed in different periods from daily to annual depending on a using method. An undoubted advantage of these approaches is that they are relatively easy to access data and well-known measuring methods. They are also easy to implement (de Ville et al. 2017). The disadvantages are mostly uncertainties due to assumptions (Kracht and Gujer 2005) specific to each method and also measurement uncertainties. Kracht et al. (2007) indicate that the minimum night flow method is oversimplified in the context of growing agglomerations and it might lead to erroneous results.
Materials and methods
The aim of the research was to attempt an assessment of extraneous water that inflowing to the sanitary sewer system. The investigation is based on data obtained from a wastewater treatment plant (WWTP) in a considered town. The daily precipitation data were received from the Polish Institute of Meteorology and Water Management (Historical Meteorological Data from Poland Area n.d.).
Five different methods and approaches were considered: • Annual water consumption and sewage discharge balance • Triangle method • Minimum night flow method • Moving minimum method • Analysis of sewage supply to WWTP in dry and wet weather The annual balance was the started point which indicated that there exists a problem of extraneous water. Other procedures were chosen mostly because they were well described and explained in available literature sources. Thus, it was possible to use them properly. Some approaches were rejected due to sufficient data collected.
Characteristics of the catchment and annual balance
The catchment is located in north-western Poland. The analysed system is a separate sewerage system. It serves one city and six smaller rural areas around. The total length of the main system is approximately 83 km. Additionally, there is about 139 km of sewer house connections. The total community served by the system is about 25,000 inhabitants. It was built in the 70 s of the twentieth century. Sewer pipes are made of PCV, stoneware, and concrete. Their diameters range from 0.2 to 0.8 m. Sewage from whole the system reaches a collective WWTP located in the largest city. This WWTP was modernized in 2012 r. Its actual capacity is 900 m 3 /h.
A considerable difference between sewage supply and water consumption was observed in the catchment. It led to the supposition that there is an inflow of extraneous water. The share of extraneous water was calculated by the formula (1): where S e -a share of extraneous water, % Q in -annual sewage inflow to WWTP, m 3 Q c -annual water consumption, m 3 The addition of extraneous water was calculated by the formula (2): where A e -the addition of extraneous water, % The results of the calculations are presented in Table 1.
Research concerns years 2014-2019. The estimated share of extraneous water was in a range between 38 and 53%. The estimated addition of extraneous water was between 62 and 115%. The maximum value of external water was occurred in the year (2017) with the highest precipitation (856 mm). On the other hand, in the year (2018) with the lowest precipitation (403 mm), there was the second-largest volume of extraneous water. The relation between the yearly sum of rainfall and sewage supply to WWTP is not clear in this catchment. Probably due to insufficient data. The collection of data consist of only 6 years, so it is not enough to provide precise conclusions.
Minimum night flow method
This method was described by Hager et al. (1985). It was also mention by De Benedittis and Bertrand-Krajewski (2005). This approach is based on the assumption that infiltration inflow is constant. During a dry period in hours 02:00 and 06:00, there occurs the minimum discharge. It is only composed of sanitary sewage and infiltration water. Night sanitary sewage can be calculated using indicators of flow per capita. Then infiltration inflow can be simply calculated by measured sewage supply to WWTP minus sanitary sewage inflow. Measurements should be made free of workdays to reduce inflow from service establishments or processing plants (Kaczor and Bugajski 2012). There are several various definitions of dry weather. Stier and Fisher (1995) provide guidelines that 1 day with precipitation less than 1 mm can be considered as dry weather. According to Kaczor and Bugajski (2012) observations, larger flows in sewerage occur 2 to 5 days after rainfall. Base on the daily inflow to WWTP and daily precipitation described in point 2.5, the above remark in this catchment was not confirmed. It was decided to consider three successive days with rainfall lower than 1 mm as a dry period according to the definition by Brian and Bertrand-Krajewski (2010).
To determine the night flow of strictly sanitary sewage, Fisher's (1990) indicators were applied. For the number of inhabitants between 5000 and 100,000, the indicator equals 0.5 dm 3 /s for each of 1000 inhabitants.
Triangle method
The triangle method is one of the quantitative methods. It is used to estimate both infiltration and accidental inflow. The application of this method was described by some authors (De Benedittis and Bertrand-Krajewski 2005;Wang et al. 2019;Weiss et al. 2002). In this method, daily sewage inflow to WWTP is ranked in ascending order. It is based on the assumption that sanitary sewage inflow is on average constant. Sanitary sewage inflow is calculated simply by the number of inhabitants times average potable water consumption. The next assumption is that sewage inflows larger than water consumption are firstly caused by infiltration inflow and secondly are caused by accidental inflow.
In this study, 6 years of daily measurements were elaborated.
Moving minimum method
The moving minimum method was described by Weiss et al. (2002). It is based on the assumption that the "sum of sanitary sewage plus infiltration flow at any day is equal to the minimum daily inflow during the past 21 days". This method can be applied to estimate the share of rainwater in the total amount of sewage. It has an advantage compared to the triangle method that rainy days and dry weather days are equally included. The moving minimum method was applied to whole the data from 6 years. The results are shown in point 3.4.
Short time variable of sewage supply to WWTP in dry and wet weather
Analysis of variable sewage supply during wet and dry weather was referred to Bugajski et al. (2017). They investigated changes in sewage inflow during dry and rainy periods. It was noticed that on specific daily periods with high precipitation, the share of accidental water in a total volume of sewage was up to 75%.
The data from the considered 6 years were divided into dry and wet periods. The criteria for beginning dry weather were 3 days after the last precipitation without rain until the next precipitation occurs. Days with rain and 3 days after were considered a wet period. Sometimes there were long periods up to 20 or 30 days without or with a single day with relatively small rain which does not cause larger flows to WWTP. These were also treated as dry weather if no influence on inflow to WWTP was observed.
Minimum night flow method
Attempting to choose appropriate periods to the minimum night flow method with relatively low variation was difficult. Finally, 29 night flows were considered in 2018 and 2019. The measurements were read from WWTP flow graphs, so the results might be subject to an error estimated as 10 m 3 /h. The outcomes are presented in Fig. 1. The average flow equalled 110.3 m 3 /h, and the standard deviation of the mean is 5.7 m 3 /h. The total measurement uncertainty is 11.5 m 3 /h. The maximum value of minimum night flow occurs on 5.03.2018, and it was 175 m 3 /h. It is about 59% more than the mean value. The minimum value was 60 m 3 /h (in 17.06.2018), and it is 46% less than the mean value. The maximum night flow was about 3 times higher than the minimum recorded value.
The estimated number of inhabitants is 25,000; thus, according to Fisher's approach (Fisher 1990), the minimum night flow of sanitary sewage equals about 45 m 3 /h. The infiltration inflow was calculated by the following formula: where q inf -infiltration inflow, m 3 /h.
(3) q inf = Q in − q s , m 3 h Q in -total measured inflow to WWTP, m 3 /h. q s -sanitary sewage inflow, m 3 /h. The infiltration inflow equals 65.3 ± 11.5 m 3 /h. Thus, the share of infiltration water in the total volume of sewage is in a range between 54 to 63% and 59% on average. The addition of infiltration water is in the range of 119% to 171% and 145% on average.
Triangle method
In each of the 6 years, the triangle method was applied. Wastewater inflow was calculated by total annual water use divided by 365 days. It represents the average daily flow (Q dav ). The green vertical line indicates the maximum infiltration inflow (Q maxi ). It is also an edge between days with and without precipitation. The yellow line means the border between infiltration and accidental water inflow. Results of plotting are illustrated in Figs. 2, 3, 4, 5, 6, 7.
The volumes of each stream were calculated by the sum of the area below the plotted lines. The sanitary sewage inflow is an area under the orange line; infiltration water volume-a surface between blue curve and orange line but from 0 to yellow line; accidental water-a surface between yellow and blue curve. Share and addition of infiltration water and accidental water for each year have been calculated using the formulas (1) and (2) with the numerator changed for infiltration or accidental water volume. The results are presented in Table 2.
The year with the largest amount of extraneous water was 2017. It reflects in triangle methods calculations. There is the largest share and addition for both infiltration water (37% and 80%) and accidental water (16% and 35%). According to this method, accidental water share is much lower than the infiltration water share every year. The volume of groundwater is from 2.3 (2017) to 3.86 (2018) times higher than accidental. Base on this method, it seems that infiltration is the main source of extraneous water in this catchment.
Moving minimum method
It was assumed, similarly to the triangle method, that sanitary flow is equal to mean water used by inhabitants. Results of plotting the moving minimum on a daily flow graph for each of 6 years are illustrated in Figs. 8,9,10,11,12,and 13. Infiltration water inflow for each day was calculated by the difference between moving minimum inflow minus sanitary sewage. Accidental water stream was calculated by the difference between daily inflow to WWTP minus moving minimum. Table 3 shows the calculation results.
The largest amount of infiltration water was in the year 2017 (33%), but the largest amount of accidental water was in the year 2015 (22%). The differences between infiltration water shares (18%-33%) in each year are larger than in accidental water (17%-22%). According to this method, infiltration inflow is generally more significant than accidental water, but differences between both streams are not very high (from 1 to 13 percentage points).
Variability of sewage supply to WWTP in dry and wet weather
All 6 years were considered, and the average daily supply to WWTP was calculated during dry and rainy periods every year. Table 4 shows the summary of calculations.
Years 2017 and 2018 which had maximum and minimum yearly precipitation have simultaneously the largest variability measured by standard deviation-1560 and 1547 m 3 /d. In other years the variability was significantly lower-from 758 to 909 m 3 /d. Also, a standard deviation of inflow in rainy weather is larger than in dry weather each year. The difference between dry and wet periods equals merely 7-14%. Figures 14 and 15 show the daily supply of WWTP and daily precipitation during, respectively, dry and rainy weather.
The year 2018 was the dries one in the considered period. The daily sewage supply shown in Fig. 14 in May 2018 was in a range of about 5500 to 7200 m 3 /d. Looking at days with precipitation, it does not seems to influence the sewage inflow. In the year 2017, there was the largest precipitation. The daily sewage supply shown in Fig. 15 was in a range between 5300 and 18,300 m 3 /d. In the first 2 weeks from 15.06 to 28.06, precipitation does not have a clear influence on sewage inflow. In the year 2017 mean daily sewage flow in dry weather was 6451 ± 677 and during the largest rain in 29.06 daily flow was about 18,322; thus, the share of accidental water equals 65%. It corresponds to the 75% that was obtained by Bugajski et al. (2017).
Based on Figs. 14, 15 and Table 4, it was found that only the largest rains have an explicit impact on the inflow to WWTP.
Summary of results
The most significant results are summarized in Table 5.
The annual balance reveals that there is a significant extraneous water inflow in the system. Applied methods confirm that observation, but there are some qualitative differences.
Results from the minimum night flow methods indicate that there is a significant addition of infiltration water. On the other hand, the data have a large variability which implies considerable uncertainties. There is also an issue that there are a few methods to calculate the sanitary sewage night flow. In this study, Fisher's approach was chosen in reference to another study (Kaczor and Bugajski 2012). However, results obtained by this method are compared with the analysed variability during wet and dry weather. It seems that most of the extraneous water poses the infiltration water because the difference was only between 7 and 14%. Merely the largest precipitation causes an explicit impact on the sewage supply. It is also visible in the triangle method.
In every year the infiltration inflow was considerably larger. The infiltration volume was about 3 times higher (except for the year 2017 with the highest precipitation) than the accidental water volume. There is a visible discrepancy between the moving minimum method and the others. In this method amount of infiltration water and accidental water is rather similar. Only in the years 2017 and 2018, there is a difference in favour of groundwater water.
Conclusions
In the system, there is a large amount of extraneous water in a range between 38 and 53% of total sewage volume. All of the methods except variability in wet and dry weather confirm that there was significant input of extraneous water. It was established in three methods (triangle method, minimum night flow method, and variability in wet and dry weather) that the most significant source of extraneous water is infiltration inflow. The share of infiltration water in these methods was in the range of 30% to 68%. There was a visible discrepancy between these three methods and the moving minimum method. In this approach, the streams were similar. Share of infiltration inflow equalled 17% to 33%. In the author's opinion, the volume of external water is large enough to be considered in designing or modifying sewage systems and WWTPs. Even if the catchment is specific and not all conclusions should be transferred to other systems, the literature review shows that the problem concerns many systems in all the world. The streams of extraneous water are variable in time and amount. This is an important issue in WWTP, where most objects and processes need constant operation conditions. To provide that properly selected retention tank is needed, especially in small WWTPs where fluctuations of inflow are generally more significant than in larger WWTPs.
|
2022-11-08T15:10:23.403Z
|
2022-11-08T00:00:00.000
|
{
"year": 2022,
"sha1": "226d766b7186af7141cb9e67a97880743f09a4a5",
"oa_license": "CCBY",
"oa_url": "https://link.springer.com/content/pdf/10.1007/s13201-022-01793-2.pdf",
"oa_status": "GOLD",
"pdf_src": "Springer",
"pdf_hash": "152be98038612374c8f3e42cf8f20f4d971eab66",
"s2fieldsofstudy": [
"Environmental Science"
],
"extfieldsofstudy": []
}
|
204062338
|
pes2o/s2orc
|
v3-fos-license
|
Initial experience with introducing national guidelines for CT- and MRI-based delineation of organs at risk in radiotherapy
A fundamental problem in radiotherapy is the variation of organ at risk (OAR) volumes. Here we present our initial experience in engaging a large Radiation Oncology (RO) community to agree on national guidelines for OAR delineations. Our project builds on associated standardization initiatives and invites professionals from all radiotherapy departments nationwide. Presently, one guideline (rectum) has successfully been agreed on by a majority vote. Reaching out to all relevant parties in a timely manner and motivating funding agencies to support the work represented early challenges. Population-based data and a scalable methodological approach are major strengths of the proposed strategy.
Introduction
A fundamental problem in radiotherapy (RT) is the variation in volumes for organs at risk (OAR) [1,2]. OARs are ideally contoured based on consensus guidelines and published delineation atlases, but opinions about which anatomical landmarks to use as references still diverge across different studies, countries, and continents [3]. In reality, OAR delineations are often left to the judgement of treating healthcare professionals resulting in significant volume variations between practitioners. It is well recognized that the real-life quality of data regarding delivered doses to OARs, therefore, can be of variable quality. This makes it hard to compare and relate dose and volume metrics to treatment outcomes both in clinical practice as well as in clinical studies [1,2,4] and can potentially even affect treatment decisions.
To date, multiple consensus guidelines for various body sites and OARs have been published [5][6][7][8][9][10]. There are also guidelines for the naming of OARs [11,12]. Educational efforts at larger international conferences, such as those offered by ESTRO and ASTRO, also take a large responsibility for teaching professionals how to contour and name OARs according to many of these, but the wider radiation oncology (RO) community lacks a strategy for how to implement such guidelines on a larger scale. Vaguely defined standard OAR delineation references and suboptimal compliance with these poses a particular problem for the growing number of "big data" efforts in RO. The lack of delineation standards for normal structures in this context was pointed out by several scientists at a recent workshop, where both the need for foundational work and large community efforts were acknowledged as means to counteract this [3,13].
This paper describes an approach to reach the agreement on OAR delineation guidelines on a national scale and the early experiences with executing such an activity. Our aim was to find a strategy involving the whole RO community nationally to reduce variability in clinically-used OAR volumes. The present work has a developmental character with the overall project being based on a multi-step plan inviting clinically active professionals from all national RT departments.
Materials and methods
This project started as a national initiative by members of the Swedish RT-register steering committee in Sweden in 2016 and is referred to as STRÅNG (acronym for the project in English: STRONG).
Overall project outline
We devised a multi-step plan, where key decision makers and RO professionals were identified and approached for project approval purposes. Applications for funding and ethical approval were submitted to relevant agencies/authorities in 2017. The latter was created as a multi-centre application to allow participation from all RT departments (n = 17, including both University and County Hospitals; granted by the Regional Ethical Agency in Gothenburg, Ref no.: Dnr 641-17/ T1115-18).
The core of the plan evolves around clinical residents in RO being invited to execute sub-projects for a limited number of clinically-relevant OARs as part of a compulsory learning objective during their clinical rotation. This is to be followed by a national comment round involving a panel of RO experts from all RT departments to reach a decision of acceptance or rejection of the proposed guideline based on a majority vote. As a final step, contour variability between clinicallyused OAR volumes and OAR volumes according to an accepted guideline will be quantified, as will potential effects on treatment decisions and associated toxicities. This last aspect includes ongoing work and will not be reported here.
Identifying OARs used in clinical practise
A survey was directed to all RT departments to collect baseline data and identify regularly-used OARs in clinical practise. The departments were asked to fill in a , and Other) and report to what extent and with what frequency these or other OARs were contoured. The pre-written number of entries was 65 with 12 structures listed for ≥2 body sites (Bladder, Bowel, BowelBag, BrainStem, BronchialPlexus, Esophagus, Eye, FemoralHead, Heart, Lens, Lung, and SpinalCord); possible answers were "always", "sometimes", and "never".
Planning and executing sub-projects in detail
Each sub-project will include approximately five OARs within a certain body site and will be conducted in a preparatory phase (all OARs) and in an evaluating phase (OARs where contour variability can be evaluated; Fig. 1).
Anatomical landmarks for selected OARs (identified as described above) will be suggested based on a structured literature search by participating clinical residents in RO guided by their local supervisors. Publicly available imaging datasets (CT and MRI) for visualisation of examples will be used when possible, but we expect that additional data from a small number of patients may also need to be collected. After approval of written and visual examples by a radiologist with expertise within the body site of interest, proposed guidelines will be sent to RO experts within the relevant anatomical site. These are then to be returned with either a recommendation of acceptance or suggestions of amendments of identified limitations. Agreement will be established as ≥50% of experts approving both text and image example(s) for a given guideline. Guidelines for each structure by this majority vote will finally be made publically available through RO-specific organizations and communicated back to the RT departments through educational workshops/meetings.
Current state of project
To date, we have arranged local and national meetings with residents in training at multiple occasions and we have participated at national RO meetings to inform about project outline and recruit local project members. In addition, discussions with representatives from the Radiation Safety Authority and the Regional Cancer Centers, two key stakeholders in the Swedish RO community, are ongoing to discuss possibilities to present the results of this project to healthcare professionals on a larger scale and through their established infrastructures. Both are official bodies of the Swedish government and work to protect citizens from the harmful effects of radiation and to assist in providing cancer care on equal terms to all patients, respectively. A first subproject with guidelines for rectum (anatomical landmarks for male pelvis as suggested by RTOG, ASTRO and ESTRO ACROP) [8,14] have been completed using CT and MRI data from the Gold-Atlas project [15].
OARs used in clinical practice
The survey was conducted in fall 2018 with responses from all 17 RT departments and revealed large local differences in terms of OARs contoured in clinical practice. Of the initial 65 OAR listings, 16/10/13 entries was reported as "always"/"sometimes"/"never" contoured in clinical routine by a majority of departments (≥9/17) The results for OARs reported as "always" are summarized in Table 1.
First sub-project on male pelvis and rectum volumes
For a first comment round to agree on the proposed guidelines, written information about anatomical landmarks for rectum (in Swedish) were presented to GI/GU experts at all RT departments on December 14, 2018, together with accordingly delineated volumes on one non-surgically-treated prostate cancer case. Instructions of how to respond to our request were included. Within two months' time, expert opinions from a first evaluation round resulted in a majority vote in favour of the proposed guideline (in total, 10 departments responded by April 9, 2019 when we decided to close the round). The expert evaluations revealed a substantial discrepancy primarily concerning length differences between the suggested guideline and the local clinical practice for delineation of rectum, with anatomical landmarks either being determined by the extent of the planning target volume or by referring to the whole anorectum.
Discussion
The importance of efforts aiming at standardization in RT including minimizing OAR volume variability cannot be overstated. Standardized delineation of treatment volumes and OARs is of highest importance to compare treatment planning, dosimetric results and, to some extent, clinical outcomes. To the best our knowledge, the STRÅNG initiative is first out to suggest a strategy of how to implement delineation guidelines exclusively dedicated to OARs at all treatment sites on a national scale. So far, we have established a multi-professional collaborative network and a strategy to involve one large RO community to reach national agreement on written-and visually-presented guidelines. Logistics for the proposed strategy has been tested successfully in a pilot project setting including male pelvis and rectum as a first OAR and the strategy is now being extended to involve RO residents in training. When a RO community adheres to guidelines for the naming of OARs, the first step towards achieving findable, accessible, interoperable, and reusable (FAIR) RT data [16] is taken. However, if the underlying volumes are created without associated guidelines for the delineation of OARs, the use of such collected data will be limited by effects of inconsistently reported dose. Our proposed strategy aims to obtain agreement on both aspects in one large community using accepted OAR nomenclature and OAR delineation guidelines and has the potential to provide national references both for automated image segmentation [17] and large-scale imaging processing algorithms in the future. In an era of "big data" efforts and where transition into AI-based delineations of both target and OARs are becoming a reality, data harmonizing initiatives such as STRÅNG, to assure consistency and to provide benchmark datasets for validation and testing of these new technologies, are urgently needed. Another area of importance in this context is to agree on which OARs to be contoured for different disease sites. A recent publication by Wright et al. [18] summarizes recommendations for this and includes various resources that define tissue delineations. In contrast to our approach here, they neither present associated anatomical land marks as such nor do they suggest how to implement their recommendations on a larger scale. In fact, a PubMed search on June 26, 2019, found no relevant publications on this subject using various combinations of the search terms (and equivalents thereof) "radiotherapy", "standardization", "normal tissue", "guideline", and "implementation strategy".
Internationally, there are several examples of strategies to reduce contour variability of target volumes, most of them being focused on clinical target volume (CTV) standardization for one particular treatment site in a specific study protocol [19][20][21][22]. For instance, the Belgian initiative PROCARE [21] took a national approach to improve outcome in rectal cancer patients between 2003 and 2014, including the creation of a central review platform to reduce variability for CTV delineation. All 25 national centers were invited to participate, and the full operation was successfully established between 20 centers in 2011 leading to increased uniformity in CTV delineations between them. Keeping in mind that our work only concerns the identification of guidelines for OARs, it is still interesting to note that there is a similarity between our voluntary multidisciplinary strategies with respect to identifying guidelines by both a literature review and the use of visual examples. Otherwise, the two strategies are fundamentally different with PRO-CARE providing feedback on uploaded cases to clinicians and our strategy having more of an educational aspect for residents in training. Knowing that the use of digital platforms such as the Anatom-e and eContour have as well been shown to increase the consistency of delineation processes [19,20,23] we also aim to take the educational aspect of STRÅNG further and have ongoing discussions about how to integrate our guidelines into such RO educational tools.
The proposed approach of STRÅNG has several strengths. If RO residents in training can be made aware of the importance of standardized OAR volumes when being introduced into clinical RT work, they can be expected to comply to such guidelines in the future. Acceptance among senior colleagues is also more likely if they are invited to Swedish RT-register Randomly select patients (per project) participate in the decision process of a proposed guideline. Furthermore, if expert groups in oncology can be convinced to integrate guidelines with other clinical recommendations on cancer diagnosis, treatment, and follow up, they can reach the wider RO community through already established infrastructures. In Sweden, the latter is partly coordinated by the Regional Cancer Centers. Challenges with the proposed strategy include difficulties to coordinate the number of parties involved and to reach out to all of them in a timely manner. The administrative efforts are in direct proportion to the scale of the project. It will take a lot of networking and coordination to build the multiprofessional network needed to reach the goal. For our conditions, we have identified that a full-time national coordinator will be required to orchestrate this in the long-term. Educational initiatives can be expected to add to this through the need for planning, organizing, and executing national meetings/workshops/on-site visits.
Structure contours before consensus guidelines (control)
The successful outcome of the project so far can partly be attributed to previous RT data standardization initiatives in Sweden where improvements on the national level have already been generally accepted by our community, the national naming convention [24] and the Medical Information Quality Archive (MIQA) [25]. The level of acceptance and commitment from the individuals and official bodies involved in the project was, therefore, high already from the beginning. How a similar strategy would fare in other countries will depend on their previous experience of data standardization initiatives and existing RO infrastructure. Finding financial support for infrastructural work can, however, be expected to be difficult in early stages of projects like these [3,13]; this is something we encountered here as well. Still, the advantage of having one unified approach to harmonize RT data per country is unquestionable if challenges of world-wide data sharing/ integration to optimally tailor treatments according to patient's needs are to be met.
In summary, we present the initial experiences with introducing national guidelines for CT-and MRI-based delineation of organs at risk in RT. The presented initiative aims to create a solid foundation for how to delineate OARs by taking a scientific approach when identifying and evaluating appropriate guidelines and by involving relevant parts of our RO community when searching for agreement and acceptance of them. National population-based data and scalable methodological approach including educational aspects for residents in RO are the major strengths of the project. Our hope is that this initiative will inspire more countries to devise strategies to obtain consistent OAR volumes in RT for reliable high-quality data in multicentre clinical trials, clinical follow up as well as quality assurance studies. Only when there is an agreement on which OARs to delineate, how to delineate them, and what to call them, each RO community has done their part in contributing towards FAIR RT data.
|
2019-09-26T09:06:21.337Z
|
2019-07-01T00:00:00.000
|
{
"year": 2019,
"sha1": "1bebd43a27b715e4efce88d97852baf8a2778e2e",
"oa_license": "CCBYNCND",
"oa_url": "http://phiro.science/article/S2405631619300338/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "32c7155d6d72162322ee0d51c48ace283f7888d7",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
132645825
|
pes2o/s2orc
|
v3-fos-license
|
Biofouling on Coated Carbon Steel in Cooling Water Cycles Using Brackish Seawater
Water cooling utilizing natural waters is typically used for cooling large industrial facilities such as power plants. The cooling water cycles are susceptible to biofouling and scaling, which may reduce heat transfer capacity and enhance corrosion. The performance of two fouling-release coatings combined with hypochlorite treatment were studied in a power plant utilizing brackish sea water from the Baltic Sea for cooling. The effect of hypochlorite as an antifouling biocide on material performance and species composition of microfouling formed on coated surfaces was studied during the summer and autumn. Microfouling on surfaces of the studied fouling-release coatings was intensive in the cooling water cycle during the warm summer months. As in most cases in a natural water environment the fouling consisted of both inorganic fouling and biofouling. Chlorination decreased the bacterial number on the surfaces by 10–1000 fold, but the efficacy depended on the coating. In addition to decreasing the bacterial number, the chlorination also changed the microbial species composition, forming the biofilm on the surfaces of two fouling-release coatings. TeknoTar coating was proven to be more efficient in combination with the hypochlorite treatment against microfouling under these experimental conditions.
Introduction
Water cooling using natural waters is typically utilized for cooling large industrial facilities such as power plants.Once-through systems are a common choice, when abundant water resources such as lakes, rivers, or seas are accessible.Industrial coolers are systems in which the excess heat is transferred through metal surfaces to colder, circulating water [1].Cooling systems utilizing natural water are susceptible to biofouling, which refers to the undesirable accumulation of biotic deposits on a surface [2].There is a known sequence of biofouling formation: first the surfaces are colonized by microorganisms that prepare the surface conditions suitable for settlement of other biofouling species [2,3].Microfouling, i.e., microbial biofilm formation, on the surfaces of materials in the water cycles can enhance corrosion of these materials.In the industrial systems, the biofilm formation together with inorganic fouling and scaling can cause problems such as an increase in the flow resistance of pipelines or a decrease in the thermal transfer capacity of the heat exchangers [4].It has been estimated that 20% of all corrosion damage of the heat exchangers is caused or influenced by microorganisms [4].The efficiency and availability of a power plant depends to a great extent on the performance and cleanness of materials used.More than 5% of the efficiency of a power plant can be lost due to biofouling on surfaces [5].
Chemical cleaning of the affected area is the main method used for removing unwanted biofilms in the cooling water systems.Currently the most common biocide to prevent biofouling in once-through cooling systems is hypochlorite [5,6].The biocidal effect of hypochlorite is due to its oxidative effect that causes it to react with bromide, which is naturally present in seawater and also possibly with other compounds, both organic and inorganic, leading to the formation of hypobromite and halogenated organic compounds [5,7].The high oxidative power of the hypobromite, formed through the above-mentioned reaction, cause fast inactivation of enzymes and other cellular components of microorganisms.The usage of hypochlorite, however, raises environmental concerns due to production of by-products that are toxic to aquatic organisms [7,8].
In addition to biocides, biofouling may be prevented by application of coatings developed to prevent the settlement of fouling organisms.The functionality of coatings is based on either biocide release, which inhibits fouling settlement by releasing biocidal products into the marine environment, or nontoxic fouling release coatings, whose antifouling efficiency relies on low adhesion strength and easy release of fouling organisms from such a coating [9].Coatings also allow the use of less corrosion resistant metals such as carbon steel, since the base material is not in direct contact with seawater.However, biofilm forming microorganisms are known to deteriorate various polymeric materials, including some coating materials [10,11].Even though the microfouling species would not directly cause deterioration of the coatings, the original antifouling properties such as hydrophobicity of surface can be masked [10].
For effective antifouling treatment, all factors including the environmental impact of biocides and the cost of biocide treatments and antifouling coatings have to be considered [7].Given the costs of biofouling treatment, power plants require biofouling and biocorrosion control solutions that are economically efficient, easy to apply to the system, as well as environmentally sustainable.This can be achieved by adjusting the timing and magnitude of chlorination in combination with special antifouling surfaces [6,[12][13][14].There is a need, however, for deeper understanding of the structure of the microbial community responsible for biofouling of the surfaces under real industrial conditions.
The aim of this research was: (1) to compare the suitability of two commercially available coatings to be used in cooling water systems using hypochlorite as a biocide; (2) to characterize the primary biofilm forming a microbial community in system with or without a hypochlorite treatment and on different coatings.Here we focused on microbial fouling community (bacteria, archaea, and fungi) that attached to the surface of coated (fouling release) carbon steel in a power plant utilizing brackish seawater from the Baltic Sea.Hypochlorite treatment was used to prevent biological fouling in the cooling water system during summer months, when the seawater temperature rises and biofouling is more intensive.The results of a three-month survey of the effect of chlorination on coating performance and species composition of microfouling on coated surfaces are presented and discussed.
Materials
The test material was coated with low alloy carbon steel (corresponding with the composition of carbon steel AISI 1005) from cold-rolled thin sheet (1 mm).The surface of the carbon steel was as received and the coupon size was 25 mm × 75 mm.Two different coatings, a biocide free silicon coating, BioCleanEco (CMP, Tokyo, Japan) and a synthetic purified epoxy tar coating TeknoTar100 (Teknos, Helsinki, Finland), were applied to the carbon steel specimens as instructed by the manufacturers.The coupons were cleaned with the FreeBact-20 (AquaFix, Satsjöbaden, Sweden) and sterile ultra pure water and air-dried before exposure.Each biofouling cell contained eight coupons of each coating.Three parallel coupons were used for material analyses, three coupons for molecular biological analyses, and two for microscopy subsequently.The coupons were photographed prior to the experiment.
Experimental Procedure
To study the fouling in the cooling water cycle, a specially developed biofouling cell was installed into the cooling water cycle (Figure 1) at a power plant where the hypochlorite treatment is routinely used to control biofouling.Two units were installed in the cycle before the chlorination point ("controls") and two units after the hypochlorite dosage point ("hypochlorite").The water flow through the biofouling cell was 0.1 m•s −1 .
Experimental Procedure
To study the fouling in the cooling water cycle, a specially developed biofouling cell was installed into the cooling water cycle (Figure 1) at a power plant where the hypochlorite treatment is routinely used to control biofouling.Two units were installed in the cycle before the chlorination point ("controls") and two units after the hypochlorite dosage point ("hypochlorite").The water flow through the biofouling cell was 0.1 m•s −1 .Na-hypochlorite (15%) was dosed in the cooling system to obtain a final target concentration of 3 mg•L −1 .During the first 30 days chlorination was applied for two hours with 10 h breaks between the treatments, and after that for one hour with 11 h breaks in between.The experiment period started at July 2014, a week after chlorination had already started.The first set of biofouling cells, one control cell and one cell after the hypochlorite dosage point were removed from the cooling water cycle after 32 days of exposure (one month).The second sampling was performed after 90 days of exposure (3 months).The biofouling cells were removed from the water cycle and sealed so that water remained inside the cells.The cells were dismantled aseptically in the laboratory.
The temperature of seawater used for cooling (seawater IN) and the temperature of cooling water after the cycle (seawater OUT) were monitored continuously during the experiment period.The chemical parameters of seawater used for cooling were measured once a month during the experiment period.
Surface Characterization
The coupons for the material studies were immersed in 96% ethanol, air-dried, and stored in a glass desiccator until studied.All coupons for material analyses were photographed and examined with a stereomicroscope.
The condition of the exposed coated surfaces was examined using a Zeiss ULTRAplus field emission scanning electron microscope (Carl Zeiss Microscopy GmbH, Jena, Germany).
For the confocal microscopy, the samples were gently rinsed in water and stained with the DNA specific fluorescence dye SYTO9 (Thermo Fisher Scientific, Eugene, OR, USA) for 25 min in dark at room temperature.Visualization of the biofilms was done by confocal laser scanning microscopy (CLSM) using a Zeiss LSM 710 microscope (Carl Zeiss Microscopy GmbH) using the EC Epiplan-Neofluar 20 × 0.5 NA or the 63 × 0.9 NA (oil) objectives.The 488 nm Ar laser was used and a green emission was collected at 498-566 nm.The images were acquired and analyzed using the ZenPro (black edition) software (Carl Zeiss Microscopy GmbH).Z-stacks (vertical sections) were recorded with the optimal parameters suggested by the software.The number of the Z-stacks depended on the biofilm thickness.Auto-fluorescence of the coatings was checked using unused samples.The TerknoTar100 coating showed auto-fluorescence in the green area of the spectrum.Na-hypochlorite (15%) was dosed in the cooling system to obtain a final target concentration of 3 mg•L −1 .During the first 30 days chlorination was applied for two hours with 10 h breaks between the treatments, and after that for one hour with 11 h breaks in between.The experiment period started at July 2014, a week after chlorination had already started.The first set of biofouling cells, one control cell and one cell after the hypochlorite dosage point were removed from the cooling water cycle after 32 days of exposure (one month).The second sampling was performed after 90 days of exposure (3 months).The biofouling cells were removed from the water cycle and sealed so that water remained inside the cells.The cells were dismantled aseptically in the laboratory.
The temperature of seawater used for cooling (seawater IN) and the temperature of cooling water after the cycle (seawater OUT) were monitored continuously during the experiment period.The chemical parameters of seawater used for cooling were measured once a month during the experiment period.
Surface Characterization
The coupons for the material studies were immersed in 96% ethanol, air-dried, and stored in a glass desiccator until studied.All coupons for material analyses were photographed and examined with a stereomicroscope.
The condition of the exposed coated surfaces was examined using a Zeiss ULTRAplus field emission scanning electron microscope (Carl Zeiss Microscopy GmbH, Jena, Germany).
For the confocal microscopy, the samples were gently rinsed in water and stained with the DNA specific fluorescence dye SYTO9 (Thermo Fisher Scientific, Eugene, OR, USA) for 25 min in dark at room temperature.Visualization of the biofilms was done by confocal laser scanning microscopy (CLSM) using a Zeiss LSM 710 microscope (Carl Zeiss Microscopy GmbH) using the EC Epiplan-Neofluar 20 × 0.5 NA or the 63 × 0.9 NA (oil) objectives.The 488 nm Ar laser was used and a green emission was collected at 498-566 nm.The images were acquired and analyzed using the ZenPro (black edition) software (Carl Zeiss Microscopy GmbH).Z-stacks (vertical sections) were recorded with the optimal parameters suggested by the software.The number of the Z-stacks depended on the biofilm thickness.Auto-fluorescence of the coatings was checked using unused samples.The TerknoTar100 coating showed auto-fluorescence in the green area of the spectrum.
DNA Extraction
The coupons for microbiological analyses were subjected to the biofilm extraction immediately after they were removed from the biofouling cell.The biomass from the coupon surfaces was removed by brushing for 60 s using Omron Sonic Style B458 brush (Omron, Wegalaan, The Netherlands).The brushes were placed aseptically into tubes containing 20 mL sterile Ringer-solution (Merck, Darmstadt, Germany).The tubes were subsequently vortexed for 60 s.The suspension containing the biomass was filtered on the Sterivex filter unit (Sterivex-GP 0.22 µm; Millipore, Billerica, MA, USA).From both sampling times (32 days and 90 days) 500 mL of chlorinated and non-chlorinated seawater was filtered onto Sterivex or Corning (Corning 0.22 µm CA 430626/PES 431161; Corning, NY, USA) filter units as two replicates.Sterivex filtering units and Corning filters were cut with a sterile scalpel and stored frozen (−80 • C) until DNA extraction.For DNA extraction, Sterivex filter units were aseptically broken with a hammer, and both Sterivex and Corning filters were cut into pieces and placed into the lysing tube of the DNA extraction kit.The DNA extraction was performed using a Fast DNA Spin kit for soil (MP Biomedicals, Santa Ana, CA, USA) according to manufacturer's protocol with the modification that the cells were homogenised in a FastPrep-24 instrument (MP Biomedicals, Santa Ana, CA, USA) at 6 m•s −1 for 3 min.
Quantitative PCR
The number of bacteria forming biofilm on the surfaces was estimated by quantitative polymerase chain reaction (qPCR) of bacterial 16S ribosomal RNA (rRNA).An approximately 200 bp fragment of the 16S rRNA gene was amplified with primers 358F and 534R [15] and detected with SYBR green based detection of double-stranded DNA, using LightCycler ® 480 qPCR machine (Roche Diagnostics, Basel, Switzerland).The DNA concentration of all samples was measured with Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific) and adjusted to ≤10 ng•µL −1 with molecular-biology-grade water (Sigma, St. Louis, MO, USA) prior to qPCR in order to avoid PCR inhibition.The 10-µL reaction mixture contained 1 µL of DNA template, 1 × KAPA SYBR ® FAST Universal qPCR Master Mix (KAPA Biosystems, MA, USA), and 150 nM of both primers.The PCR program consisted of initial 15 min incubation at 95 • C, followed by 45 cycles of denaturation at 95 • C for 10 s, annealing at 57 • C for 35 s and extension at 72 • C for 30 s, followed by a final extension at 72 • C for 3 min and a melting curve analysis.As an external standard, a dilution series (100-10 8 colony forming units (cfu)•µL −1 ) of Escherichia coli VTT E-90418 genomic DNA was used.The results were calculated using Absolute quantification/second derivative maximum in LightCycler ® 480 Software 1.5.0 (Roche Diagnostics).The statistical significance of differences of the means was calculated applying two-way ANOVA (IBM SPSS Statistics version 21, Armonk, NY, USA).
Amplification Library Preparation
The amplification libraries for high-throughput sequencing with Ion Torrent PGM were prepared by PCR from the DNA samples.Bacterial 16S rRNA genes were amplified with the primers [16], targeting the variable region V3-V4 of the 16S rDNA gene, archaeal 16S rRNA genes with the primers [17], targeting the V4 region of the gene and fungal internal transcribed spacer (ITS) gene markers with the primer pair ITS1 and ITS2 targeting the fungal ITS1 region [18].PCR amplification was performed in parallel 25 µL reactions for every sample containing 1 × MyTaq™ Red Mix (Bioline, London, UK), 20 pmol of each primer, up to 25 µL molecular-biology-grade water (Sigma) and 2 µL of template.The PCR program consisted of an initial denaturation step at 95 • C for 3 min, 35 cycles for bacteria and fungi, and 40 cycles for archaea of 15 s at 95 • C, 15 s at 50 • C, and 15 s at 72 • C. A final elongation step of 30 s was performed at 72 • C. The PCR products were verified with agarose gel electrophoresis.Amplicons were sequenced using Ion Torrent PGM sequencing platform (Thermo Fisher Scinetific) at Bioser (University of Oulu, Finland) and amplicons were purified before sequencing.
Sequence Processing and Analysis
The sequence reads obtained from Ion Torrent sequencing were first subjected to quality control using the QIIME software [19].During this step, adapters, barcodes, and primers were removed from the sequence reads, and the quality of base-calls was assessed in order to remove erroneous reads from the data set.Subsequently, chimeric sequence reads, which are a type of sequencing artefact arising from sequences from separate sources fusing into one, were removed from the data set with the USEARCH algorithm version 5.2.236 [20] by de novo detection and through similarity searches against the Greengenes reference database [21] with bacterial and archaeal sequences and UNITE database [22] with fungal sequences.
Groups of similar sequences-i.e., Operational Taxonomic Units (OTUs)-were selected from the chimera-filtered sequence data set following open-reference OTU-picking protocol of QIIME version 1.8.0 against the 97% identity Greengenes or UNITE reference database sets.OTU clustering was performed with UCLUST version 1.2.22q [20] and the seed sequences were selected as the representative OTU sequences.All reads that failed to hit the Greengenes or UNITE reference database with a minimum of 60% identity threshold were discarded as sequencing error.Next, singleton OTUs-i.e., OTUs that were represented by a single sequence-were filtered from the data set.Finally, taxonomy from the domain-level up to species-level was assigned to OTUs via representative OTU sequences with the Ribosomal Database Project classifier algorithm at minimum of 80% confidence [23] with bacterial and archaeal sequences.With fungal ITS sequences, taxonomic assignments were made by the BLASTN algorithm with a maximum E-value of 0.001 [20].
Seawater Characteristics
Seawater temperature varied between 15 and 25 • C for the first 70 days, after which the temperature gradually decreased to 5 • C by the end of the experiment period (Figure 2).Conductivity of seawater was 1010-1020 mS•m −1 , salinity 5.8‰-5.9‰,pH 8, alkalinity 1.4 mmol•L −1 , and total organic carbon content 4-4.2 mg•L −1 during the experiment period.
The sequence reads obtained from Ion Torrent sequencing were first subjected to quality control using the QIIME software [19].During this step, adapters, barcodes, and primers were removed from the sequence reads, and the quality of base-calls was assessed in order to remove erroneous reads from the data set.Subsequently, chimeric sequence reads, which are a type of sequencing artefact arising from sequences from separate sources fusing into one, were removed from the data set with the USEARCH algorithm version 5.2.236 [20] by de novo detection and through similarity searches against the Greengenes reference database [21] with bacterial and archaeal sequences and UNITE database [22] with fungal sequences.
Groups of similar sequences-i.e., Operational Taxonomic Units (OTUs)-were selected from the chimera-filtered sequence data set following open-reference OTU-picking protocol of QIIME version 1.8.0 against the 97% identity Greengenes or UNITE reference database sets.OTU clustering was performed with UCLUST version 1.2.22q [20] and the seed sequences were selected as the representative OTU sequences.All reads that failed to hit the Greengenes or UNITE reference database with a minimum of 60% identity threshold were discarded as sequencing error.Next, singleton OTUs-i.e., OTUs that were represented by a single sequence-were filtered from the data set.Finally, taxonomy from the domain-level up to species-level was assigned to OTUs via representative OTU sequences with the Ribosomal Database Project classifier algorithm at minimum of 80% confidence [23] with bacterial and archaeal sequences.With fungal ITS sequences, taxonomic assignments were made by the BLASTN algorithm with a maximum E-value of 0.001 [20].
Seawater Characteristics
Seawater temperature varied between 15 and 25 °C for the first 70 days, after which the temperature gradually decreased to 5 °C by the end of the experiment period (Figure 2).Conductivity of seawater was 1010-1020 mS•m −1 , salinity 5.8‰-5.9‰,pH 8, alkalinity 1.4 mmol•L −1 , and total organic carbon content 4-4.2 mg•L −1 during the experiment period.
Performance of the Coatings
The fouling on the coatings was examined using a stereo-microscope and a confocal microscope.The inspection revealed fouling on all surfaces (Figures 3 and 4).However, the fouling was less intensive on the hypochlorite-treated surfaces and had an even appearance whereas in the untreated
Performance of the Coatings
The fouling on the coatings was examined using a stereo-microscope and a confocal microscope.The inspection revealed fouling on all surfaces (Figures 3 and 4).However, the fouling was less intensive on the hypochlorite-treated surfaces and had an even appearance whereas in the untreated samples the fouling was present in form of thicker patches on the surfaces (Figure 4).No visual difference between the BioCleanEco and TeknoTar coatings could be observed (Figure 3).samples the fouling was present in form of thicker patches on the surfaces (Figure 4).No visual difference between the BioCleanEco and TeknoTar coatings could be observed (Figure 3).The deposit on the surfaces formed a thin uniform light brown layer on all surfaces (Figure 4) in the hypochlorite-treated environment.In the untreated control system, the thicker dark brown deposits partly covered the surfaces.The deposits resembling both inorganic fouling and biofouling were detected on all surfaces.The deposit on the surfaces formed a thin uniform light brown layer on all surfaces (Figure 4) in the hypochlorite-treated environment.In the untreated control system, the thicker dark brown deposits partly covered the surfaces.The deposits resembling both inorganic fouling and biofouling were detected on all surfaces.samples the fouling was present in form of thicker patches on the surfaces (Figure 4).No visual difference between the BioCleanEco and TeknoTar coatings could be observed (Figure 3).The deposit on the surfaces formed a thin uniform light brown layer on all surfaces (Figure 4) in the hypochlorite-treated environment.In the untreated control system, the thicker dark brown deposits partly covered the surfaces.The deposits resembling both inorganic fouling and biofouling were detected on all surfaces.The confocal microscopy analyses revealed that the biofilms were formed on both the BioCleanEco and on the TeknoTar coated samples exposed to the untreated control system (Figures 5 and 6).The detected biofilm was uneven and ranged between 2 µm and 35 µm in thickness.No difference between the biofilms formed on the TeknoTar coating during one month and three months of the exposure to the control system was observed (Figure 6).Slightly better formed microcolonies were found on the surface of BioCleanEco coated samples after three-month exposure (Figure 5A,B).
Hypochlorite treatment of cooling water clearly reduced biofouling of the BioCleanEco surfaces.After a one-month exposure, detectable biofilm was formed on the surface (Figure 5A).After a three-month exposure, the fouling in hypochlorite-treated coupons was substantially reduced as compared to the control system (Figure 5B).There was, however, accumulation of organic matter on the surface on three-month old BioCleanEco coating, as unspecific fluorescence was seen under the microscope (Figure 5B).
The confocal microscopy analyses revealed that the biofilms were formed on both the BioCleanEco and on the TeknoTar coated samples exposed to the untreated control system (Figures 5 and 6).The detected biofilm was uneven and ranged between 2 μm and 35 μm in thickness.No difference between the biofilms formed on the TeknoTar coating during one month and three months of the exposure to the control system was observed (Figure 6).Slightly better formed microcolonies were found on the surface of BioCleanEco coated samples after three-month exposure (Figures 5a,b).
Hypochlorite treatment of cooling water clearly reduced biofouling of the BioCleanEco surfaces.After a one-month exposure, detectable biofilm was formed on the surface (Figure 5a).After a three-month exposure, the fouling in hypochlorite-treated coupons was substantially reduced as compared to the control system (Figure 5b).There was, however, accumulation of organic matter on the surface on three-month old BioCleanEco coating, as unspecific fluorescence was seen under the microscope (Figure 5b).The effect of hypochlorite treatment on the biofouling of the TeknoTar coating was also notable-only small number of microorganisms was seen on the surfaces (Figure 6).There were also surface areas free of fouling.The accumulation on the organic layer on the TeknoTar surface was difficult to judge due to the autofluorescence of the TeknoTar coating.
The condition of cleaned coated surfaces after the experiment was studied using Scanning Electron Microscopy (SEM).No coating damage due exposure or hypochlorite treatment was observed compared to the unexposed coated specimens.The effect of hypochlorite treatment on the biofouling of the TeknoTar coating was also notable-only small number of microorganisms was seen on the surfaces (Figure 6).There were also surface areas free of fouling.The accumulation on the organic layer on the TeknoTar surface was difficult to judge due to the autofluorescence of the TeknoTar coating.
Characterization of Microbial Population
Seawater, both hypochlorite-treated and controls, contained 2 × 10 7 bacteria•mL −1 at the one-month sampling, and 7 × 10 6 bacteria•mL −1 at the three-month sampling, as determined by qPCR of the 16S rRNA gene.The number of bacteria on the coupons from untreated control system was at the level of 10 8 cm 2 (Figure 7).Hypochlorite treatment reduced the bacterial numbers on the surfaces by 10-1000 fold, depending on the coating and exposure time.Lowest number of bacteria, 10 4 cm 2 , was detected from TeknoTar coatings in hypochlorite-treated system exposed for three months.When interpreting the results from the quantitative PCR assay, it must be borne in mind that the 16S rRNA gene copy numbers vary in different microbial species.Thus, the results do not give absolute cfu numbers, but still give an overview estimate of the trend and can be well used for comparing different treatments.
The diversity of bacterial, archaeal, fungal, and other eukaryotic communities from biofilms growing on the coupons both from the hypochlorite-treated and the untreated control system, were studied using DNA-based high-throughput amplicon sequencing analysis.The total number of bacterial OTUs was 14,409; archaeal OTUs 10,891; and fungal OTUs 5267.From the bacterial sequencing the highest Shannon diversity index (H' = 7.3), which indicates the abundance and evenness of the species present, was obtained from the hypochlorite-treated and control system TeknoTar coupons after a one-month exposure.The lowest bacterial diversity (H' = 4.1) was detected on the hypochlorite-treated BioCleanEco coupons after a three-months exposure.The highest archaeal diversity (H' = 6.7) was detected on BioCleanEco coupons from control system after a onemonth exposure and lowest on the hypochlorite-treated TeknoTar coupons (H' = 0-2.9).Fungal diversity (H' = 6.6) was highest on the treated BioCleanEco coupons after one month of exposure and The condition of cleaned coated surfaces after the experiment was studied using Scanning Electron Microscopy (SEM).No coating damage due exposure or hypochlorite treatment was observed compared to the unexposed coated specimens.
Characterization of Microbial Population
Seawater, both hypochlorite-treated and controls, contained 2 × 10 7 bacteria•mL −1 at the one-month sampling, and 7 × 10 6 bacteria•mL −1 at the three-month sampling, as determined by qPCR of the 16S rRNA gene.The number of bacteria on the coupons from untreated control system was at the level of 10 8 cm 2 (Figure 7).Hypochlorite treatment reduced the bacterial numbers on the surfaces by 10-1000 fold, depending on the coating and exposure time.Lowest number of bacteria, 10 4 cm 2 , was detected from TeknoTar coatings in hypochlorite-treated system exposed for three months.When interpreting the results from the quantitative PCR assay, it must be borne in mind that the 16S rRNA gene copy numbers vary in different microbial species.Thus, the results do not give absolute cfu numbers, but still give an overview estimate of the trend and can be well used for comparing different treatments.
lowest (H' = 0.9) in the hypochlorite-treated TeknoTar coupons after a one-month exposure.When comparing the Chao1 OTU richness estimate values to true detected OTU numbers, 44%-66% of bacterial, 47%-100% of the archaeal, and 47%-100% fungal estimated OTUs were obtained from the subsampled sequence data, meaning that sequencing depth was sufficient enough to fully characterize the microbial communities in most of the samples.The main bacterial group in the biofilms growing on both the hypochlorite-treated and control coupons was Proteobacteria-70%-96% of identified sequences (Figure 8).The bacterial community forming biofilm on the TeknoTar and BioCleanEco coupons in control system, comprised of Alphaproteobacteria (42%-87%), with Rhodobacteraceae being the most common family after both a one-month and three-month exposure.After a one-month exposure, Acidobacteria (5%-20%), with Solibacterales as a most common family, was identified on surfaces of coupons from control system.Also, species belonging to Betaproteobacteria group (6%-17%) were present after a one-month exposure but were not detected after a three-month exposure.Gammaproteobacteria (4%-32%) were detected after a one-month exposure on surface of BioCleanEco and after three months of exposure on surface the of TeknoTar coupons from control system.Species belonging to the class Flavobacteriia (2%-8%) of the phylum Bacteroidetes were present on both, BioCleanEco and TeknoTar coupons, in the control system exposed for three months.
In the hypochlorite-treated coupons, the main bacterial group was Alphaproteobacteria (19%-76%) but unlike in coupons from the control system, more species from the phylum Betaproteobacteria were detected (10%-28%), especially on the hypochlorite-treated BioCleanEco coupons.More species belonging to the class Gammaproteobacteria (19%-65%) were identified on hypochlorite-treated BioCleanEco coupons than control system.The number of Flavobacteriia was higher in hypochloritetreated coupons especially on BioCleanEco coupons after a one-month exposure.No species from the phylum Acidobacteria were detected in hypochlorite-treated coupons.The diversity of bacterial, archaeal, fungal, and other eukaryotic communities from biofilms growing on the coupons both from the hypochlorite-treated and the untreated control system, were studied using DNA-based high-throughput amplicon sequencing analysis.The total number of bacterial OTUs was 14,409; archaeal OTUs 10,891; and fungal OTUs 5267.From the bacterial sequencing the highest Shannon diversity index (H = 7.3), which indicates the abundance and evenness of the species present, was obtained from the hypochlorite-treated and control system TeknoTar coupons after a one-month exposure.The lowest bacterial diversity (H = 4.1) was detected on the hypochlorite-treated BioCleanEco coupons after a three-months exposure.The highest archaeal diversity (H = 6.7) was detected on BioCleanEco coupons from control system after a one-month exposure and lowest on the hypochlorite-treated TeknoTar coupons (H = 0-2.9).Fungal diversity (H = 6.6) was highest on the hypochlorite-treated BioCleanEco coupons after one month of exposure and lowest (H = 0.9) in the hypochlorite-treated TeknoTar coupons after a one-month exposure.When comparing the Chao1 OTU richness estimate values to true detected OTU numbers, 44%-66% of bacterial, 47%-100% of the archaeal, and 47%-100% fungal estimated OTUs were obtained from the subsampled sequence data, meaning that sequencing depth was sufficient enough to fully characterize the microbial communities in most of the samples.
The main bacterial group in the biofilms growing on both the hypochlorite-treated and control coupons was Proteobacteria-70%-96% of identified sequences (Figure 8).The bacterial community forming biofilm on the TeknoTar and BioCleanEco coupons in control system, comprised of Alphaproteobacteria (42%-87%), with Rhodobacteraceae being the most common family after both a one-month and three-month exposure.After a one-month exposure, Acidobacteria (5%-20%), with Solibacterales as a most common family, was identified on surfaces of coupons from control system.Also, species belonging to Betaproteobacteria group (6%-17%) were present after a one-month exposure but were not detected after a three-month exposure.Gammaproteobacteria (4%-32%) were detected after a one-month exposure on surface of BioCleanEco and after three months of exposure on surface the of TeknoTar coupons from control system.Species belonging to the class Flavobacteriia (2%-8%) of the phylum Bacteroidetes were present on both, BioCleanEco and TeknoTar coupons, in the control system exposed for three months.
In the hypochlorite-treated coupons, the main bacterial group was Alphaproteobacteria (19%-76%) but unlike in coupons from the control system, more species from the phylum Betaproteobacteria were detected (10%-28%), especially on the hypochlorite-treated BioCleanEco coupons.More species belonging to the class Gammaproteobacteria (19%-65%) were identified on hypochlorite-treated BioCleanEco coupons than control system.The number of Flavobacteriia was higher in hypochlorite-treated coupons especially on BioCleanEco coupons after a one-month exposure.No species from the phylum Acidobacteria were detected in hypochlorite-treated coupons.Archaeal communities in the control environment on the surfaces of TeknoTar and BioCleanEco coated coupons were quite similar (Figure 9).Sequences affiliating with the phylum Crenarchaeota (43%-97%) dominated the archaeal community in all of the coupons after a one-month exposure.Also, species from Thaumarcheaota and Miscellaneous Crenarchaeota group (MCG) classes were detected and the most common family was Nitrosopumilus.Other major archaeal group, especially on the control coupons exposed for three months was the phylum Euryarchaeota with Methanomicrobia (20%-54%) being the most common class detected.Species from Methanosarcina, Methanocella, and Methanoculleus were most common families identified.Archaeal population extracted from the hypochlorite-treated coupons differed from archaeal population from the control system.Main archaeal group in the biofilms extracted from the hypochlorite-treated TeknoTar and BioCleanEco coupons was Crenarchaeota but only below 2% of sequences were identified as Euryarchaeota.In addition, in the TeknoTar coupons, sequences belonging to Parvarcheaota (4%-11%) were observed.A large number of archaeal sequences remained unidentified (19%-59%) from both TeknoTar and BioCleanEco samples, but this can be due to insufficient representation of archaeal sequences in the sequence databases.After one month of exposure in the control system, the most common fungal classes identified on the surfaces of the TeknoTar coupons were unidentified Ascomycota group (7%-16%) and Dothideomycetes (3%-10%), with Cladosporium as the most common family detected (Figure 10).After three months of Archaeal communities in the control environment on the surfaces of TeknoTar and BioCleanEco coated coupons were quite similar (Figure 9).Sequences affiliating with the phylum Crenarchaeota (43%-97%) dominated the archaeal community in all of the coupons after a one-month exposure.Also, species from Thaumarcheaota and Miscellaneous Crenarchaeota group (MCG) classes were detected and the most common family was Nitrosopumilus.Other major archaeal group, especially on the control coupons exposed for three months was the phylum Euryarchaeota with Methanomicrobia (20%-54%) being the most common class detected.Species from Methanosarcina, Methanocella, and Methanoculleus were most common families identified.Archaeal population extracted from the hypochlorite-treated coupons differed from archaeal population from the control system.Main archaeal group in the biofilms extracted from the hypochlorite-treated TeknoTar and BioCleanEco coupons was Crenarchaeota but only below 2% of sequences were identified as Euryarchaeota.In addition, in the TeknoTar coupons, sequences belonging to Parvarcheaota (4%-11%) were observed.A large number of archaeal sequences remained unidentified (19%-59%) from both TeknoTar and BioCleanEco samples, but this can be due to insufficient representation of archaeal sequences in the sequence databases.Archaeal communities in the control environment on the surfaces of TeknoTar and BioCleanEco coated coupons were quite similar (Figure 9).Sequences affiliating with the phylum Crenarchaeota (43%-97%) dominated the archaeal community in all of the coupons after a one-month exposure.Also, species from Thaumarcheaota and Miscellaneous Crenarchaeota group (MCG) classes were detected and the most common family was Nitrosopumilus.Other major archaeal group, especially on the control coupons exposed for three months was the phylum Euryarchaeota with Methanomicrobia (20%-54%) being the most common class detected.Species from Methanosarcina, Methanocella, and Methanoculleus were most common families identified.Archaeal population extracted from the hypochlorite-treated coupons differed from archaeal population from the control system.Main archaeal group in the biofilms extracted from the hypochlorite-treated TeknoTar and BioCleanEco coupons was Crenarchaeota but only below 2% of sequences were identified as Euryarchaeota.In addition, in the TeknoTar coupons, sequences belonging to Parvarcheaota (4%-11%) were observed.A large number of archaeal sequences remained unidentified (19%-59%) from both TeknoTar and BioCleanEco samples, but this can be due to insufficient representation of archaeal sequences in the sequence databases.After one month of exposure in the control system, the most common fungal classes identified on the surfaces of the TeknoTar coupons were unidentified Ascomycota group (7%-16%) and Dothideomycetes (3%-10%), with Cladosporium as the most common family detected (Figure 10).After three months of After one month of exposure in the control system, the most common fungal classes identified on the surfaces of the TeknoTar coupons were unidentified Ascomycota group (7%-16%) and Dothideomycetes (3%-10%), with Cladosporium as the most common family detected (Figure 10).After three months of exposure, the dominating fungal classes on the TeknoTar coupons were Monoblepharidomycetes (9%-45%) and Blastocladiomycetes (60% in one sample).Most common fungal group on the BioCleanEco coupons from control system exposed for one month were unidentified Ascomycota group (7%-23%) and after three months, Monoblepharidomycetes (11%-27%).Only small differences of fungal communities between the control system and the hypochlorite-treated coupons were detected.The most common fungal groups on the hypochlorite-treated TeknoTar and BioCleanEco coupons were unidentified Ascomycota group (9%-63%) and species from Monoblepharidomycetes group (3%-55%).In addition, after three months of exposure on the hypochlorite-treated TeknoTar surfaces, species belonging to the Malasseziales order (34% in one sample) were detected.A large number of fungal sequences remained unidentified in all coupons (13%-84%).exposure, the dominating fungal classes on the TeknoTar coupons were Monoblepharidomycetes (9%-45%) and Blastocladiomycetes (60% in one sample).Most common fungal group on the BioCleanEco coupons from control system exposed for one month were unidentified Ascomycota group (7%-23%) and after three months, Monoblepharidomycetes (11%-27%).Only small differences of fungal communities between the control system and the hypochlorite-treated coupons were detected.The most common fungal groups on the hypochlorite-treated TeknoTar and BioCleanEco coupons were unidentified Ascomycota group (9%-63%) and species from Monoblepharidomycetes group (3%-55%).In addition, after three months of exposure on the hypochlorite-treated TeknoTar surfaces, species belonging to the Malasseziales order (34% in one sample) were detected.A large number of fungal sequences remained unidentified in all coupons (13%-84%).
Discussion
Microfouling on the surfaces of studied anti-fouling coatings was intensive in the cooling water cycle during the warm summer months.As in most cases in natural water environments the fouling was formed by both inorganic fouling and biofouling.The formed biofilm was thick on the surfaces of untreated control samples.Our results indicated that the amount of bacteria on the surfaces from the control system was 10 8 cells•cm −2 .Similar numbers of bacterial microfouling have been reported on anti-fouling surfaces in oceanic waters, also in much warmer sites [24][25][26].Hypochlorite treatment reduced the bacterial numbers by 10-1000-fold, but the efficiency depended on the coating.The combination of TeknoTar coating and hypochlorite treatment was most efficient in reducing the attachment of bacteria, causing a 100-1000-fold decrease in number of attached bacteria compared to the untreated control, depending on the exposure time.With BioCleanEco the difference between the untreated control and hypochlorite-treated samples was 10-fold at both exposure times.The confocal results suggest that the chlorination of the cooling water probably did not prevent the bacterial adherence to the surface, but rather killed cells later, resulting in accumulation of the cell debris on the surface.
Biocidal coatings have been demonstrated to decrease the bacterial diversity of a microfouling community when compared to biocide-free coating [27].The antifouling coating used here did not produce as straightforward results combined with the hypochlorite treatment.The highest bacterial diversity was detected on TeknoTar coatings after a one-month exposure irrespective to whether they had received a hypochlorite treatment or not, even though the number of bacteria was significantly decreased due to the hypochlorite.On the other hand, the lowest bacterial diversity was detected on the three-month hypochlorite-treated BioCleanEco specimens.
Discussion
Microfouling on the surfaces of studied anti-fouling coatings was intensive in the cooling water cycle during the warm summer months.As in most cases in natural water environments the fouling was formed by both inorganic fouling and biofouling.The formed biofilm was thick on the surfaces of untreated control samples.Our results indicated that the amount of bacteria on the surfaces from the control system was 10 8 cells•cm −2 .Similar numbers of bacterial microfouling have been reported on anti-fouling surfaces in oceanic waters, also in much warmer sites [24][25][26].Hypochlorite treatment reduced the bacterial numbers by 10-1000-fold, but the efficiency depended on the coating.The combination of TeknoTar coating and hypochlorite treatment was most efficient in reducing the attachment of bacteria, causing a 100-1000-fold decrease in number of attached bacteria compared to the untreated control, depending on the exposure time.With BioCleanEco the difference between the untreated control and hypochlorite-treated samples was 10-fold at both exposure times.The confocal results suggest that the chlorination of the cooling water probably did not prevent the bacterial adherence to the surface, but rather killed cells later, resulting in accumulation of the cell debris on the surface.
Biocidal coatings have been demonstrated to decrease the bacterial diversity of a microfouling community when compared to biocide-free coating [27].The antifouling coating used here did not produce as straightforward results combined with the hypochlorite treatment.The highest bacterial diversity was detected on TeknoTar coatings after a one-month exposure irrespective to whether they had received a hypochlorite treatment or not, even though the number of bacteria was significantly decreased due to the hypochlorite.On the other hand, the lowest bacterial diversity was detected on the three-month hypochlorite-treated BioCleanEco specimens.
In addition to decreasing the bacterial number, the hypochlorite treatment also changed the species composition forming the biofilm on the surfaces of two antifouling coatings.Whereas mainly Alphaproteobacteria formed the bacterial biofilm in the non-chlorinated system, in the hypochlorite-treated system the relative abundance of Beta-and Gammaproteobacteria increased.Cassé et al. [27] compared the number of microfouling bacteria and the species composition on different coatings in a marine environment.Similar to our experiment, these researchers demonstrated that there was none or only little difference in the number of microfouling bacteria, but the species composition differed between the coatings [27].
The dominant alphaproteobacterial family was Rhodobacteraceae, which has previously been linked especially to early development of biofilm in marine environments, but also detected from this same environment earlier [28][29][30][31].Many bacteria belonging to Rhodobacteraceae are known to perform phototrophic iron oxidation [32].In addition, it has been speculated that the members of Rhodobacteraceae form the initial biofilm since they are able to survive on nutrient poor conditions of early stages of biofilm formation.Moreover, these microorganisms are capable of forming biofilm on copper-based biocidal antifouling coating [30,33].
Archaeal fouling has not been studied as extensively as bacterial fouling and only few reports describe the archaeal community in marine fouling [2,31].The family Nitrosopumilus was the most commonly detected archaea.They perform chemoautotrophic ammonia oxidation and could thus be important partners in the biofilm-forming community, providing the source of nitrogen compounds to the other microorganisms [34].
Monoblepharidomycetes and unidentified Ascomycota were the dominant fungi detected.Most aquatic fungi belong to Ascomycetes and Chytridiomycetes, as do the fungi detected here [35].Only small differences were detected in fungal communities between the control system and the hypochlorite-treated coupons.Similar fungal composition has been reported earlier from the same environment but different from current results; Monoblepharidomycetes were earlier detected only from the hypochlorite-treated environment [31].Fungi has been connected to deterioration of polymeric coatings [36].Fungi also exert physical forces, unlike bacteria or archaea, and have the ability to form hyphae, to penetrate hard substrates, and to spread on and through materials [37].Overall, fungi have an important role in the biofilm since they are able to degrade autochthonous and allochthonous organic matter [35].
The biofouling and biodeterioration of man-made materials, including metal and their alloys, due to biofilm formation has a great environmental and economical implications [38].Antifouling coatings are widely used on marine structures to control biofouling [39].Both coatings included in this study were fouling release type of coatings [9].The effectiveness of most of the non-stick coatings is based on a combination of surface free energy, surface structure, and surface roughness [40].Here, the anti-fouling coatings alone did not prevent the settlement of microfouling organisms.Since microbial fouling affects the performance of commercial antifouling coatings the quantitative and qualitative analyses of marine microbial fouling communities is of great interest [25,39].As summarized by Flemming [4], detailed knowledge about biofilms is crucial for understanding and preventing biofouling, as well as to choose a successful anti-fouling strategy.Microorganisms may cause deterioration of polymeric substances, such as the coatings [10].However, during the relatively short survey time (three months) in this study, no degradation of coatings was evident.Even though the microfouling species would not directly cause deterioration of the coatings, the original surface properties, such as hydrophobicity, can be masked [10].
The present study demonstrates that microfouling combined to inorganic fouling was extensive and formed by a diverse group of microorganisms in a cooling water cycle utilizing brackish Baltic Sea water.The composition of a microfouling community, not only the number of microorganisms, has been identified to be an important factor in either promoting or inhibiting the macrofouling settlement depending on the species composition [3].The hypochlorite treatment combined to anti-fouling coatings proved to be efficient at least during this three-month survey period.
Conclusions
The microfouling on the surfaces of the studied fouling-release coatings was intensive during the warm summer months.The fouling consisted of both inorganic fouling and biofouling.According to the results presented here, the chlorination reduced the microfouling in the brackish water by 10-1000-fold, but also altered the structure of the remaining microbial community forming the biofilm on the surfaces.The chlorination did not prevent the adhesion, but rather killed the biofouling species after the attachment.The combination of TeknoTar coating and the hypochlorite treatment was most efficient in reducing the attachment of bacteria.The highest bacterial diversity was detected on surfaces of the TeknoTar coating, irrespective to whether they had received a hypochlorite treatment or not, even though the number of bacteria was significantly decreased due to the hypochlorite.
Figure 1 .
Figure 1.Schematic illustration of experiment setup, not in scale.
Figure 1 .
Figure 1.Schematic illustration of experiment setup.Not in scale.
Figure 2 .
Figure 2. Temperature of seawater used for cooling (IN) and temperature of the seawater after cooling circle (OUT).
Figure 2 .
Figure 2. Temperature of seawater used for cooling (IN) and temperature of the seawater after cooling circle (OUT).
Figure 3 .
Figure 3. Specimens before (A,D) and after the three-month experiments (B,C,E,F).First column shows BioCleanEco coating (white coating), and second column shows TeknoTar coating (black coating).In addition, (B,E) and (C,F) show specimens exposed to control and hypochlorite systems, respectively.
Figure 3 .
Figure 3. Specimens before (A,D) and after the three-month experiments (B,C,E,F).First column shows BioCleanEco coating (white coating), and second column shows TeknoTar coating (black coating).In addition, (B,E) and (C,F) show specimens exposed to control and hypochlorite systems, respectively.
Figure 3 .
Figure 3. Specimens before (A,D) and after the three-month experiments (B,C,E,F).First column shows BioCleanEco coating (white coating), and second column shows TeknoTar coating (black coating).In addition, (B,E) and (C,F) show specimens exposed to control and hypochlorite systems, respectively.
Figure 5 .
Figure 5. Confocal microscopy images of biofilm on the surface of BioCleanEco coated specimens.The biofilms were washed and stained with SYTO9, and Z-stacks of each were acquired by confocal laser scanning microscopy (CLSM).Upper panels, orthogonal projections of Z-stacks; bottom panels, 3D projections.Two random microscope fields are presented, (A) after one month; (B) after three months.
Figure 5 .
Figure 5. Confocal microscopy images of biofilm on the surface of BioCleanEco coated specimens.The biofilms were washed and stained with SYTO9, and Z-stacks of each were acquired by confocal laser scanning microscopy (CLSM).Upper panels, orthogonal projections of Z-stacks; bottom panels, 3D projections.Two random microscope fields are presented, (A) after one month; (B) after three months.
Figure 6 .
Figure 6.Confocal microscopy images of biofilm on the surface of TeknoTar coated specimens.The biofilms were washed and stained with SYTO9, and Z-stacks of each were acquired by CLSM.Upper panels, orthogonal projections of Z-stacks; bottom panels, 3D projections.Two random microscope fields are presented, (A) after one month; (B) after three months.
Figure 6 .
Figure 6.Confocal microscopy images of biofilm on the surface of TeknoTar coated specimens.The biofilms were washed and stained with SYTO9, and Z-stacks of each were acquired by CLSM.Upper panels, orthogonal projections of Z-stacks; bottom panels, 3D projections.Two random microscope fields are presented, (A) after one month; (B) after three months.
Figure 7 .
Figure 7. Numbers of bacteria on coupon surfaces after (a) one-month and (b) three-month exposure in control system or hypochlorite-treated sea water, as determined using quantitative PCR.Bars show standard deviations.cfu: colony forming units.
Figure 7 .
Figure 7. Numbers of bacteria on coupon surfaces after (A) one-month and (B) three-month exposure in control system or hypochlorite-treated sea water, as determined using quantitative PCR.Bars show standard deviations; cfu: colony forming units.
J 15 Figure 8 .
Figure 8.The bacterial community composition on surfaces of BioCleanEco coating and TeknoTar coating in hypochlorite-treated cooling water system and in control system.
Figure 9 .
Figure 9.The archaeal community composition on surfaces of BioCleanEco coating and TeknoTar coating in hypochlorite-treated cooling water system and in control system.
Figure 8 .
Figure 8.The bacterial community composition on surfaces of BioCleanEco coating and TeknoTar coating in hypochlorite-treated cooling water system and in control system.
J 15 Figure 8 .
Figure 8.The bacterial community composition on surfaces of BioCleanEco coating and TeknoTar coating in hypochlorite-treated cooling water system and in control system.
Figure 9 .
Figure 9.The archaeal community composition on surfaces of BioCleanEco coating and TeknoTar coating in hypochlorite-treated cooling water system and in control system.
Figure 9 .
Figure 9.The archaeal community composition on surfaces of BioCleanEco coating and TeknoTar coating in hypochlorite-treated cooling water system and in control system.
Figure 10 .
Figure 10.The fungal community composition on surfaces of BioCleanEco coating and TeknoTar coating in in hypochlorite-treated cooling water system and in control system.
Figure 10 .
Figure 10.The fungal community composition on surfaces of BioCleanEco coating and TeknoTar coating in in hypochlorite-treated cooling water system and in control system.
|
2019-04-26T14:24:21.520Z
|
2016-11-11T00:00:00.000
|
{
"year": 2016,
"sha1": "04da37888182c28e2a72ed2a911eef15fbbcd1bf",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2077-1312/4/4/74/pdf?version=1478853703",
"oa_status": "GOLD",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "04da37888182c28e2a72ed2a911eef15fbbcd1bf",
"s2fieldsofstudy": [
"Engineering"
],
"extfieldsofstudy": [
"Materials Science"
]
}
|
215409055
|
pes2o/s2orc
|
v3-fos-license
|
Advances on Chelation and Chelator Metal Complexes in Medicine
Metal ions such as iron, copper and zinc are essential for life. Chelators (Chele, greek χειλή-claw of a crab) are organic molecules possessing specific ligands which have high affinity and can bind/carry metal ions and play very important roles in living systems e.g., haemoglobin, transferrin, phytochelators and microbial siderophores [...].
Metal ions such as iron, copper and zinc are essential for life. Chelators (Chele, greek χειλή-claw of a crab) are organic molecules possessing specific ligands which have high affinity and can bind/carry metal ions and play very important roles in living systems e.g., haemoglobin, transferrin, phytochelators and microbial siderophores [1].
The acquisition of daily dietary requirements and the maintenance of a specific range of concentrations of metal ions in the tissues ensure normal daily biological, metabolic and physiological activities and bodily functions, as well as healthy living. For example, it is estimated that zinc is required for the turnover of more than 300 catalytically active zinc metalloproteins, as well as more than 2000 zinc dependent transcription factors.
Metal metabolic imbalance is associated with serious conditions such as iron deficiency anaemia, which affects about a third to a quarter of the world's population. Similarly, many hundreds of thousands of patients are affected by iron overload due to regular red blood cell transfusions. Among the principal categories of patients with negative consequences from transfusional iron overload are chronic haematological and malignant diseases including thalassaemia, haematopoietic stem cell transplantation, sickle cell anaemia, aplastic anaemia and cancer [1]. There are also many cases of iron and copper overload as a result of increased gastrointestinal absorption, such as iron overload in idiopathic haemochromatosis or copper overload in Wilson's disease. The progressive accumulation of iron and copper in the body could lead to major toxicities and fatalities, unless removal therapies of these metals are introduced.
In general, the accumulation of excess metal ions including iron, copper, zinc and aluminium in any organ, as well as in cellular and sub-cellular compartments is a negative prognostic factor for many diseases [1].
The therapeutic application of chelating drugs can in many of the above and also similar other cases restore metal metabolic imbalance and treat the associated diseases (Table 1). For example, the chelating drugs deferoxamine, deferiprone, deferasirox and their combinations are used daily for the treatment of iron overload in thalassaemia and the chelating drugs penicillamine and triethylenetetramine for the treatment of copper overload in Wilson's disease. Many other chelating drugs are also used for the detoxification of other metals [2].
Examples of the Clinical Applications of Chelators
Removal of metals in diseases associated with essential metal overload All diseases related to free radical pathology Cancer Infectious diseases Neurodegenerative diseases Acute kidney disease Myocardial infarction Ageing Prevention of metal absorption in the gastrointestinal tract Xenobiotic metal decorporation originating from food and drink products, environmental pollution, the nuclear industry and weapons Decorporation of xenobiotic metals used in medical diagnosis
Examples of the clinical applications of chelator metal complexes
Increase metal absorption in diseases of essential metal deficiency Delivery of redox active metal complexes against cancer Delivery of xenobiotic metals to disrupt essential metal pathways in cancer Diagnostic metal delivery in diseases including inflammation and cancer Radiolabeling of cells using metal radiotracers Theranostic metal delivery in diseases Metal ions are also essential for the growth and proliferation of microbes and cancer cells, which in effect makes them a target for the design of new pharmaceuticals for treating infections and cancer (Table 1) [1,3,4]. The latter two are considered to be on the top of the list of major diseases affecting humans.
There is variation in the requirements for metal ions in each cell, including different microbial cell species and cancer cell types. Selective targeting for the inhibition of the growth and proliferation of cancer cells and microbes by deprivation of essential metals can be accomplished by chelating agents with specific properties and physicochemical characteristics [3,4]. In contrast, some natural or synthetic chelating agents and medicinal drugs could have the opposite effect and increase metal delivery thus promoting the growth of cancer and microbial cells [1,4].
The treatment of cancer is a major goal for many research institutions and investigators worldwide. Anticancer therapeutic strategies involve surgery, chemotherapy, radiotherapy and their combinations. The aim of these therapeutic strategies is to reduce cancer cell proliferation and cancer growth, and, if possible, eliminate altogether the presence of cancer cells. In most cases of chemotherapy, specific anticancer drug protocols are directed to various targets in cancer cells such as those related to key metabolic protein pathways, DNA synthesis, signal transduction etc. Damage to normal cells and other toxic effects are also observed during the anticancer treatments. The risk/benefit assessments for most anticancer therapies are not optimal and new drugs and approaches are required, including the possible application of multidrug anticancer regimens aiming at different targets. Specificity in cancer cell targeting is also important and such approaches are considered to have a better therapeutic outcome. The efficacy and risk/benefit assessments of such combination approaches which may involve metal ions and chelators is in progress, with some of these at an investigational level and others in different stages of clinical development and use [1,4,5].
The therapeutic strategies for targeting cancer cell growth and proliferation by chelators may involve the modulation of protein function or other associated pathways such as the inhibition of key enzymes e.g., the iron containing ribonucleotide reductase involved in DNA synthesis, incorporation of xenobiotic metals such as Ga which could bind to nucleotide substrates and also disrupt iron and other essential metal metabolic pathways and lysosomic damage by redox active metal complexes (Table 1) [1,[4][5][6]. Chelating drugs could also be used in many other anticancer approaches such as modulation of macrophage anticancer function and also in the protection from the cardiotoxicity of anticancer drugs [7][8][9]. Of particular interest in clinical oncology are platinum complexes [10]. It is estimated that about 50% of cancer patients receive platinum complex based anticancer therapy. Such platinum complexes appear to cause crosslinking of DNA, which causes the inhibition of DNA synthesis and DNA repair in cancer cells. Each of several drugs containing different platinum-based complexes appear to have selective anticancer specificity for each cancer type [10]. Similarly, the anticancer chelating drug hydroxycarbamide (hydroxyurea) which inhibits ribonucleotide reductase is also widely used in different cancer types [1,11].
Another major area affecting health and disease, which involves metal ions and chelators is free radical pathology. Increased production of free radicals and reactive oxygen species (FR/ROS) has been implicated in almost all diseases, including those involving tissue damage, as well as in cancer and aging. In almost all biological systems the catalysis of free radical production including the effects leading to oxidative stress toxicity is carried out by iron, copper and enzymes containing these metals. The inhibition of free radical toxicity by specific iron and copper chelators, supports their selective use as main or adjuvant antioxidant therapy in free radical pathology (Table 1) [1,12].
Oxidative stress toxicity due to excess FR/ROS has been shown to cause widespread molecular, sub-cellular and cellular damage, which may lead to cell apoptosis, autophagy and necrosis. In particular, cell death due to iron toxicity called "ferroptosis" has been identified in many conditions including cancer, acute kidney disease, stroke etc. Ferroptosis appears to have specific characteristics and to be different from apoptosis or other cell death pathways. In this context, the design of antioxidant therapeutic protocols which exclude the control of iron toxicity through iron chelation, appear to be limited and ambiguous in clinical settings [1,12].
A major paradox in the treatment of diseases associated with free radical pathology is the absence of antioxidant drugs in clinical practice. However, millions of people are using daily natural antioxidants for the prevention and treatment of many diseases in the context of traditional folk medicine. The rapid expansion in the use of natural antioxidants such as plant nutraceuticals, which are sold over the counter as dietary supplements for potential nutritional and therapeutic effects, as well as thousands of related reports in the medical literature suggest that many scientists and the public have in general a positive attitude in respect of such therapeutic approaches. Many natural antioxidants including plant nutraceuticals possess metal chelating properties. Some of these phytochelators such as quercetin, curcumin, 8-hydroxyquinoline and ellagic acid are the subject of ongoing clinical investigations. Similarly, several other phytochelators such as maltol, fisetin, caffeic acid, phytic acid and silibinin appear to play a role not only in antioxidant activity but other physiological processes such as metal transport and delivery and also metal detoxification [1,[12][13][14][15].
Xenobiotic metal detoxification is an important area of therapeutic applications of chelating drugs, which is associated with environmental toxicology issues including those arising from heavy metals such as lead and mercury, carcinogenic metals such as cadmium and nickel and radioactive metals such as plutonium and uranium (Table 1) [1,2]. Xenobiotic metal toxicity arising from the ingestion of food and drink product contamination, as well as inhalation of polluted air is a negative prognostic factor and is thought to be associated with many diseases. In this context, millions of people are attending alternative medicine clinics worldwide for ethylenediaminetetraacetic acid (EDTA) chelation therapy, which is widely used for xenobiotic metal detoxification. Similarly, air pollution from industrial and automobile combustion waste products and especially the inhalation of specific size micro particles which contain xenobiotic metals, toxic forms of iron and also other toxins appear to be a serious health hazard and the cause of many fatalities every year. The removal of xenobiotic metals and toxic iron forms from different organs as a result of pollution is a serious challenge for personalized chelation therapy and future intoxication targeting strategies [1,2,12].
Similar intoxication chelating strategies have to be designed also in relation to toxicity concerns arising from the excessive use of diagnostic metals such as gadolinium and technetium, which are routinely used in clinical diagnosis [16].
Specific chelation targeting strategies are also required for the treatment of diseases associated with the excess accumulation of xenobiotic and essential metals in specific organs e.g., excess iron, copper, zinc and aluminium in the brain, all of which have been implicated in neurodegenerative diseases [1,17]. Such strategies should include the selection of specific chelators that can cross the blood-brain barrier such as deferiprone [18,19]. Substantial improvements have been observed in different categories of patients with neurodegenerative diseases using deferiprone including pantothenate kinase-associated neurodegeneration (PKAN) and Friedreich's Ataxia [12,[20][21][22][23][24]. Several other ongoing and planned clinical trials for the use of deferiprone and other chelating drugs in other neurodegenerative diseases with increased morbidity and mortality such as Parkinson's and Alzheimer's disease patients are also in progress [25,26].
Similar targeting chelation strategies could be adopted in relation to different stages of the transport, storage, utilisation or detoxification pathways of metal ions and associated proteins. For example, the interactions of chelators, metal ions and chelator metal complexes with transferrin will affect their metabolic, therapeutic, diagnostic and detoxification properties [17]. In this context, in a total of about 100 elements of the periodic table, about 40 metal ions including all essential and xenobiotic metals mentioned above e.g., iron, copper, zinc, aluminium, gallium, technetium, indium, gadolinium, nickel, cadmium, plutonium, uranium and platinum have been identified to bind or interact with plasma transferrin. The interactions are different in each case and these differences could affect associated targeting strategies including diagnostic, therapeutic and detoxification uses [17]. In particular, the ability of high doses deferiprone to remove iron and other metals from transferrin are of physiological and clinical importance [17].
The development of monitoring methods for assessing the efficacy and mode of action of chelation therapy in the context of personalised medicine is very important for future metal intoxication targeting strategies. For example, new insights in the mode of action of chelating drugs such as the monitoring of the changes in excess iron in organs during chelation therapy using the magnetic resonance imaging (MRI) T2 and T2* techniques have increased the selectivity prospects and targeting therapeutic potential of chelating drugs for different organs including the heart, brain, liver, spleen pancreas and different diseases of gross body iron overload or of focal iron toxicity such as neurodegenerative diseases [21][22][23][24][25]27,28].
A major area of the application of chelators in medicine is the diagnostic and theranostic fields (Table 1) [29,30]. The design and targeted use of different chelating agents which form complexes of variable physicochemical properties appear to affect the bodily distribution of metal radiotracers and are increasingly applied for the diagnosis of different diseases and tracking their progress. Similar targeted chelator metal complexes have been identified for the theranostic application of radiotracer metals in cancer, inflammation and other diseases [29,30].
There are major challenges ahead in relation to the new applications of chelators, chelating drugs and their metal complexes in medicine. In addition to biomedical research developments, the continuation of basic chemical research in the design of new chelators and the characterization of their physicochemical and other properties including those of their metal complexes are essential for increasing their targeting potential and therapeutic and/or diagnostic or theranostic application in medicine [1,[29][30][31].
Many In each case the mode of action of chelators, chelating drugs and their metal complexes will be affected by competition with other natural chelators including chelating proteins and other metal ions. Similarly, in each case the absorption, distribution, metabolism, excretion and toxicity (ADMET) profile will be different and also be affected by pharmacogenomic, metallomic, proteogenomic, metabolomic and redoxomic factors (Figure 1).
The target organs of toxicity are different for each chelator, their metabolites and chelator metal complexes. The same differences in toxicities apply to the accumulation of excess metal ions in organs. Efficacy and toxicity aspects are also affected by the underlying disease and organ function.
Recent new developments in the clinical use, application, drug interactions, metabolic and other effects of chelating drugs and their metal complexes has increased our knowledge in relation to the pivotal role of chelators and metal ions played in living systems, as well as the diagnosis and therapy of many diseases. However, many challenges lie ahead in the optimization aspects of the diagnostic and therapeutic use of chelator and chelator metal complexes.
For example, future chelator targeting processes could be designed aiming for specific organs, cells and cellular compartments, for use in the removal or delivery of essential or xenobiotic metals and for achieving better therapeutic or diagnostic or theranostic clinical results. Similarly, improved chelator platinum complexes and hydroxyurea derivatives could be designed for targeted delivery and better treatment in different types of cancer [10,11]. Investigations using deferiprone and other chelators are also in progress for targeting oxidative stress toxicity and malfunction in mitochondria, which have been implicated in cancer and other diseases [12,[31][32][33][34][35][36]. New therapeutic possibilities include the design of polymeric chelators for the extraction and removal of excess essential and xenobiotic metals in the gastrointestinal tract and in environmental pollution. In contrast lipophilic chelators such as maltol could increase the absorption of iron and other metals [37,38] In each case the mode of action of chelators, chelating drugs and their metal complexes will be affected by competition with other natural chelators including chelating proteins and other metal ions. Similarly, in each case the absorption, distribution, metabolism, excretion and toxicity (ADMET) profile will be different and also be affected by pharmacogenomic, metallomic, proteogenomic, metabolomic and redoxomic factors (Figure 1).
The target organs of toxicity are different for each chelator, their metabolites and chelator metal complexes. The same differences in toxicities apply to the accumulation of excess metal ions in organs. Efficacy and toxicity aspects are also affected by the underlying disease and organ function.
Recent new developments in the clinical use, application, drug interactions, metabolic and other effects of chelating drugs and their metal complexes has increased our knowledge in relation to the pivotal role of chelators and metal ions played in living systems, as well as the diagnosis and therapy of many diseases. However, many challenges lie ahead in the optimization aspects of the diagnostic and therapeutic use of chelator and chelator metal complexes.
For example, future chelator targeting processes could be designed aiming for specific organs, cells and cellular compartments, for use in the removal or delivery of essential or xenobiotic metals and for achieving better therapeutic or diagnostic or theranostic clinical results. Similarly, improved chelator platinum complexes and hydroxyurea derivatives could be designed for targeted delivery and better treatment in different types of cancer [10,11]. Investigations using deferiprone and other chelators are also in progress for targeting oxidative stress toxicity and malfunction in mitochondria, which have been implicated in cancer and other diseases [12,[31][32][33][34][35][36]. New therapeutic possibilities include the design of polymeric chelators for the extraction and removal of excess essential and xenobiotic metals in the gastrointestinal tract and in environmental pollution. In contrast lipophilic chelators such as maltol could increase the absorption of iron and other metals [37,38] Chelating drugs could generally be used as main, alternative or adjuvant therapy in many diseases, including those associated with free radical pathology, in metal detoxification, as antioxidants, anticancer and anti-infective agents, and as modulators of protein function or pathways associated with disease. In this context, improved therapeutic strategies could be developed and adopted based on combination therapies, target-specific and prodrug design methods and other aspects that could fulfil personalized medicine characteristics [31].
However, notwithstanding that such academic research initiatives may benefit millions of patients, pharmaceutical development is generally slow and based on commercial and not academic or ethical criteria [39]. In this context, the current use of available approved drugs such as deferiprone and initiatives for academic drug development in diseases with no effective treatments could speed up the pharmaceutical process and benefit related groups of patients [39].
Funding:
The study was supported from internal funds of the Postgraduate Research Institute of Science, Technology, Environment and Medicine, a non-profit, charitable organization. The article was a result of an invitation by the journal editors and publication costs of the study in open access were waived.
Conflicts of Interest:
The author declares no conflict of interest.
|
2020-04-08T19:10:50.886Z
|
2020-04-01T00:00:00.000
|
{
"year": 2020,
"sha1": "0022de3ab5c9c9ab9ee3912c47f1310b2d3b734c",
"oa_license": "CCBY",
"oa_url": "https://res.mdpi.com/d_attachment/ijms/ijms-21-02499/article_deploy/ijms-21-02499-v2.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "9806159eae7626b401ec8aa442125d043e25657b",
"s2fieldsofstudy": [
"Medicine",
"Chemistry"
],
"extfieldsofstudy": [
"Medicine",
"Chemistry"
]
}
|
119691916
|
pes2o/s2orc
|
v3-fos-license
|
Representations of Conformal Nets, Universal C*-Algebras and K-Theory
We study the representation theory of a conformal net A on the circle from a K-theoretical point of view using its universal C*-algebra C*(A). We prove that if A satisfies the split property then, for every representation \pi of A with finite statistical dimension, \pi(C*(A)) is weakly closed and hence a finite direct sum of type I_\infty factors. We define the more manageable locally normal universal C*-algebra C*_ln(A) as the quotient of C*(A) by its largest ideal vanishing in all locally normal representations and we investigate its structure. In particular, if A is completely rational with n sectors, then C*_ln(A) is a direct sum of n type I_\infty factors. Its ideal K_A of compact operators has nontrivial K-theory, and we prove that the DHR endomorphisms of C*(A) with finite statistical dimension act on K_A, giving rise to an action of the fusion semiring of DHR sectors on K_0(K_A)$. Moreover, we show that this action corresponds to the regular representation of the associated fusion algebra.
Introduction
Conformal field theory (CFT) is the theory of quantum fields with conformal symmetry, see e.g. [DMS96]. This subject is interesting on its own and because it provides further insight into the structural aspects of more general quantum field theories. Moreover it is deeply related with many areas of theoretical physics (e.g. string theory, critical phenomena) and of mathematics (e.g. number theory, topology of three-dimensional manifolds, infinite-dimensional Lie algebras and Lie groups, quantum groups, subfactors). Of particular importance is the two-dimensional case. Here, the chiral fields, namely the fields depending on one light ray coordinate only, play a special role and one is led to study the corresponding chiral theories on the light ray R or more frequently on its compactification S 1 where the conformal symmetry naturally acts.
The operator algebraic approach to two-dimensional chiral CFT (the case we are going to study here) goes through conformal nets on S 1 , namely inclusion-preserving maps I → A(I) from the set of proper (nonempty, nondense, open) intervals of the unit circle S 1 into the family of von Neumann algebras (actually type III 1 factors) acting on a fixed separable Hilbert space and covariant under a given representation of the conformal group, cf. [FRS92,FG93,GL96]. Although there are various important differences mainly due to the low-dimensional spacetime topology, these are the one-dimensional analogues of the nets of local von Neumann algebras on Minkowski spacetime (Haag-Kastler nets) which are the basic ingredients in the so called algebraic quantum field theory (local quantum physics), see [Haa96] for a standard reference book. As a consequence, besides the powerful functional analytic methods of the theory of operator algebras, the operator algebraic approach allows the study of CFT through the theory of superselection sectors of Doplicher, Haag and Roberts (DHR), [DHR71,DHR74].
From the beginning the operator algebraic approach to CFT has shown to provide a natural framework for classification [BMT88] and structural analysis [FRS92,GL92,FG93,Reh90]. A central role has been played by the deep relations with the theory of subfactors initiated by Jones in [Jo83] and the Tomita-Takesaki modular theory of von Neumann algebras [Tak70]. On the one hand the connections with the theory of subfactors have occurred through the analysis of the representations of the braid group which emerge in the study of the DHR statistics in low spacetime dimensions [FRS89,FRS92] and, more explicitly, through the relationship between the Jones index and the statistical dimension of superselection sectors discovered by Longo [Lon89,Lon90]. On the other hand, as shown by Brunetti, Guido and Longo [BGL93] and (independently) by Fröhlich and Gabbiani, [FG93] the action of the modular group of local von Neumann algebras with respect to the vacuum state, can always be described in terms of the conformal symmetry. Subsequently this approach produced a large number of remarkable results not only for the study of CFT but also for its impact on the theory of subfactors. A very important step in these later developments was made by Wassermann [Was98] with the computation of the Connes fusion for the positive energy representations of the loop groups LSU(N). As a consequence of this work the composition of DHR sectors of the associated conformal nets turns out to correspond to the Verlinde fusion. On the basis of this result many other DHR fusion rings have been subsequently determined and found out to be isomorphic to the corresponding Verlinde rings. It also served as a starting point for the work of Xu [Xu00] on Jones-Wassermann subfactors for disconnected intervals which provided inspiration and examples for the notion of completely rational nets (conformal net analogues of rational chiral conformal field theories) introduced shortly after by Kawahigashi, Longo and Müger [KLM01]. We refrain from giving further details on the various achievements of the operator algebraic approach to CFT but we just mention two of the most representative recent results: the classification of conformal nets with central charge c < 1 by Kawahigashi and Longo [KL04] and the development of the technique of mirror extensions for the construction of new nets [Xu07].
Although various areas in the theory of operator algebras, in particular the Tomita-Takesaki modular theory for von Neumann algebras and the theory of subfactors, have played a central role, noncommutative geometry [Con94] and K-theory for operator algebras [Bla98] experienced -with few exceptions -only marginal considerations in the conformal nets setting. In [Lon01] Longo suggested to consider quantum field theories and their superselection sectors as quantum analogue of infinite-dimensional manifolds and elliptic operators respectively in order to look for a geometrical interpretation of the Jones index for subfactors through the statistical dimension of superselection sectors (quantum index theorem). In [KL05] Kawahigashi and Longo, with the above picture as a background, studied the asymptotic behavior of the characters of completely rational conformal nets on S 1 guided by the classical analogy of Weyl's asymptotics for the trace of the heat kernel. Later, Carpi, Kawahigashi and Longo [CKL08] have began a systematic study of local nets on S 1 with superconformal symmetry (superconformal nets). One of the motivations for their work concerned the relations with the noncommutative geometrical framework of Connes. In particular they proved a formula relating the Fredholm index of the (upper off diagonal part of) the supercharge operator in Ramond representations to the Jones index of subfactors and they proposed to study the spectral triples having these supercharge operators as Dirac operators together with the corresponding JLO cocycles in entire cyclic cohomology (Chern characters in quantum K-theory) [JLO88]. An important step in this direction has been done by Carpi, Hillier, Kawahigashi and Longo in [CHKL10] with the definition of nets of spectral triples associated to the unitary representations of the (N=1) super-Virasoro algebras. This requires the solution of nontrivial technical domain problems. In the introduction of the same paper (see also the outlook) an outline of a general "noncommutative geometrization" program for CFT is also given. This program has been recently carried out by Carpi, Hillier and Longo [CHL12] through a general procedure which allows to associate spectral triples and hence entire cyclic cohomology classes of suitably chosen global algebras of differentiable operators associated to the given net and its superselection sectors. Although this procedure gives rise to serious domain problems it is shown by using the index pairing with K-theory [Con94] that the cohomology classes associated to superselection sectors actually provide nontrivial geometric invariants.
At this point a natural question arises. Would it be possible to take a dual (topological) point of view and study the superselection sectors directly in K-theory without using the spectral triples (differentiable structure) and the corresponding entire cyclic cohomology classes as a starting point? Such an approach would seem to involve various advantages: it would eliminate the need to restrict to nets on S 1 with superconformal symmetry; it would avoid the serious domain problems inherent in the use of Connes' spectral triples; it could give further insights into the spectral triples approach; it could shed light on possible connections with other situations where K-theory appears in quantum field theory, see e.g. [Free02]. In this paper we propose a first investigation in this direction.
In order to explain some of the main ideas underlying this work we briefly describe the analogous situation on four-dimensional Minkowski spacetime, for which the DHR theory of superselection sectors was originally formulated. Here one deals with nets O → A(O) of von Neumann algebras where O runs over the set of spacetime double cones. In order to capture the global nature of the superselection sectors one has to define a suitable global C*-algebra. To this end, since the set of double cones is directed under inclusion, one can consider the C*-inductive limit of the local von Neumann algebras A(O), which here, as usual, will be denoted by A as the underlying net. The superselection sectors of the theory are realized as DHR (i.e. localized and transportable) unital endomorphisms of the C*-algebra A. As a consequence the DHR-sectors act on the K-theory of A and one might be tempted to use this action in order to obtain K-theoretical invariants for the DHR sectors. To simplify our discussion we assume the split property which is expected to hold for any physically reasonable net (see [Haa96,V.5]). Then it turns out that the quasi-local C*-algebra A is a simple C*-algebra with trivial K-theory and it is moreover independent of the underlying net (cf. Proposition 2.5). In particular the action of DHR sectors on the K-theory of A gives no information at all.
For a conformal net A on S 1 the situation is different: although the set of intervals is not directed, one can define in a suitable way a universal C*-algebra C * (A). It is generated by the local algebras associated to all proper intervals of S 1 and the equivalence classes of its irreducible locally normal representations are in one-to-one correspondence with the superselection sectors of the net. It therefore contains a lot of information about the net A, cf. [Fre90,FRS92]. As in the case of the quasi-local C*-algebra the superselection sectors can be realized as DHR endomorphisms of C * (A), but in general C * (A) is no longer simple and the vacuum representation is not faithful either. Moreover, as an abstract C*-algebra, it depends on the representation theory of the underlying net (cf. Proposition 2.3).
Unfortunately we are not able to determine C * (A) or its K-theory even in special cases. The problem is that the definition of C * (A) involves plenty of representations of A which are not locally normal and thus difficult to handle. On the other hand these non-locally normal representations do not appear to have direct relevance for CFT. It seems therefore natural to consider a related global C*-algebra here called the locally normal universal C*-algebra of the net A and denoted by C * ln (A). It can be defined as the quotient of C * (A) by the largest of its ideals vanishing in all locally normal representations and it admits by definition faithful locally normal representations. Moreover, if one restricts to locally normal representations of the net A then C * ln (A) has all the relevant properties of C * (A). In particular the superselection sectors of A can also be realized as DHR-endomorphisms of C * ln (A). One might be tempted to believe that the nontrivial topology of S 1 may result in a nontrivial K-theory for the algebra C * ln (A). However, one of the main results of this paper is the proof of the following surprising fact: for every completely rational conformal net A on S 1 the locally normal universal C*-algebra C * ln (A) is isomorphic to a finite direct sum of countably decomposable infinite-dimensional type I factors (one for each sector of A), see Theorem 2.15. As a consequence K 0 (C * ln (A)) = 0 = K 1 (C * ln (A)) (Theorem 4.3). Actually the above result follows from a more general fact (Theorem 2.13): if A is a conformal net on S 1 satisfying the split property (but not necessarily completely rational) then π(C * (A)) is weakly closed for every locally normal representation π with finite statistical dimension.
In order to overcome the problem of trivial K-theory we consider, for a given completely rational net A, the largest norm-separable ideal K A of C * ln (A). If A has n sectors then K A is isomorphic to the direct sum of n copies of the C*-algebra K(H) of compact operators on a separable infinite-dimensional Hilbert space H. We show that the DHR endomorphisms of C * ln (A) with finite statistical dimension restrict to endomorphisms of K A and they give rise to an action of the fusion semiring R A of sectors with finite statistical dimension on K 0 (K A ) which corresponds to the regular representation of the fusion algebra generated by R A (Theorem 4.4).
We end this introduction with some comments on possible connections with the work of Freed, Hopkins and Teleman on loop groups and twisted K-theory [FHT11], cf. also [EG09] for relations with subfactors and modular invariants. If G is a simply connected compact Lie group and LG = C ∞ (S 1 , G) is the corresponding loop group then the projective unitary representation of LG with lowest energy 0 (vacuum representation) and level k gives rise to a conformal net A G k on S 1 , see [FG93]. Assume that A G k is completely rational and that the ring whose elements are the formal differences of elements in the semiring R A G k generated by the DHR sectors of A G k is isomorphic to the Verlinde fusion ring R k (LG) (this is known to be true e.g. for G = SU(N ) and all positive integers k as a consequence of the results in [Was98, Xu00] mentioned above, cf. also the comments at the end of Sect. 1 and Remark 2.2 (4)). As shown in [FHT11] R k (LG) is isomorphic to a twisted equivariant K-theory of the Lie group G. Hence our results indirectly exhibit a relation between the above twisted equivariant K-theory of G and the K-theory of the noncommutative C*-algebra K A for A = A G k .
Preliminaries on conformal nets
We fix the following notation: N and N 0 stand for the positive integers and the nonnegative integers, respectively. Given a (complex) Hilbert space H, B(H) will stand for the C*algebra of bounded linear operators on H, K(H) (or sometimes simply K, if H is infinitedimensional separable) for the C*-subalgebra of compacts on H, ⊗ the minimal (spatial) tensor product of C*-algebras (i.e., the completion with respect to the minimal C*-norm on the algebraic tensor product), and⊗ the (weakly closed) tensor product of von Neumann algebras. We say that two * -isomorphisms φ i : A → B, i = 1, 2 of the C*-algebra A into the unital C*-algebra B are unitarily equivalent in B if there is a unitary u ∈ B such that When composing homomorphisms of algebras or groups, we will usually drop the sign " • " between the two homomorphisms. Finally, we denote by Z(A) the center of any associative algebra A.
Möbius covariant nets
Let I be the set of open, nonempty and nondense intervals of the unit circle S 1 = {z ∈ C : |z| = 1}. Diffeomorphisms of S 1 of the form z → az+b bz+a with a, b ∈ C, |a| 2 − |b| 2 = 1, are called Möbius transformations; they form a Lie group isomorphic to PSL(2, R), and we shall identify these two groups henceforth. In particular, rotations are Möbius transformations and we shall use the symbol R α to denote the anticlockwise rotation of S 1 by an angle α.
A Möbius covariant net A over S 1 (cf. [FG93,GL96]) consists of a family of von Neumann algebras A(I) (I ∈ I) acting on a common separable Hilbert space H together with a given strongly continuous representation U of PSL(2, R) satisfying • locality: elements of A(I 1 ) commute with those of A(I 2 ) whenever I 1 ∩ I 2 = ∅, • covariance: U (g)A(I)U (g) * = A(g(I)) for all g ∈ PSL(2, R) and I ∈ I, • positivity of the energy: the conformal Hamiltonian L 0 , defined by the equation U (R α ) = e iαL 0 (∀α ∈ R), is positive, • existence, uniqueness and cyclicity of the vacuum: up to phase there exists a unique unit vector Ω ∈ H called the "vacuum vector" which is invariant under the action of U ; moreover, it is cyclic for the von Neumann algebra I∈I A(I).
There are many known important consequences of the above definition. In particular, we shall often use the following ones (for proofs see [FG93,FJ96,GL96]). There are some further properties of A which do not follow from the above axioms, but are often satisfied in the physically interesting cases. Strong additivity: A(I) = A(I 1 ) ∨ A(I 2 ) whenever I 1 , I 2 , I ∈ I are such that the closure of I coincides with the closure of I 1 ∪ I 2 . Split property: for any inclusionĪ − ⊂ I + , there is a type I factor F such that A(I − ) ⊂ F ⊂ A(I + ); there is actually a canonical choice of F , cf. [DL84], but this is irrelevant in the present publication. Complete rationality: the net is strongly additive, split and its µ-index, that is the index of the so-called "2-interval inclusions", is finite. The µ-index of a Möbius covariant net A on S 1 with the split property is defined as follows: let I 1 , I 2 , I 3 , I 4 ∈ I be four intervals in anticlockwise order, obtained by removing four points from S 1 . Then A(I 1 ) ∨ A(I 3 ) ⊂ (A(I 2 ) ∨ A(I 4 )) ′ is an irreducible inclusion of type III factors ("2-interval inclusion") whose index [(A(I 2 ) ∨ A(I 4 )) ′ : A(I 1 ) ∨ A(I 3 )] does not depend on the choice of the intervals and is called the µ-index of A, cf. [KLM01] for more details.
Conformal nets
Let Diff(S 1 ) be the group of orientation preserving (smooth) diffeomorphisms of S 1 , which we shall consider with the usual C ∞ -topology. In what follows, we shall consider strongly continuous projective unitary representations of Diff(S 1 ). If V is such a representation, then sometimes for a g ∈ Diff(S 1 ) we shall think of V (g) as a unitary operator. Although there is more than one way to "fix phases", note that expressions like Ad(V (g)) or V (g) ∈ M for a von Neumann algebra M ⊂ B(H) are unambiguous. We shall also say that V is an extension of the unitary representation U of PSL(2, R) if we can arrange the phases in such a way that V (g) = U (g), or without mentioning phases: Ad(V (g)) = Ad(U (g)), for all g ∈ PSL(2, R).
A conformal net is a Möbius covariant net (A, U ) such that U extends to a continuous projective unitary representation of Diff(S 1 ) denoted again by U and satisfying • U (g)A(I)U (g) * = A(gI), • g| I = id I ⇒ Ad(U (g))| A(I) = id A(I) , for all g ∈ Diff(S 1 ) and I ∈ I. Note that though there are some "pathological " cases in which such an extension does not exist, when it does exist, it is unique, cf. [CW05,Wei05]. Note also that by Haag-duality, if a diffeomorphism is localized in the interval I -i.e. it acts trivially elsewhere -then, by the second listed property, the corresponding unitary is also localized in I in the sense that it belongs to A(I). From now on all nets are supposed to be conformal.
Representations of conformal nets
A representation of A is a family π = (π I ) I∈I of (unital) * -representations π I of A(I) on a common Hilbert space H π such that π I 2 |A(I 1 ) = π I 1 whenever I 1 ⊂ I 2 . The representation π is called locally normal if π I is normal for every I ∈ I; by [Tak79, Theorem 5.1], this is always the case if H π is separable. Conversely if π is a cyclic representation of A, i.e. I∈I π I (A(I)) has a cyclic vector in H π , then H π is separable as a consequence of the fact that A(I) has separable predual for all I ∈ I. Accordingly a representation π of A is locally normal if and only if it is a direct sum of separable representations. The vacuum representation π 0 on the vacuum Hilbert space H π 0 := H is defined by π 0,I (x) = x, for all I ∈ I and all x ∈ A(I) and it is obviously locally normal.
Two representations π 1 and π 2 of the same net are equivalent if there exists a unitary operator u : H π 1 → H π 2 such that uπ 1,I (x) = π 2,I (x)u for every I ∈ I and x ∈ A(I). We denote by [π] the unitary equivalence class of a representation π of A. The representation π is called irreducible, if ∩ I∈I π I (A(I)) ′ = C1 Hπ . The equivalence classes of irreducible locally normal representations of A are called sector (or superselection sector or DHR sector). The vacuum representation π 0 is irreducible and the corresponding sector [π 0 ] is called the vacuum sector.
Let π be a representation of A and let I 1 , I 2 ∈ I be such that I 2 ⊂ I ′ 1 . If I 2 = I ′ 1 then there is an interval I 0 ∈ I containing I 1 ∪ I 2 . Hence π I 1 (A(I 1 )) = π I 0 (A(I 1 )) commutes with π I 2 (A(I 2 )) = π I 0 (A(I 2 )). Moreover, for every I ∈ I, π I is faithful because A(I) is a countably decomposable type III factor and hence a simple C*-algebra.
If π is locally normal then π I (A(I)) is a type III 1 factor with separable predual for all I ∈ I. Moreover it follows easily from the definition and additivity of conformal nets that if I ⊂ β I β , I, I β ∈ I, then π I (A(I)) ⊂ β π I β (A(I β )). As a consequence we also have π I (A(I)) ⊂ π I ′ (A(I ′ )) ′ for every I ∈ I a property which in principle may fail for representations which are not locally normal.
A representation π on the vacuum Hilbert space H is said to be localized in I 0 if π I ′ = id. In this case, by Haag duality, we have π I (A(I)) ⊂ A(I), for all I ∈ I containing I 0 , i.e. π I is an endomorphism of A(I). A representation of A localized in some interval in I is also called a localized endomorphism or DHR endomorphism of the net A.
If π is any representation of A on a separable Hilbert space H π and I 0 is any interval in I then there exists a representation localized in I 0 and unitarily equivalent to π.
A locally normal representation π is always covariant with respect to the Möbius group in the sense that there exists a unique strongly continuous unitary representation U π of the universal covering PSL(2, R) (∞) such that whereġ is the image of g in PSL(2, R) under the covering map, and such that U π (g) ∈ I∈I π I (A(I)), for all g ∈ PSL(2, R) (∞) , cf. [DFK04]. This way, the infinitesimal generator L π 0 of the lifting of the rotation subgroup turns out to be well-defined and positive [Wei06].
The sum of d(π) 2 over all sectors [π] is called the global index of A. The (finite) µ-index of a completely rational net coincides with the global index [KLM01] and hence such a net has only finitely many sectors, each with finite statistical dimension. Finally, for conformal nets, the split property implies equality of the µ-index and the global index, and split together with finiteness of the global index imply complete rationality [LX04].
Many interesting examples of unitary rational chiral CFT models have been proved to correspond to completely rational conformal nets including various loop groups models (e.g. L SU(n) for every positive integer level k) and the corresponding cosets models, CFTs associated to the discrete series representations of the Virasoro and super-Virasoro algebras, the moonshine vertex operator algebra V ♮ and lattice models. The class of completely rational conformal nets is moreover closed under tensor product, irreducible extensions, orbifold construction and finite index subnets, see e.g. [CKL08, DX06, KL04, KL06, Lon03, Xu00b, Xu05].
2 The universal C*-algebra and its image in locally normal representations As mentioned above, the single algebras A(I), I ∈ I, are all isomorphic, so we cannot expect them to carry any information about the net and its representations. In fact, the characteristic properties and the representation structure of a specific local conformal net lie in the family of inclusions between these local algebras. A successful way of capturing this structure is to study a global C*-algebra (see [Fre90,FRS92,GL92] and also [RV10] for a generalization of this notion) with certain universality properties.
Definition 2.1. The universal C*-algebra of A is the unique (up to isomorphism) unital C*-algebra C * (A) such that -for every I ∈ I, there are unital embeddings ι I : A(I) → C * (A), such that ι I 2 |A(I 1 ) = ι I 1 if I 1 ⊂ I 2 , and all ι I (A(I)) ⊂ C * (A) together generate C * (A) as C*-algebra; -for every representation π of A on H π , there is a unique representationπ : . 5] is well-defined, we may construct it as follows: consider the * -algebra * A generated freely by all A(I), let ι I be the natural inclusions, and consider the C*-seminorm where the supremum is taken over all * -representations π of * A such that π • ι I 2 |A(I 1 ) = π • ι I 1 if I 1 ⊂ I 2 (it is easy to show that the sup in the above formula is finite). Then the algebra C * (A) can be taken as the completion of * A/ ker(|| · ||). This is the general way of defining a universal C*-algebra given a suitable family of generators and relations [Bla06, II.8.3.1]. Note that C * (A) can be defined in this way for every Möbius covariant net, yet in the present paper we shall only consider conformal nets.
(2) For convenience, when no confusion can arise, we shall often identify A(I) with its image ι I (A(I)) in C * (A). Moreover we drop the symbol· over π.
(3) We call a representation π of C * (A) locally normal if π • ι I is normal, for all I ∈ I, in other words, if it comes from a locally normal representation (π I ) I∈I of the net A.
We say that π is localized in There is a natural one-to-one correspondence between representations π of C * (A) localized in a given interval I ∈ I and covariant endomorphisms ρ of C * (A) localized in I, i.e. such that ρ • ι I ′ = id A(I ′ ) . Namely, given π, ρ is the unique endomorphism of C * (A) localized in I satisfying π = π 0 • ρ and the covariance condition where z(π, g) is the canonical cocycle associated to π and α : PSL(2, R) → Aut(C * (A)) is the automorphism group corresponding to the Möbius covariance of the net A, see [GL92, Sect. 8] (cf. also [CHL12]). Clearly, the localized covariant endomorphism corresponding to the vacuum representation π 0 of A is the identity automorphism id : C * (A) → C * (A). If ρ 1 and ρ 2 are localized covariant endomorphisms of C * (A) then π 0 • ρ 1 is equivalent to π 0 • ρ 2 iff ρ 1 = Ad(u) • ρ 2 for some unitary u ∈ C * (A) (i.e. ρ 1 and ρ 2 are unitarily equivalent in C * (A)).
The statistical dimension d(ρ) of the localized endomorphism ρ of C * (A) is defined to be the statistical dimension d(π 0 • ρ) of the localized representation π 0 • ρ. Note that if π is any locally normal representation of C * (A) and ρ is a covariant localized endomorphism of C * (A) (not necessarily corresponding to π) then the representation π • ρ is locally normal and the equivalence class [π • ρ] depends on [π] and [ρ] only.
(4) If ρ 1 and ρ 2 are covariant endomorphisms of C * (A) with finite statistical dimension and localized in the same interval I 0 ∈ I then the composition ρ 1 ρ 2 is a covariant endomorphism localized in As a consequence, with the operations the set of equivalence classes of locally normal representations of A with finite statistical dimension becomes a unital semiring (without 0) R A , which is commutative because of the DHR braiding [FRS92,GL96,Reh90] and is equipped with a dimen- namely it is involutive on the equivalence classes. If π 0 • ρ is irreducible then, up to unitary equivalence in C * (A),ρ is determined by the irreducibility of π 0 •ρ and the requirement that π 0 is equivalent to a subrepresentation of π 0 •ρρ. If this is the case, the multiplicity of π 0 as a subrepresentation of π 0 •ρρ is 1. If π 0 • ρ is reducible and equivalent to π 0 •ρ 1 ⊕π 0 •ρ 2 then π 0 •ρ is equivalent to π 0 •ρ 1 ⊕π 0 •ρ 2 . d(ρ) = d(ρ) for any localized covariant endomorphism ρ of C * (A) with finite statistical dimension. We call R A the fusion semiring of A.
Let A be conformal on S 1 with finitely many sectors with finite statistical dimension, s 1 , s 2 , . . . s n . This is the case if e.g. A is completely rational. Suppose that s 1 = [π 0 ] and define· : {1, 2, . . . n} → {1, 2, . . . n} by sj :=s j . Addition and multiplication in R A are related by the so-called fusion rules, namely there are coefficients N k ij ∈ N 0 such that From the properties of the conjugation of endomorphisms and commutativity we have As a consequence we also have In all computed cases the DHR fusion rules of a completely rational conformal net coincide with the Verlinde fusion rules of the the corresponding CFT model, see e.g.
(5) In the following we shall consider the subalgebra Alg(A) of C * (A) generated by the local subalgebras A(I), I ∈ I. Clearly, Alg(A) is a unital norm dense * -subalgebra of C * (A).
(6) Let I 1 , I 2 ∈ I be such that I 1 ⊂ I ′ 2 . If I 1 = I ′ 2 then there is an interval I 0 ∈ I containing I 1 ∪ I 2 . Hence ι I 1 (A(I 1 )) = ι I 0 (A(I 1 )) commutes with ι I 2 (A(I 2 )) = ι I 0 (A(I 2 )) so that we have a weak version of locality inside C * (A). However, it seems to be unknown whether or not ι I (A(I)) commutes with ι I ′ (A(I ′ )) for I ∈ I. Accordingly, the strong version of locality may fail in C * (A). On the other hand, if π is any locally normal representation of A then π(ι I (A(I))) ⊂ π(ι I ′ (A(I ′ ))) ′ for all I ∈ I by additivity, cf. Section 1.
As an abstract C*-algebra C * (A) depends in general on the conformal net A on S 1 . For example if A 1 (resp. A 2 ) is a completely rational conformal net with n 1 (resp. n 2 ) sectors then C * (A 1 ) (resp. C * (A 2 )) is a C*-algebra with n 1 (resp. n 2 ) equivalence classes of irreducible representations on a separable Hilbert space. Accordingly, if n 1 = n 2 , C * (A 1 ) and C * (A 2 ) cannot be isomorphic.
However, as shown by the following proposition the universal C*-algebra C * (A) has various properties which do not depend on the choice of the conformal net A on S 1 . Proposition 2.3. Let A be a conformal net on S 1 . Then the universal C*-algebra C * (A) is unital, properly infinite, has stable rank ∞ and it is generated by projections. It is neither separable, nor simple, nor exact, nor purely infinite.
Proof. Concerning all the relevant definitions and facts about these C*-algebraic properties we have to refer to [Bla06] in order to keep our discussion tight. Moreover, we will use Corollary 2.18 which is stated and proved (of course, independently of this proposition) at the end of this section.
C * (A) is unital by definition. It is nonseparable and properly infinite because it contains the closed unital subalgebras ι I (A(I)), I ∈ I which are nonseparable and properly infinite C*-algebras because they are isomorphic to the type III factors A(I). According to [Bla06, V.3.1.11], a properly infinite C*-algebra has stable rank ∞. Moreover, C * (A) is generated by projections because it is generated by the W * -algebras ι I (A(I)), which of course are generated by projections.
By Corollary 2.18, π 0 (C * (A)) contains K(H), so that C * (A) contains the nonzero closed ideal π −1 0 (K(H)). This ideal is not equal to C * (A) e.g. because K(H) has a trivial intersection with any of the type III factors A(I) = π 0 (ι(A(I))). Hence C * (A) is not simple. As another consequence, the finite projections in K(H) are finite projections in π 0 (C * (A)) so that the latter is not purely infinite, [ Finally, suppose C * (A) were exact. Let I be an interval in I and let A(I) be the corresponding type III factor. Then A(I), being a purely infinite von Neumann algebra, contains a countably decomposable infinite-dimensional type I factor F , see [Bla06, III.1.5.6 (ii)] (we are not assuming the split property). Since exactness is hereditary to subalgebras, ι I (F ) and hence F would be exact. But type I ∞ factors are not exact [Bla06, II.9.6.6], a contradiction. So C * (A) is not exact.
Remark 2.4. For many interesting conformal nets A on S 1 one can use [FRS92, Proposition 5.1.] to show that C * (A) has nontrivial center.
Another C*-algebra which is useful in the study of a conformal net A on S 1 is obtained as the quasi-local C*-algebra C * (A| R ) of the restriction of A to R ≃ S 1 \ {−1}. More precisely let I R = {I ∈ I : −1 / ∈ I}. Then I R is directed under inclusion and C * (A| R ) is defined as the norm closure of the unital * -algebra I∈I R A(I). As shown by the following proposition the quasi-local C*-algebras C * (A| R ) share many properties with the universal C*-algebras C * (A) but there are also various important differences.
Proposition 2.5. Let A be a conformal net on S 1 . Then the quasi-local C*-algebra C * (A| R ) is a simple unital, properly infinite and purely infinite C*-algebra, which has stable rank ∞, real rank 0, it is generated by projections and has trivial K-theory. It is neither separable nor exact. If A 1 and A 2 are conformal nets on S 1 with the split property then Proof. The proof follows from the fact that C * (A| R ) is the direct limit of countably decomposable type III factors, which are simple, nonseparable properly infinite and purely infinite C*-algebras [Bla06, III.1.7.11&p.431]. Moreover, type III factors are not exact being properly infinite von Neumann algebras (cf. the proof of Proposition 2.3 above). The K-groups of a direct limit are the direct limit of the K-groups, which are 0 in the present case [Bla06, V. . . and it is fairly easy to see that the corresponding C*-inductive limit B is isomorphic to B 1 and hence to C * (A 1 | R ). If A 2 is another conformal net on S 1 with the split property then in the same way we see that C * (A 2 | R ) is isomorphic to B and consequently it is isomorphic to C * (A 1 | R ). Note that incidentally we have shown that the above C*-inductive limit B of infinite-dimensional countably decomposable type I factors has all the properties in the proposition. In particular it is a simple and purely infinite C*-algebra.
The properties in Proposition 2.3 are far from being enough to determine the possible isomorphism classes of the universal C*-algebras of conformal nets even if we restrict to the very special (but important) class of completely rational nets. However, as we shall see in the rest of this section we can get much more information for their images π(C * (A)) in locally normal representations π, especially in the cases where the representations have finite statistical dimension. We begin with an abstract result on intermediate * -subalgebras of inclusions of von Neumann algebras having finite index in the sense of Pimsner and Popa [PP86], see [Pop95, Definition 1.1.1], which will play a crucial role in our analysis.
It follows that N 0 must contain some element from the open set of invertibles in N and hence, being a right ideal, it must coincide with N . Accordingly B 0 = B and the conclusion follows.
We now come back to the universal C*-algebras of conformal nets on S 1 .
Proposition 2.7. Let A be a conformal net on S 1 with the split property and let F be a type I factor such that A(I 1 ) ⊂ F ⊂ A(I 2 ) for a given pair I 1 , I 2 ∈ I with I 1 ⊂ I 2 . Then π(C * (A)) ′′ = π I 2 (F ) ∨ (π I 2 (F ) ′ ∩ π(Alg(A))) ′′ for every locally normal representation π of A.
Proof. Let π be a locally normal representation of A, then π I 2 (F ) is an infinite-dimensional countably decomposable type I factor and hence is isomorphic to B(H) where H is the (separable) vacuum Hilbert space of A. Accordingly there exists a Hilbert spaceH and a unitary w : H π → H ⊗H such that wπ I 2 (F )w * = B(H) ⊗ 1. Let I ∈ I be any interval containing I 2 , so that F ⊂ A(I) and π I 2 (F ) = π I (F ). Since π I (A(I)) ′ ⊂ π I (F ) ′ , wπ I (A(I)) ′ w * = 1 ⊗ R for some von Neumann algebra R onH. Accordingly, and by taking the commutants in the above equality we find π I (A(I)) = π I (F ) ∨ π I (F ) ′ ∩ π I (A(I)) ⊂ π I (F ) ∨ π I (F ) ′ ∩ π(Alg(A)) ′′ .
The remaining inclusion is obvious.
Lemma 2.8. Let A be a conformal net on S 1 and let π be a locally normal representations of A with finite statistical dimension. Then the * -algebra π I (A(I)) ′ ∩π(Alg(A)) is σ-weakly closed in B(H π ) for all I ∈ I.
Proposition 2.9. Let A be a conformal net on S 1 with the split property and let F be a type I factor such that A(I 1 ) ⊂ F ⊂ A(I 2 ) for a given pair I 1 , I 2 ∈ I with I 1 ⊂ I 2 . Then the * -algebra π I 2 (F ) ′ ∩ π(Alg(A)) is σ-weakly closed in B(H π ), for every locally normal representation π of A with finite statistical dimension.
Corollary 2.10. Let A be a conformal net on S 1 with the split property and let F be a type I factor such that A(I 1 ) ⊂ F ⊂ A(I 2 ) for a given pair I 1 , I 2 ∈ I such that I 1 ⊂ I 2 .
Then π I 2 (F ) ′ ⊂ π(Alg(A)) for every irreducible locally normal representation π of A with finite statistical dimension.
In order to understand the reducible case we need to consider the center of π(C * (A)).
We are now ready to prove one of the main results of this paper.
Theorem 2.13. Let A be a conformal net on S 1 with the split property. Then π(Alg(A)) = π(C * (A)) = π(C * (A)) ′′ for every locally normal representation π of A with finite statistical dimension.
Corollary 2.14. Let A be a conformal net on S 1 with the split property. Then π(C * (A)) is a direct sum of finitely many countably decomposable infinite-dimensional type I factors for every locally normal representation π of A with finite statistical dimension.
For completely rational nets the assumption of finiteness of the statistical dimension in Theorem 2.13 and Corollary 2.14 turns out to be unnecessary as shown by the following theorem.
Theorem 2.15. Let A be a completely rational conformal net on S 1 and let π be a locally normal representation of A. Then π(Alg(A)) = π(C * (A)) = π(C * (A)) ′′ and moreover π(C * (A)) is a direct sum of finitely many countably decomposable infinite-dimensional type I factors.
Proof. Let π be a locally normal representation of the completely rational conformal net A. Then according to [KLM01,Cor.39] it is equivalent to a direct sum of irreducible locally normal representations with finite statistical dimension. Since there are only a finite number of equivalence classes of these representations it follows that π(C * (A)) ′′ is a direct sum of finitely many type I factors which are countably decomposable and infinitedimensional because irreducible locally normal representations of a conformal net A acts on infinite-dimensional separable Hilbert spaces. Moreover, π is quasi-equivalent to a locally normal representation with finite index and hence π(Alg(A)) = π(C * (A)) = π(C * (A)) ′′ by Theorem 2.13.
If A is a completely rational conformal net on S 1 then Theorem 2.15 gives a complete description of π(C * (A)) for all representations which are relevant for CFT and hence in view of the applications to quantum field theory the result appears to be completely satisfactory. If A is a conformal net which is not completely rational then in order to have a complete description, we can apply Theorem 2.13 if we assume the split property and if we restrict ourselves to representations which are quasi-equivalent to locally normal representations with finite statistical dimension. The split property does not appear to be a restrictive assumption. As far as we know there is no conformal net for which it has been proved that the split property fails. Moreover the standard construction of net without the split property by means of infinite tensor products gives rise to nondiffeomorphism covariant nets as shown in [CW05,Section 6]. On the other hand, although locally normal irreducible representations with infinite statistical dimension are in a certain sense pathological they naturally appear in non-rational conformal net models [Fre95,Car03,Car04,LX04]. For this reason it could be useful to have some information about the structure of π(C * (A)) also in the case in which the locally normal representation π is not assumed to have finite statistical dimension and in the remaining part of this section we will tackle this problem. This will also lead us to Corollary 2.18 which was used in the proof of Proposition 2.3 in order to show that the universal C*-algebra C * (A) of a conformal net on S 1 is neither simple nor purely infinite.
First, we shall show that π(C * (A)) contains "heat kernels". We write L π 0 for the conformal Hamiltonian (the infinitesimal generator of the rotation subgroup of Diff(S 1 ) as explained in Section 1) of A in the locally normal representation π. Since for t ∈ R e i tL π 0 ∈ I∈I π I (A(I)) = π(C * (A)) ′′ , recalling that L π 0 ≥ 0, we also have e −tL π 0 ∈ π(C * (A)) ′′ for all t ≥ 0. Hence, as a consequence of Theorem 2.13, if π has finite statistical dimension then e −tL π 0 ∈ π(Alg(A)) for all t ≥ 0.
If π is irreducible but is not assumed to have finite statistical dimension then it follows from [DFK04] that, for any t ∈ R, e i tL π 0 is a finite product of local unitaries and therefore lies in π (Alg(A)). This result extends in the following way (the split property is not assumed): Theorem 2.16. Let A be a conformal net on S 1 , π a (not necessarily irreducible) locally normal representation of A. Then e −tL π 0 ∈ π(Alg(A)), for every t ≥ 0.
Proof. There are two important ingredients for our proof: one of them can be found in [BDL07], the other in [Wei06]. To keep the argument relatively short, we will extensively refer to those two papers. At the same moment, however, we shall also try to avoid being "unreadable" and recall at least the basic concepts and ideas.
Consider the group PSL(2, R) (2) ⊂ Diff(S 1 ) defined as the set of diffeomorphisms g ∈ Diff(S 1 ) for which there exists a Möbius transformation h ∈ PSL(2, R) so that It is the second cover of PSL(2, R), and as such, it is isomorphic to SL(2, R). Through a lifting procedure, to each one-parameter group of PSL(2, R) one can define the corresponding one-parameter group of PSL(2, R) (2) . Thus corresponding to the usually introduced translations x → τ x and rotations α → R α , one has the one-parameter groups x → τ (2) x ∈ PSL(2, R) (2) and α → R (2) α ∈ PSL(2, R) (2) ; see further details in [Wei06]. Of course, for rotations one has the simple relation R (2) α = R α/2 . Let S 1 ± = {z ∈ S 1 : ±Im(z) > 0} be the upper and lower half-circles, respectively. A "2-translation" τ (2) x leaves the points ±1 ∈ S 1 fixed and hence can be "cut into two": it can be written as the composition of two homeomorphisms localized in the upper and the lower half-circle, respectively. However, these homeomorphisms are not smooth and hence one cannot substitute them in the positive energy representation U given with the conformal net A. Nevertheless, in [Wei06] it was proved that there exist two self-adjoint operators K ± affiliated with A(S 1 ± ) respectively, and bounded from below, such that For every locally normal representation π of A, there exist two unique strongly continuous projective representations U (1) π := U π and U (2) π of PSL(2, R) and PSL(2, R) (2) , respectively, that U (n) π (PSL(2, R) (n) ) ⊂ π(C * (A)) ′′ and Ad(U (n) π (g)) • π I = π gI • Ad(U (g)), for n = 1, 2, g ∈ PSL(2, R) (n) and intervals I ⊂ S 1 . In fact in [Wei06,Prop.3.4], apart from existence, it was also shown that the representations U (n) π are of positive energy type and that for all x, α ∈ R. Here of course the equations are meant in the projective sense, i.e., up to a phase. Note that, by construction, (2.5) is a product of local elements, hence in π(C * (A)). The projective representations U (n) π lift to strongly continuous positive energy representations of the universal covering of PSL(2, R). It is exactly through this lift that the self-adjoint generator of rotations L π 0 ≥ 0 is defined and we have that e i αL π 0 = U (n) π (R α ), α ∈ R, n = 1, 2, in the projective sense.
On the other hand, [BDL07,Th.3.3] applied to the representation U (2) π of PSL(2, R) (2) provides us with a decomposition of e −tL π 0 into the product of 3 operators; using (2.5), by which we can further split each of those 3 operators into a product of two, we finally get that, for all t > 0, whereK ± = U (R π/2 )K ± U (R π/2 ) * are positive operators affiliated to A(i S 1 ± ), the algebras over the left and right half-circle, and the scalar c t ∈ R + depends only on π and t.
If A is a conformal net on S 1 with the split property and π is a locally normal representation of A with finite statistical dimension, then by Theorem 2.11, π(C * (A)) = B(H π ) so that in particular K(H π ) ⊂ π(C * (A)). In the case of infinite statistical dimension we are not able to prove that π(C * (A)) = B(H π ) but thanks to Theorem 2.16 we can still prove that K(H π ) ⊂ π(C * (A)) under the mild assumption that L π 0 has a finite-dimensional eigenspace. Note that, to the best of our knowledge, L π 0 has no infinite-dimensional eigenspace in all studied examples with diffeomorphism symmetry.
Proposition 2.17. Let A be a conformal net on S 1 and let π be an irreducible locally normal representation of A. Assume that L π 0 has a finite-dimensional eigenspace. Then K(H π ) ⊂ π(C * (A)).
Corollary 2.18. Let A be a conformal net on S 1 and let π 0 be the vacuum representation of A on the Hilbert space H. Then K(H) ⊂ π 0 (C * (A)).
Proof. π 0 is irreducible and locally normal. Moreover by the definition of conformal nets on S 1 the eigenspace of L π 0 0 = L 0 corresponding to the eigenvalue 0 is one-dimensional and the conclusion follows from Proposition 2.17.
The locally normal universal C*-algebra
As already pointed out in the introduction the non-locally normal representations do not apperar to have direct rilevance for CFT . Therefore, it seems natural to ask for a more manageable C*-algebra than C * (A) taking account only of the locally normal representations. In this section we define the most natural candidate for this purpose and analyze some of its properties.
Definition 3.1. The locally normal universal representation (π ln , H ln ) of C * (A) is the direct sum of GNS representations π ϕ over all states ϕ of C * (A) for which π ϕ • ι I is a normal representation of A(I), for all I ∈ I. It defines the locally normal universal C*-algebra C * ln (A) := π ln (C * (A)) of A. Remark 3.2. (1) Clearly, C * ln (A) ≃ C * (A)/ ker(π ln ). Consequently, C * ln (A) satisfies the following universal property for locally normal representations: for every locally normal representation π of A on H π , there is a unique representationπ : C * ln (A) → B(H π ) such that π I =π • ι I , I ∈ I, where, with a slight abuse of notation, we have also denoted by ι I the natural inclusion of A(I) in C * ln (A). Accordingly, for most purposes, if one deals only with locally normal representations one can replace C * (A) with C * ln (A). As before then, we shall drop the symbol "· " and identify A(I) with its image in C * ln (A).
(2) If A is completely rational, then according to [KLM01,Cor.39] every locally normal representation of A is equivalent to a direct sum of irreducible locally normal representations with finite statistical dimension. Considering then the reduced locally normal universal representation (π red , H red ) of C * (A), namely the direct sum of a maximal family of mutually inequivalent irreducible locally normal representations of C * (A), we see that π red and π ln are quasi-equivalent so that π red (C * (A)) ≃ π ln (C * (A)).
Moreover, every locally normal representation of C * (A) is quasi-equivalent to a subrepresentation of π red . The reason of considering this reduced representation, in which we get rid of an infinite multiplicity, shall become clear in Theorem 3.4. We write C * red (A) := π red (C * (A)). Based on the new concepts introduced in the present setting, we can reformulate the results from Section 2 as follows: Theorem 3.3. Let A be a conformal net on S 1 . Then the following hold.
(3) Suppose A is completely rational, with n sectors. Then C * ln (A) is weakly closed, Proof.
(1) follows by the same arguments given in the proof of Proposition 2.3.
(2) As a sum of locally normal representation, π ln is locally normal again, so the statement follows from Theorem 2.16.
(3) From the definition of C * ln (A) and Theorem 2.15 we know that C * ln (A) is weakly closed and that Proposition 3.4. Let A be a completely rational conformal net on S 1 with n sectors. Then the following hold.
(1) Every irreducible locally normal representation of C * (A) contains the compact operators. The restriction of representations of C * ln (A) ≃ C * red (A) to the subalgebra of compacts K A := K(H red ) ∩ C * red (A) gives rise to an isomorphism from the category of locally normal (unital) representations of C * (A) onto the category of nondegenerate representations of K A . In particular there is a one-to one correspondence between equivalence classes of nondegenerate irreducible representations of K A and sectors of A.
(1) From Theorem 3.3 (2), we have K A ≃ K ⊕n since H red is a finite direct sum of separable Hilbert spaces. As a consequence, every nondegenerate representation of K A has a (unique) normal extension to K ′′ A = C * red (A) ≃ B(H) ⊕n , and the claim follows.
(2) The locally normal representation π red •ρ is quasi-equivalent to a subrepresentation of π red . Accordingly, there is a unique normal representationρ (corresponding through the quasi-equivalence to the restriction to the subrepresentation subspace) of C * red (A) on H red satisfyingρ (π red (x)) = π red (ρ(x)), x ∈ C * (A), and consequentlyρ(C * red (A)) ⊂ C * red (A). We now show thatρ is injective. Since A is completely rational, the localized representation π 0 • ρ has a direct summand π 0 • σ with σ a localized covariant endomorphism of C * (A) with finite statistical dimension. Letσ be the corresponding conjugate endomorphism. Then π red •σρ contains a subrepresentation unitarily equivalent to π red . On the other hand, the locally normal representation π red •σ is quasi-equivalent to a subrepresentation of π red so that π red •σρ is quasi-equivalent to a subrepresentation of π red • ρ. Accordingly π red is quasi-equivalent to a subrepresentation of π red • ρ. As a consequence the identical representation of C * red (A) on H red is quasi-equivalent to a subrepresentation of the representationρ and hence the latter is faithful.
Let e ∈ C * red (A) be any orthogonal projection with finite-dimensional range eH red . Then e is a finite projection in the von Neumann algebra C * red (A) and henceρ(e) is finite inρ(C * red (A)). If ρ has finite statistical dimension thenρ(π red (C * (A))) ′ = π red (ρ(C * (A))) ′ is finite-dimensional and henceρ(e) has finite-dimensional range. It follows that The nondegenerate irreducible representations of K A are in one-to-one correspondence with the sectors of A; moreover, while C * ln (A) cannot be a good candidate for a Ktheoretical description of the superselection structure of chiral CFTs (see Theorem 4.3), K A will turn out in Section 4 to be fine. In contrast to its nice C*-algebraic properties, however, K A completely forgets the net structure of A, in the sense that it has zero intersection with any local algebra A(I), I ∈ I.
Superselection sectors and K-theory K-theory basics
Let B be a * -algebra. We say that p ∈ B is a projection if it is a self-adjoint idempotent, i.e. p = p * and p 2 = p. We say that two projections p, q ∈ B are (Murray -von Neumann) equivalent if there exists a w ∈ A such that w * w = p and ww * = q. If this is the case we write p ∼ q. We denote by Proj(B) the set of projections of B and by V 0 (B) the set Proj(B)/ ∼ of equivalence classes of projections of B.
Let A be a stably unital C*-algebra, i.e., a C*-algebra with an approximate unit of projections; denote its unitization by A + if it is not unital. While details (and the appropriate definitions when A is not stably unital) can be found in [Bla98] or [Bla06, Sect.V.1], we provide here only the main A)). There is a binary operation (p 1 , p 2 ) ∈ Proj (M r 1 (A)) × Proj (M r 2 (A)) → diag(p 1 , p 2 ) ∈ Proj (M r 1 +r 2 (A)) , which turns V (A) into an abelian semigroup. Then the K 0 -group of A is defined as where the Grothendieck group of an arbitrary additive semigroup H is the group of formal differences of elements of H, i.e.
We write [p] for the element in K 0 (A) induced by a projection p ∈ A.
(K 1 ) Let U r (A) denote the set of unitary elements in M r (A + ) of the form 1 r + x with x ∈ M r (A), and let U r (A) 0 be the connected component of 1 r in U r (A). With the diagonal inclusion u ∈ U r (A) → u ⊕ 1 ∈ U r+1 (A), this gives a directed system, and with the operation (u 1 , u 2 ) ∈ U r 1 (A) × U r 2 (A) → u 1 ⊕ u 2 ∈ U r 1 +r 2 (A), the union of all U r (A) becomes a group. Then the K 1 -group of A is defined as the direct limit K-theory is a functor from the category of stably unital C*-algebras to abelian groups. Moreover, it can be regarded as a cohomology theory on C*-algebras; in particular, it satisfies the additivity property, namely for C*-algebras A, B. It defines an important invariant for the description and classification of C*-algebras.
One might be tempted to apply this property of * -homomorphisms in order to achieve a K-theoretic interpretation of the representation theory of conformal nets on S 1 and its localized endomorphisms. However, since infinite factors have trivial K-theory [Bla98, 5.3. 2. (b), 8.1.2 (b)], as a straightforward consequence of additivity of K-theory and Theorem 2.15 we have Theorem 4.3. Let A be a completely rational conformal net on S 1 . Then K * (π(C * (A))) = 0, for every locally normal representation π of A. In particular, K * (C * ln (A)) = 0. We shall see, however, that one can overcome these problems by considering the C*subalgebra K A ⊂ C * red (A) introduced in Proposition 3.4. The purpose of the following subsection is to show that we get indeed a nontrivial semiring action of the fusion ring R A (cf. Remark 2.2 (4) for definition and some properties) on K * (K A ).
Fusion semiring action on K 0 (K A )
In contrast to the algebra C * ln (A), which has trivial K-theory, we have This follows from additivity because K A ≃ K ⊕n and Note that, in the identification N 0 = V (K), 1 ∈ N 0 must correspond to the class of an arbitrary minimal projection in M ∞ (K) and hence to the class of an arbitrary minimal projection in K. Accordingly K 0 (K) is generated by V 0 (K) through the embedding (in fact surjective) V 0 (K) ⊂ V (K), cf. Remark 4.2. Similarly, V 0 (K A ) generates K 0 (K A ). Actually the n equivalence classes in V 0 (K A ) corresponding to the minimal (one-dimensional) projections in K A generate K 0 (K A ).
By the above facts nontrivial actions on K 0 (K A ) are not a priori excluded. In fact, we have Theorem 4.4. Let A be a completely rational net on S 1 . Then the push-forward of * -homomorphisms of K A gives rise to a faithful semiring action of the fusion semiring R A on K 0 (K A ), namely an injective semiring homomorphism η : R A → End(K 0 (K A )) satisfying η [ρ] = (ρ| K A ) * for every localized covariant endomorphism ρ of C * (A). It corresponds to the regular representation of the fusion algebra associated to R A .
Proof. Every localized endomorphism ρ of C * (A) with finite statistical dimension localized in an arbitrary but fixed interval I 0 ∈ I induces a normal * -homomorphismρ of C * red (A) satisfyingρ • π red = π red • ρ (cf. Proposition 3.4 (2)); moreover, it preserves the subalgebra K A ⊂ C * red (A). Let ρ i , for i = 1, ..., n, be endomorphisms of C * (A) all covariant and localized in I 0 , with ρ 1 is the identity endomorphism id and such that π 0 • ρ i , i = 1, ..., n, are pairwise inequivalent irreducible representations. Accordingly we can assume that We have to show that there is an injective semiring homomorphism η : R A → End(K 0 (K A )), which will then define the action. We proceed in four steps.
(2) Additivity. Since η [ρ] is a group homomorphism, it suffices to check additivity of the action on the generators in V 0 (K A ) of K 0 (K A ). Given [ρ], [σ] ∈ R A , recall from Remark 2.2 (4) that their sum in R A is defined as where λ is any covariant endomorphism of C * (A) localized in I 0 and such that π 0 • λ is equivalent to π 0 • ρ ⊕ π 0 • σ. Then we can write λ = v 1 ρ(·)v * 1 + v 2 σ(·)v * 2 with isometries v 1 , v 2 ∈ A(I 0 ) defining a Cuntz algebra as a unital subalgebra of A(I 0 ). Since, for any i, v i,red := π red (v i ) is again an isometry (now in C * red (A)), given a minimal projection p ∈ K A , v i,red pv * i,red is again a minimal projection in K A having the same central support of p in C * red (A). Accordingly p and v i,red pv * i,red are equivalent in C * red (A) and hence they are equivalent in the ideal K A . Therefore i.e., additivity holds.
(3) Multiplicativity. According to the above general discussion of push-forwards of * -homomorphisms to K-theory, we always have, for [ρ], [σ] ∈ R A : Summing up, η : R A → End(K 0 (K A )) is a well-defined nontrivial homomorphism of semirings.
Let z 1 , z 2 , . . . , z n be the minimal projections of Z(K A ) ordered in the natural way. Accordingly z i is the central support corresponding to the subrepresentation π 0 • ρ i for k = 1, . . . , n. Now letz i ∈ Z(W * (A)) be the central support corresponding to any subrepresentation of the universal representation of C * (A) unitarily equivalent to π 0 • ρ i . Then, ifπ red denotes the unique normal extension of π red to the enveloping von Neumann algebra W * (A) of C * (A) thenπ red (z i ) = z i . More generally, for every representation π of C * (A) on a Hilbert space H π ,π(z i ) is the central support corresponding to any subrepresentation of π equivalent to π 0 • ρ i . Now recalling that for k = 1, . . . , n we havê ρ k •π red = π red •ρ k we can conclude thatρ k (z 1 ) is the central support corresponding to any subrepresentation of π red • ρ k unitarily equivalent to the vacuum representation π 0 . But, by the properties of the conjugate endomorphisms π 0 is equivalent to a subrepresentation of π 0 • ρ iρk iff i = k. Moreover, for i = k, the multiplicity is one. Accordingly,ρ k (z 1 ) is a minimal projection inρ k (C * red (A)) ′ andρ k (z 1 ) ≤ z k . It follows that if p 1 is a minimal projection in K A with nonzero entry in K 1 , i.e. p 1 ≤ z 1 , thenρ k (p 1 ) =ρ k (p 1 )ρ k (z 1 ) is a minimal projection in K A such thatρ k (p 1 )z k =ρ k (p 1 ).
The equivalence class in K 0 (K A ) ofρ k (p 1 ) does not depend on the actual choice of the minimal projection and is consequently precisely the k-th generator of K 0 (K A ) ≃ Z n . In other words, [ρ k (p 1 )], with k = 1, ..., n, is the canonical basis of K 0 (K A ). Notice that involutivity of conjugation on the equivalence classes implies that, for every l there is a unique k such that [ρ l ] = [ρ k ]. This shows that e k := [ρ k (p 1 )] = η [ρ k ] (e 1 ), k = 1, ..., n, defines again a basis of K 0 (K A ), which is just a permutation of the canonical one.
With this choice, η is easily seen to correspond to the regular representation of the fusion rules of R A , namely for i, j = 1, ..., n, we have Here we made use of additivity and multiplicativity of η, and of the fusion matrix introduced in (2.1). In particular, for j = 1 and arbitrary [ρ] ∈ R A , which can always be uniquely decomposed as n i=1 m i [ρ i ] with certain m i ∈ N 0 , we find η [ρ] (e 1 ) = i m i e i , so the map [ρ] ∈ R A → η [ρ] ∈ End(K 0 (K A )) is actually injective. Now letR A be the Grothendieck group of the additive semigroup R A . From the above discussion we see that the the map [ρ] → η [ρ] (e 1 ) is a semigroup isomorphism β of additive semigroups from R A into K 0 (K A ) whose extension toR A gives rise to a surjective group isomorphismβ :R A → K 0 (K A ) and the action η of R A uniquely extends to an actioñ η :R A → End(K 0 (K A )). Now, C ⊗ ZRA is a fusion algebra with distinguished basis 1 ⊗ Z [ρ 1 ] . . . 1 ⊗ Z [ρ n ], see e.g. [Yam99] for the definition, and id ⊗η is a representation of C⊗ ZRA on the n-dimensional complex vector space C⊗ Z K 0 (K A for all x, y ∈ C ⊗ ZRA , namely id ⊗ Zβ −1 (id ⊗ Zη ) id ⊗ Zβ is the regular representation of the commutative fusion algebra C ⊗ ZRA .
The group isomorphismR A ≃ K 0 (K A ) might be used in order to define a fusion ring structure on K 0 (K A ). Obviously this structure is not encoded in the C*-algebra K A on its own but depends on the underlying net A through the action on K A of the DHR endomorphisms. If A 1 and A 2 are two completely rational conformal nets with the same number of sectors then K A 1 and K A 2 are isomorphic C*-algebras but typicallyR A 1 and R A 2 are not isomorphic fusion rings.
|
2012-09-06T21:53:48.000Z
|
2012-02-12T00:00:00.000
|
{
"year": 2012,
"sha1": "209a425692ce517b414ab04b3bc9f1fb5ab7ab47",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/1202.2543",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "209a425692ce517b414ab04b3bc9f1fb5ab7ab47",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics",
"Physics"
]
}
|
259709505
|
pes2o/s2orc
|
v3-fos-license
|
Teleconnection between the Low Index Phase of Southern Oscillation and Precipitation Patterns over the Southeastern United States
: The purpose of this analysis is to examine the spatiotemporal extent to which the extreme phase of Southern Oscillation affects precipitation patterns over the southeastern region of the United States. Composite and harmonic analyses using monthly precipitation data were conducted based on the 2-year composite resulting in harmonic dials, which represent the spatial coherence of the candidate regions. For the candidate regions identified in the harmonic dial maps, aggregate composite and index time series of precipitation were analyzed to determine the core regions which represent the temporal consistency in relation to the low index phase of Southern Oscillation (LISO) events. As a result, the precipitation patterns of the core regions, Upper Region (UR), Middle Region (MR), and Lower Region (LR), were influenced by the occurrence of the extreme phase of SO events, with LR having the greatest response. During the LISO cycle, the harmonic dial tends to face southeast in UR, northeast in MR, and east in LR. Furthermore, the spatial coherence rates of the three core regions range from 0.94 to 0.99, and the temporal consistency rates are 0.76 to 0.83. In addition, from the cross-correlation and annual cycle analyses, the seasonal and annual effect of LISO forcing on the precipitation variability over the three core regions can be recognizable. In conclusion, Southern Oscillation phenomena have a great influence on the precipitation patterns over the southeastern region of the United States.
Introduction
The Southern Oscillation (SO) is one of the well-known climate indicators, which represents large-scale fluctuations in sea level pressure over the western and eastern Pacific Ocean. These large-scale naturally occurring phenomena have been investigated at regional and global scales since the extreme phases of the SO episodes can cause major hydrologic extremes in various parts over the world. Many scientific approaches for understanding these phenomena have been providing us a chance to prepare for the disastrous hazards such as torrential rainstorm, floods, and droughts.
As a pioneering research, Walker and Bliss [1,2] studied firstly the impacts of the Southern Oscillation (SO) on the Indian precipitation variability. Since then, a number of global scale studies related to the SO extreme phases showed various notable climatic links between precipitation patterns and SO extreme events in many areas. Since Berlage [3] found the SO extreme events correlated well with precipitation anomaly on a global basis, Rasmusson and Carpenter [4] related the precipitation and temperature patterns to the extreme phase of Southern Oscillation and identified a significant link between the two variations. Ropelewski and Halpert [5,6] investigated temporal and spatial ranges showing consistent monitors and operates the overall United States meteorology. As shown in Figure 1, the monthly precipitation data range from 1895 to 2020 and cover the overall 29 LISO episodes. To identify a consistent far-reaching effect of LISO events on precipitation anomaly over the southeastern United States, a set of extreme LISO episodes is selected considering a comprehensive scope of criteria defined by Ropelewski and Halpert [5,6], Rasmusson and Carpenter [4]. The event years for the extreme southern oscillation selected in the present analysis are listed in Table 1. Southern Oscillation Index (SOI) is applied as an indicator of representing large-scale climate variation over the central-eastern Pacific Ocean (ENSO). This present analysis applied the SOI data calculated and recorded by the NOAA Climate Prediction Center. These SOI time series are computed based on the difference of the standardized sea level pressure (SLP) anomaly between Tahiti and Darwin, Australia. 1905,1911,1914,1918,1923,1925,1930,1932,1939,1941,1951,1953,1957,1963,1965,1969,1972,1976,1982,1986,1991,1994,1997,2002,2004,2006,2009,2015,2018.
Method
The methodological approach for un¬derstanding teleconnection patterns consists of empirical orthogonal teleconnection [30], cross correlation analysis [20], composite analysis, and harmonic analysis [9,15,31,32]. Also, Kalaycı et al. [33] employed spectral analysis to identify the regional-scale interannual variability of precipitation and streamflow and some periodicities associated with the warm phase of Southern Oscillation. For investigating the spatiotemporal extent to which LISO affect precipitation patterns over the southeastern of the United States, an empirical method [9], annual cycle and cross correlation analyses are employed with some changes and additions. As shown in Figure 2, the specific procedure of this analysis consists of mainly three steps, namely, data processing, spatial and temporal analyses, and comparative statistical assessment. In the first step, the original raw data are transformed into appropriate data formats, e.g., ranked percentile, modular coefficients, and categorized SOIs. In the second step, candidate and core regions are determined using composite and harmonic analyses. Then, in the last step, LISO-related precipitation signals are compared using lag cross-correlation and annual cycle analyses. Monthly precipitation time series are transformed into modular coefficients for carrying out annual cycle analysis. These modular coefficients remove the effects of dispersed variance and mean values. The precipitation data are expressed as percentages for the annual mean values. The modular coefficient data are calculated by the rate of each monthly precipitation value to the monthly average value for the entire data. It places all divisions on a same cycle with unchangeable condition of the cyclic feature of the values at the same time. In this present study, lag cross-correlation coefficients are computed for precipitation data and categorized SOI time series on a seasonal basis. To do this, four seasonal precipitation and SOI time series are formed by averaging three-month values. The four seasons consist of December-February (DJF), March-May (MAM), June-August (JJA), and September-November (SON). Then, all SOI values are categorized into five levels based on the magnitudes of individual data [20]. The five categories of the SOI values are strong warm phase, weak warm phase, normal phase, weak cold phase, and strong cold phase. On the other hand, the seasonal precipitation data are converted into percentile ranked probability time series to remove periodicities in precipitation time series and to deal with the disparities among climate divisions. The percentile ranked probability values are based on Weibull plotting position formula. All precipitation values are ranked in ascending order and then divided by n+1 (n: size of data). Monthly precipitation percentile rank composites on 24-month basis are computed for each climate division, starting with the July preceding the event, continuing through the June following the event year, for the LISO. The July preceding the event is designated as Jul (-), while the June following the event year is expressed as Jun (+) as shown in figure 3. The composite for each climate division is then fitted with the first harmonic of an idealized 24-month LISO cycle (either warm or cold episodes). This method assumes one precipitation peak (or trough) during the duration of a LISO event and that the LISO is phase locked to the annual cycle. A 24-month compositing period was chosen since this defines the period during which one phase of the LISO goes through its entire cycle [4]. In the first harmonic cycle, the amplitude represents magnitude of the LISO-related precipitation signals, and the angular phase indicates time of the peak anomaly from the mean value ( Figure 4). The formula of the harmonic fits are as follows [34].
where y t is monthly precipitation value, y _ is the mean precipitation value, t is time of observation, i is number of harmonic fits, n is sample size, C i is amplitude of the harmonic curve (magnitude of curve), β i is time of harmonic peak (temporal phase of curve), and A i and B i are Fourier coefficients. After the climate division composites are fit with a 24-month harmonic, the amplitude and phase of the curve is plotted as a vector for each station. In the analysis convention chosen here, the vector points toward the positive part of the cycle, that is, wetter-than-normal precipitation. It is only after examining the composites, described below, that the actual sign of the LISO-precipitation relationship can be determined. This study is concerned with regional areas of the southeastern United States that exhibit strong LISO-precipitation relationships over periods of many months. Therefore, individual, or isolated, climate division that shows strong apparent relationships or areas that have short-time scale relationships are not considered for further study.
Plotting the harmonics as vectors on a map provides a method to spatially identify candidate geographic areas that appear to have a coherent LISO-precipitation response. We attempted to choose the largest areas of coherent LISOprecipitation response, where the "coherence" is estimated through the computation of the ratio of the magnitude of the average vector to the arithmetic average value of the vector magnitudes.
where the numerator is the average vector magnitude of all harmonic vectors within the candidate regions, the denominator is the arithmetic average value of the vector magnitudes, θ is angle of the vector, and V is magnitude of the vector. The analysis that follows is limited to areas for which values of the coherence were equal to or greater than 0.80 [9]. This eliminates from the analysis regions that contain harmonic vectors with large amplitudes at a few stations which have little consistency in phase, i.e., low coherence. Aggregate composites are formed to detect the LISO-related precipitation signal seasons. These signal seasons represent apparently consistent precipitation responses to the extreme LISO forcing. All LISO composite values in a candidate region are averaged and plotted on a 24-month period to cover the entire LISO cycle and identify accurately the signal season. One season within the aggregate composite is found by detecting a group of values showing more than five consecutive months with the same sign. The event year and the following year are regarded as the responding period of LISO phenomena, considering the distance between the study area in midlatitude and the LISO area in the Pacific Ocean.
Index Time Series (ITS) are computed by temporally averaging precipitation values of the signal seasons for the entire years of record and by spatially averaging the precipitation data over the candidate regions. The ITS values are used to quantify the temporal consistency of the LISO impact on precipitation patterns. Temporal consistency rates for the candidate regions are computed using the rate of the number of years exhibiting LISO signal in ITS to the number of all LISO event years. These temporal consistency rates are the determinant of the core regions showing consistent precipitation responses to LISO phenomena. In addition, extreme precipitation events are examined in association with LISO forcing as demonstrated by Ropelewski and Halpert [9]. They investigated climate linkage between LISO events and extreme precipitation occurrences. In the present study, the number of years showing the LISO-related extreme precipitation signal is counted during the signal season. To assign the highest and lowest levels of the extreme events, the ITS time series are ranked from the highest value to the lowest value, normalized by the entire data, and transformed to the probability time series [15]. The highest value is assigned to the probability of 80% ITS, while the lowest value is assigned to the probability of 20% ITS.
Hypergeometric distribution test is employed to assign the significance level of the LISO-precipitation correlation. Haan [35] conducted the hypergeometric distribution test calculating "the cumulative probability that at least m successes are obtained in n trials from a finite population of size N containing k successes". A cumulative probability computed from the hypergeometric distribution gives an occurrence significance level of the relationship previously defined for both extreme phases of the LISO. Kahya and Dracup [15] used hypergeometric test in terms of average value and high-low extreme events. In the present test, two cases (I and II) are considered according to the definition of a success. In case I, a success is defined as the occurrence of years that an ITS value associated with LISO events is lower than the median, while in case II, a success is defined as the occurrence of year in which an ITS values associated with SO events falls into the lower 20% of the distribution.
Annual cycle analysis is used as a comparative interpretation of two LISO effects on the precipitation anomaly from the perspective of magnitude and annual trend of the signal. Monthly precipitation time series are transformed into modular coefficients for carrying out annual cycle analysis. These modular coefficients remove the effects of dispersed variance and mean values. The precipitation data are expressed as percentages for the annual mean values. The modular coefficient data are calculated by the rate of each monthly precipitation value to the monthly average value for the entire data. This annual cycle plots make it possible to determine whether the extreme phases of SO events modulate precipitation increasingly.
Cross-correlation coefficients are calculated on a seasonal basis to compare the positive and negative LISO-related precipitation signals. Five categorized SOI data sets as SO index are correlated with the monthly precipitation time series expressed by percentile ranked probability. The resulting correlation coefficient values indicate the magnitude and sign of the relationship between the precipitation patterns and the LISO forcing on a seasonal basis. More detailed explanation of the data conversion and correlation procedure was described in the first section of data processing.
Precipitation response to LISO
For 29 LISO episodes, the composite and harmonic analyses were performed on the monthly precipitation data. The vectorial map for precipitation indicates three regions of the southeastern United States that appear to have a coherent LISO response. The candidate regions are the North Region (UR), the Middle Region (MR), and the South Region (LR). As shown in Table 2, composite precipitation indices for each region indicate that the UR, MR, and LR regions may have a LISO-related response. The precipitation composites for the three regions show clearly defined wet season within the LISO cycle and thus they are explained in detail for further consideration in this analysis. The time series of the November (0) to March (+) precipitation indices averaged over all stations in the UR region ( Figure 5) illustrates the notable consistency in the precipitation with respect to the LISO in this part of the southeastern United States. This season showed above normal, i.e., greater than the average precipitation for 22 out of 29 LISO events. Further, of the 25 occurrences of index values equal to or greater than the highest ITS limit (80%), 11 of them occurred in association with LISO ( Figure 6). Only one of the occurrences in the lowest ITS limit (20%) was associated with LISO. The spatial coherence and temporal consistency were 0.94 and 0.76.
Time series of the MR region precipitation index based on the climate division data for the September (0) to January (+) season ( Figure 5) shows above normal precipitation for 22 out of the 29 LISO events. While the index shows values of greater than or equal to the highest ITS limit (80%) for 10 of the LISO years, positive values of the same magnitude or greater also occur during 14 non-LISO-related years ( Figure 6). Only one of the LISO-related seasons in the MR area fall into the lowest, or driest, ITS limit (20%). The spatial coherence and temporal consistency were 0.94 and 0.77.
Time series of the precipitation index for the LR region, November (0) to March (+) in Figure 5, shows another example of conclusive result. 24 out of 29 LISO seasons are associated with above median precipitation in the time series based on the climate division data. Since the probability of getting 24 observations of the same sign strictly by chance is relatively low, the result is statistically significant. The precipitation index exceeded the 80% index value with 13 LISO episodes, and none of the occurrences in the lowest ITS limit (20%) was associated with LISO ( Figure 6). The spatial coherence and temporal consistency were 0.99 and 0.83. Thus, it appears that LISO is a reliable discriminator of the precipitation anomalies of the LR region.
Seasonal and Annual Variation
The probability that wet (dry) season occurs at random during the LISO event years was tested by the hypergeometric distribution (Table 3). In case I, the application of the hypergeometric distribution model results in a very low level of probability of occurrence by chance (less than 0.004) for LISO events. In case II, the probability is also very low for the LISO events. The extreme wet precipitation conditions appear to be almost exclusively related to LISO events in the 125-yr period. Overall results in Table 3 are also consistent with the high confirmation rates (76-83% in Table 2) for temporal consistency of the signals. All of this implies that the relationship depicted in the aggregate composites is probably due to nonrandom forcing mechanism, i.e., the tropical thermal anomalies.
Monthly precipitation time series are transformed into modular coefficients for carrying out annual cycle analysis. From this resulting series, LISO composites are formed and plotted along with the regional annual cycle (Figure 7). For a typical precipitation behavior during LISO events, a suppressed precipitation of the annual cycle during the event year is followed by an enhanced precipitation of the annual cycle from the end of event year to the beginning of the following year. This enhancement of magnitudes is roughly concurrent with the previously detected wet signal seasons in three core regions. In summary, the resulting findings suggest that the tropical heating anomalies modulate the annual precipitation cycle within the southeastern United States by increasing. Table 4 displays the results of calculating cross-correlation coefficients. These values represent intensity and sign of the correlation between the LISO phenomena and precipitation anomalies. This correlation analysis is conducted for large-scale climate indicator (LISO index) and the seasonal precipitation anomalies, which use five categorized SOI data sets and percentile ranked probabilities, respectively. As a result, the seasonal precipitation anomalies were significantly correlated with both extreme phases of SO at 0.05 significance level. The highest positive correlation coefficient values are shown in the lag-2 and lag-3 cases over the three core regions for the strong warm phase SOI condition. For the strong cold phase SOI condition, the highest negative correlation coefficient values are found at all core regions with lag-2 and lag-3 cases. As a result, the stronger warm and cold phases of SOI, the more and less precipitation with lag time 2 and 3 seasons over the southeastern United States.
Discussions
As shown in Figures 4 to 6, the results of this study show positive precipitation response to the LISO events at the UR, MR, and LR regions during early fall (0) to spring (+) seasons. Especially, amplitude of the positive precipitation departure of LISO year at the SR region is even higher than that of non-LISO year. The precipitation response in the southeastern United States to the southern oscillation is consistent with Ropelewski and Halpert [9], Douglas and Englehart [8]. Ropelewski and Halpert [9] stated that the precipitation response to SO may be more easily explained in terms of direct or shorter range effects related to the enhanced subtropical jet stream and warmer than normal surface water over the Pacific. Douglas and Englehart [8] suggested that the SO-related precipitation signal may be an indication of a more direct link to SO forcing than a Pacific North American teleconnection pattern (PNA). Active SOrelated convection is typical in the equatorial Pacific, south of the southeastern United States. This convection has been linked to stronger than normal westerlies in the southern parts of the United States including Gulf of Mexico (e.g., 200 mb SO composites in Arkin [36]) and, hence, a tendency for more frequent storms and precipitation in the southeastern United States. This possible direct link to the SO-related forcing may account for the consistent precipitation response over the southeastern US. Rasmusson and Wallace [37] found a strengthened subtropical jet stream displaced southward from its normal position during the mature phases of SO event (1982)(1983). This pronounced intensification of the jet stream drove numerous winter storms causing flooding events in the southern parts of the United States. Additionally, they indicated that this region has shown abnormal wet conditions associated with past SO events. During the SO event years, the persistent occurrence of warm and cold sea surface temperatures over the central/eastern equatorial Pacific triggers large-scale atmospheric fluctuations in the middle latitude based on complex air-sea coupled interactions. As a result, these LISO-related middle latitude circulations excite abnormal precipitation patterns over the southeastern United States.
Ropelewski and Halpert [9] documented the climatic links between the extreme southern oscillation and precipitation anomalies over North America and showed the SO-related precipitation signals. The analysis showed that above normal precipitation was associated with ENSO in the 18 out of 22 cases (81%) in the "season" starting with October of the event year to March of the following year for an area of the southeastern United States and northern Mexico. Their core region is consistent partially with the Lower Region (LR) in the present study. The present study showed that time series of the precipitation index for the LR region, November (0) to March (+) and 24 out of 29 events were associated with above median precipitation in the time series based on the climate division data. The spatial coherence and temporal consistency were 0.99 and 0.83.
In addition, Kahya and Dracup [14] examined the relationship between the extreme phase of Southern Oscillation and unimpaired streamflow over the contiguous United State. They found that the southern US region shows statistically significant correlation between tropical thermal forcing and streamflow patterns. The signal season of December (0) to April (+) was considered as a time period for strong teleconnection and 6 out of 9 event occurrences (67%) confirmed the wet season. The spatial coherence rate was 0.88. The signal season and spatial pattern are consistent with those of the LR region in the present study. The signal season for the above normal precipitation was November of the event year to March of the following year and 24 out of 29 event occurrences (83%) confirmed the wet signal season. The spatial
|
2023-07-12T08:32:15.162Z
|
2022-12-23T00:00:00.000
|
{
"year": 2022,
"sha1": "64d30ad36f261b0c57ed8412c4c8df8b1dfe3c94",
"oa_license": "CCBY",
"oa_url": "https://ojs.wiserpub.com/index.php/UJCE/article/download/1767/1200",
"oa_status": "HYBRID",
"pdf_src": "Anansi",
"pdf_hash": "4af687de0510d0b828b93fa0e299e842797fcba4",
"s2fieldsofstudy": [
"Environmental Science"
],
"extfieldsofstudy": []
}
|
214323310
|
pes2o/s2orc
|
v3-fos-license
|
Safe Space For Urban Muslim’s Society
This study is about the increase in Muslim housing complexes in Jakarta, which is marketed to specific targets. Those houses are exclusively designed for Muslims, to provide them with a house and a community to ensure they can raise a good and blessed family. Although the development of Islamic housing complexes can be traced to the 1990s, they are mostly found near the Islamic boarding schools or Islamic centers. Today, we find dozens of Muslim housing complexes built in Jakarta’s suburbs such as in Depok, Tangerang, and Bekasi. The settlements are characterized by one entrance gate, fenced with concrete wall, and are commonly known to form gated communities. It is worthwhile to learn more regarding the reasons and processes used to create Muslim gated communities. Experts usually relate this gated community with fear and threats a community has to face in urban life. Although Islam is the majority in Jakarta and Muslim solidarity is currently growing stronger, our inquiry is to focus on the reasons individuals would live in closed and exclusive housings. In addition, we would like to know the Muslims who choose to live there and what type of community is formed by this process of fencing their life so that they are separate from other groups in the city. We also found that women play a significant role to construct the safe space perception for the urban middle-class Muslim’s movement. Women are also an active agent for social movements based on Islam as way of life. Using an ethnographic approach, we interview and observe one selected housing complex in Jakarta, namely, “Cahaya Besar.” Although there is a tendency that new houses in Indonesia are usually associated with a new family, we found that a Muslim housing complex is today a preference of young Muslim couples. As the price of the houses is competitive, we also would like to describe the dynamic of the middle class Muslim today. Keywords—urban space, housing, muslim, gated
I. INTRODUCTION
Today, Indonesian people seem to be facing a phenomenon that the Muslim lifestyle is becoming a trend and moving dynamically to follow the changing times. If in the 1990s the concept of haram means something were forbidden to pigs, dogs, and alcohol, to be consumed it seems that the era of the 2000s began to increase the awareness to search for halal-verified products ranging from food to cosmetics. Additionally, there has been an emergence of various Islamic banking products, Islamic insurance, and even Muslim housing, which began to increase as discussed in this study. Departing from this phenomenon, we were finally interested in conducting research and raising the theme of Muslim housing trends in Indonesia. The name of Cahaya Besar housing and the respondents in this study are not real names. We disguise the name of the housing and respondents to maintain confidentiality. Noorhaidi (2008) described the phenomenon of the commodification of Islam, which has become a relevant symbol of the elite class as an expression and representation that links aspects of consumption with identity. The study of Muslim housing, in our opinion, has the inherent social status implications of an individual's choice of occupancy on the basis of Muslim identity preferences, which also relate to financial resources such as access to lifestyle.
"Islam appears to be no longer simply a set of rituals, beliefs and doctrines, but it is also a symbolic commodity relevant to social class demands for lifestyle, modesty and enjoyment." (Noorhaidi, 2008, p. 241) The Post-Islamist Phenomenon of Piety from Bayat (2002) also explained the strengthening of Muslim middleclass groups that looked different from other Islamic groups because of wider access to resources. Bayat also introduced the concept of "moral outrage" in response to this group in the face of secular Western domination.
The political class par excellence remains the educated middle class-state employees, students, professionals and the intelligentsia-that mobilized the "streets" in the 1950s and 1960s with overarching ideologies of nationalism, Ba'thism, socialism and social justice. Islamism has been the last of these grand worldviews. With the core support coming from the worse-off middle layers, Islamist movements succeeded for three decades in activating large numbers of the disenchanted population with what I like to call cheap Islamization, that is, by resorting to the language of moral and cultural purity (e.g., by calling to restrict alcohol, "immoral" literature and raising issues with women's public appearance), appealing to identity politics, promising a highly general utopian future and carrying out affordable charity work. (Bayat, 2002, p. 16) Noorhaidi (2008) also agreed that although many sociopolitical conditions have favored the Islamic group, it still seems that a fear of a dark outside world attempting to control the use of resources has implications for the value of piety in their social status.
"Being marked and framed by the taste and style of the rich and kept away from the traditional mosques located in the lower class area, it enables the upper middle class to be pious while maintaining their power and prestige" Noorhaidi (2008:239) This Muslim housing is a symptom of a community group that has power over spatial arrangements based on social status. The emergence of this urban young middleclass group, mentioned by Noorhaidi (2008), is a shaken-identity phenomenon; according to Ariel Heryanto, it is the fulfillment of the need to seek identity and cannot be separated from the wave of modernization and globalization.
Kelas menengah muda perkotaan, mencoba merumuskan ulang identitas mereka pada dekade pertama abad 21. Ini adalah masa yang tak terduga, penuh dengan janji akan kebebasan tapi juga, pada saat yang sama, ketakutan. Masa ini juga ditandai oleh beberapa hal seperti: peningkatan politik islami yang belum pernah sedasyat belakangan ini, perdebatan publik tentang pelanggaran hak asasi manusia di masa lalu, perpecahan yang berkepanjangan dan tak terdamaikan di kalangan elite politik, bangkitnya kekuatan ekonomi Asia, serta revolusi komunikasi digital yang disambut secara bergairah oleh kaum muda di seluruh dunia. (Heryanto, 2015:1) Before discussing the gated community (GC) concept, we first identify residents of Muslim housing that cannot be separated from the urban Muslim movement. Geertz's contribution to the results of research in a small town, Modjokuto, does not directly understand the phenomenon of Muslim GCs but can help illustrate the characteristics of urban Muslims that emerge from the profile of Muslim housing residents. Notably, the Geertz study also drew harsh criticism from Harsja Bachtiar through his comments in 1973 because abangan, santri, and priyayi were considered inappropriate because of their different classifications, and because Geertz asserted the absoluteness of the three. Criticism of the Geertz study, which was later explained by Parsudi Suparlan, became a tool to identify these Muslim community groups who deliberately sought housing in Muslim GCs.
The community formed as a result of the gates and is defined as follows: A group of people who live in a GC who are united by the equality of preferences from various backgrounds. The community interprets the gated conditions as representing their aspirations for religious life in the housing they inhabit. Notably, women have contributed to the development of religious-based housing. This study aims to describe human relations with their physical environment, which is the main theme of the discussion regarding GCs.
GCs presented by Low (2017) provided inspiration for the development of the study of urban communities regarding communities who chose to live in a type of housing with limited access. The relationship between the community and the physical environment is the core of GC studies. In a previous study, the topic of social control in modern urban areas became relevant but in anticipation of crime and in the pursuit of community order. GCs are an topic that intersects with territories, which Brower (1980) connected with spatial behavior that aims to achieve the social goals of a group. The contribution of knowledge and experience that is built or built into human interaction with cities-both physical and nonphysical-has implications for the realization of the architecture of fear. The statement about GCs as a response to "fear" and the trend of middleclass housing against the backdrop of different community contexts resulted in our interest in exploring the concept of GCs within the context of an Indonesian Muslim majority.
A discussion on Muslim housing cannot be separated from the urban Muslim community inhabiting the housing.
They also later became part of the Islamic resurgence study that we put in the context of Muslim housing in the city 1 . Although the topic is important, the study of Muslim housing focuses more on groups that deliberately separate themselves from other Muslim urban groups and then become their own community through housing. What we found in this study is that mothers have a role in the formation of the safe space because of the future of their children.
II. METHODS
Roitman, Webster, and Landman (2010) wrote an article on research methods regarding their experience in conducting a GC study. This community intentionally closed itself by imposing restrictions on access to prevent foreigners from entering their environment.
Generally, they can be understood to be closed urban residential schemes voluntarily lived in by a homogeneous social group where public space has been privatized, restricting access through the implementation of security devices. They are designed with the inten-tion of providing security for their residents and preventing penetration by non-residents. (Roitman et al 2010) According to them, to conduct closed community research, several methods can be used, such as Accessing the GC Using the help of people who work in the GC Paying attention to social practices Interpreting community members' preferences and perspectives The city described as an arena in which the processes of all activities and information occur so rigid and complement each other. The city is not just a place but a space, a part of modern society with the availability of spot mobility and interaction that allows members to create communities. Based on this consideration, it seems that only with a qualitative approach that emotions can be captured carefully to observe the possibility of the emergence of variations that occur. Referring to what was conveyed by Paul Ten Have (2004), the patterns that were observed were not numerical but a dialectic between us and the participants studied, and the evidence was presented in the context of a theory and the reality.
Low (2017) stated that ethnography is closely related to the study of space. Early ethnography, for example, uses space as the location of a culture. The next development of space (as a built environment) is part of social life, especially regarding daily activities. The idea he offered was that space is an embodied place. The space then becomes a translocal and transnational site that can be physical, metaphoric, or discursive and identified by individuals and groups.
As with other ethnographic studies, the process of collecting data through in-depth interviews and observations involved is carried out to capture the meaning of a social system. At the time of the fieldwork, our basic assumption was that the community was different and bound and had the same character, language and moral code. For this reason, our fieldwork became a social research process when we interpreted and participated in their moral system (the world of morals and meaning) by entering the community, which is also called "immersion" (Wax & Rosalie, 1980). We made limited observations to obtain a daily picture that was reflected in the activities carried out in the Muslim housing. This limitation has several causes, for example, this activity was a "membership" as the consideration of an outsider could be harmful for the community; the informant's perspective for the researcher whose has role as a mother who leaves their children at home for work and college activities resulted in limited observation at night that were intended as an immersion process; and also we did not have the freedom to interview individuals who were not our mahram 2 without being accompanied by trusted third parties.
We also went to several locations that they used to come and used their attributes to attending the event. Fieldwork is also described in this article as "observer partisipan" and began with our acceptance into the group that demands obligations as a form of commitment to the conception of equality and identity; this means that the intention of our participation was to assume a role of responsibility in the local community.
This strategy allowed us to begin understanding how the community defined its activities. Although, in the beginning, we were often met with rejection, we were finally able to obtain a possible interpretation to understand the group we were researching. Clearly, at this stage, interactions develop between the other and the self in the field. The location of the housing was selected to be able to assess the relationship between fear of the city and occupancy preferences, and in this paper, we present a new housing community that was no more than 10 years old and was inhabited by no more than 50 households: the Cahaya Besar Muslim Housing (not its real name).
The informants interviewed included several categories: (1) residents who were earlier living in housing, (2) residents who were active in activities in the community, (3) residents who withdrew from routine activities in the community, (4) individuals who were not housing residents but active in the community (e.g., ustadz 3 and area'sdministrators who reside in outside housing), and (5) residents who previously did not strictly enforce Islamic law (e.g., wearing a hijab).
A. Profile: Cahaya Besar Housing
Cahaya Besar, which means great light, is an initiative of Pak SD, who has a vision and mission regarding how this housing can become a comfortable residence for Muslims to raise a family. This housing is not offered commercially, and finally, after 3 years, housing was sold through only limited networks (eramuslim.com) and small banners in front of the housing. The names of the streets are chosen by PakSD and directly registered by the developer to the kelurahan and subdistrict during the UN submission process. The names of the blocks in the selected Cahaya Besar housing community havethe expectation value of the Pak SD idea as follows: 2 Mahram is unmarriageable kin with whom marriage or sexual activities 3 Ustadz term means chaplain, priest, pastor in Muslim community Firdaus means heaven An Naim means goodness, comfort, and tranquility Reyhan means a gift from God Salsabila means spring in heaven Al-Kautsar means lake in heaven The concept of heaven is based on physical infrastructure and the occupants; thus, this Muslim housing, from the beginning, was committed to how prospective residents of the Cahaya Besar Muslim housing could include prospective residents of heaven.
Even though it was built by a public developer, then I said that I wanted to establish an Islamic housing, not a house but in the form of a resident, I initially asked for a 500 m house standard but the market did not accept." (interview with Pak SD, 2016) Thus, although the market was difficult because it takes a long time to market this housing product (more than 3 years), the process carried out is a screening of the prospective buyers of Muslim homes in Besar Cahaya. The main criterion is to be Muslim and sign an agreement made at the beginning of the purchase. Notably, the existence of this agreement was conveyed to us by several informants, the agreement was stored at the elementary school, and the contents of the agreement were about the willingness to work together to create neighborhoods based on Islamic values.
At first I did not know what the character of Islam was, but first I had made a contract, we tried a communication forum, recitation, arisan, iftar together and I saw in this place I was a neighbor, I felt that there was peace and security even the security guard just as a symbol. (interview with Pak SD, 2016) Pak SD added that the purpose of the contract was to build the character of Cahaya Besar housing as a Muslim housing community. The contents of the agreement letter include the following: Directly build a house within a maximum period of 3 months from the purchase because the construction could disturb other residents It is not permissible to rent or sell a house to anyone because the individual must be in accordance with the Islamic environment that the community wants to create Pak SD stated that owners do not have to wear headscarves. However, one time, a prospective buyer arrived without a headscarf and wearing clothes considered inappropriate, and Pak SD said that nothing was available. Another topic is the concept of Muslim housing, which he initiated was different from other housing developer. Conceptually, Muslim housing is open to all the manhaj of Islam as long as has strong argument in Muslim holy book. Another difference is that the housing is not based on schools and figures, as found in the Bukit Az-Zikra housing concept, Muslim Darussalam housing, and so forth.
"If you invite me to environment A, you will be consumed with living in the A way. The conditions of staying in a previous house in a traffic jam and lack of Islamic socialization encourage me to make this Muslim housing. For me, one of the priceless favors is neighbors." (interview with Pak SD, 2016) To support the process, several things were offered in this Muslim housing community in the sales brochure, for example: Pak SD provide the teachers for Muslim housing residents to support the implementation of activities. Some of the methods include the recruitment of people in Pak SD's network and offered other residents who were eligible. Two of these students is Ustadz Af, who plays a role in the Islamic studies of adults and other, who play a role in providing Islamic religious education for children.
"For this purpose, a TPA or Tahfidz Cahaya Besar is held specifically for the residents of Cahaya Besar, with the instructor from the Cahaya Besar living in a place provided by us at the mosque, besides teaching also being a priest for the Cahaya Besar mosque." (interview with Pak SD, 2016) The offer of the facility is not without reason but is an inseparable part of the Muslim's thing that affects its inhabitants to achieve the ideal form of the desire of Islamic housing environment.
Through Muslim housing, the process of building a generation based on Islamic morality is fulfilled with the support of all the residents who love Allah and the Qur'an. Pak SD tells how the association of children in Muslim housing is the responsibility of all people who are the inhabitants. Words and behaviors are crucial and stimulate the ability of hafidz (the title of the Al-Qur'an memorizer).
.". let his parents be black and we direct him according to Islamic guidance … but for his children it needs to be supported to be white in the form of beneficial activities to be a qualified Muslim." (interview with Pak SD, 2016) According to our observations, the growth of Muslim housing was also followed by the emergence of several integrated Islamic-based schools that began returned to the household decisions. The social media and technology elements became part of life because their function could connect one person to other groupas part of the modern needs. Technological developments have also influenced changes in the communication between neighbors, especially regarding ideas in making Cahaya Besar Muslim housing more Islamic. The Whats App group that replaced the BBM Group was more active in delivering good advice and was coordinated by several people active in Muslim housing.
B. Muslim Housing Residents in the Community
From the profile of the occupants, the minimum separate household expenditure was as follows: The budget between a husband and wife was Rp. 25,000,000 per month and sometimes more than Rp. 50,000,000 per month. The amount of the expenditure depends on the details regarding the calculation per year, which is then calculated on a monthly basis, as we submitted.
The monthly expenditure is mostly allocated for education and personal development purposes. Our research demonstrated that the cost of obtaining children's education in the play group class at, for example, SDIT Saadah, was Rp. 15 million per school year. Notably, Al-Falah offers a price of Rp. 97 million per year per child to enroll in the playgroup. The Islamic school teaching system and Islamicway of life asknowledge basd are desirable additions to Muslim housing and complete the process of becoming a Muslim community.
Referring to the profile of the residents of the Muslim Cahaya Besar housing, they were engaged in the private sector and the government sector. The majority of the residents were engaged in the private sector, that is, multilevel marketing businesses such as Oriflame, Nu Skin, and other modern and conventional Muslim business networks; notably, some residents were owners of BTA tutoring institutions (e.g., Sekolah Cikal, Sekolah Buah Hati, and Al-Falah Boarding School), da'wah of forums, and writers and printers of Islamic studies. Other residents were engaged in the military field, for example, two residents were Kopassus officials, one resident was a police official, and other residents were academics (e.g., a famous State University Lecturer in Jakarta). Some empty lots were mentioned, for example, one owner of an empty lot was a former minister of religion.
The individuals who bought housing were typically newly married couples or had children who were toddlers. These young couples had a minimum of an undergraduate education and obtained information directly by visiting the location of the housing or through their relatives. It was only due to a direct survey. At first, I didn't even know that there was a housing because it had to be located on the side of the highway but it was a little inside; thus, it was only just the entrance to people's homes. Additionally, the writing of the housing is small (even though the writing is on the roadside). (interview with Iwed, 2016) However, some people are 50 years old, and their children are taking college courses; for them, the existence of the mosque and its activities are the added value of the housing.
C. Motivation to Live in Muslim Housing
Referring to the research findings, Muslim housing developers are affiliated with several groups active in carrying out political activities and da'wah, one of which is an Islamic majority party. According to Pak SD, the organization developing and transmitting Islamic values to other Muslims is taught in the form of cadre programs. The offer is to build a generation that has the ability to manage modern development with good morals.
The emergence of GCs is because of a gap between one group and another on the basis of location. The implication is that the existence of this housing is for the sake of security that occurs with the emergence of social segregation that is not just about an economy. However, the security meaning of each person's residence may be different depending on their experience and knowledge of placing us and them in the circle of social relations.
Forms of fear are such that choosing Muslim housing is how to prepare children to become qualified Muslims. Muslims should be faithful and devoted to Allah, and that is what is expected to be accommodated by Muslim housing. The role of the family in shaping and directing the child's future becomes the main purpose because it is carried out as early as possible in the most intimate daily sphere.
We observed how the entry of foreigners into the housing was considered a problem regarding the safety of children. Even this concern concerns the fear of their children not following the religious path expected by their parents. This Muslim housing facilitates the fear of the residents and the parents because of their role in providing Islamic religious activities controlled by the founder of the Muslim housing.
"I am more comfortable to live and live with people with the same religion to make it easier for me to educate them to be good Muslims." (interview with Ibu Marwa, 2016) In addition, schools or educational institutions where the children of the Muslim Cahaya Besar housing were integrated to learn Islamic-based knowledge also provided different activities compared with public schools. The activity of memorizing the Muslim Holy Book was mandatory because in elementary school, short Surahs and daily prayers are memorized as early as possible; in addition, extracurricular activities such as archery, swimming, and riding received attention because they are in Al -Qur'an.
In the section on Islamic reform, this movement includes modern urban Muslims who are aware of technology and science. The children growing up in this Muslim housing went to an integrated Islamic education institution that not only places moral construction but also faces the demands of the times that also bring religious.
"Anak-anak ini akan tumbuh dalam enklaf dan terlindung dari pengaruh dunia luar. Mereka akan diajarkan bagaimana hidup secara Salafi dan dikirim belajar ke pusat pengajaran Islam yang juga milik Salafi." (Noorhaidi, 2008) In a discussion, Ozy mentioned how rational choices to inhabit Muslim housing and attend SDIT were part of protecting their children from the influence of Western concepts of secularization. How he wants his children to grow up is as intelligent, skilled, and insightful Muslims based on the Qur'an as Muslim holy book and Hadith. The Muslim housing is chosen by the informant because what they feel is actually not a matter of fear of criminals or pollution. For them who lived in, Muslim housing is an effort to create a "sense of security" that can no longer be expected from the government and the state because Indonesia is not an Islamic state. Muslim housing with limited access creates a supposedly safe and comfortable environment to ensure their children and future generations are in the environment they want to realize.
Some key words about the safe and comfortable concept of several meetings are as follows: We argue that the realization of Muslim housing is to protect children as victims and to protect children as perpetrators of evil and ugliness. This concept was also reviewed by Mary Douglas in the Purity of Danger (1966), which explored a society that had similar conditions and already existed long ago. The context of the research is how these community groups have values or habits different from individuals outside them and want this purity to be maintained and sustainable. The concept of pollution and taboo becomes a system created in the community to prevent the contamination of their cultural purity because it is considered to endanger the sustainability of the group.
Cahaya Besar Muslim housing, according to informants, is not GC in the sense of protection alone but represents how to prepare for the future of their children by creating space where they can become qualified Muslims.
This Muslim residence is one of my efforts to instill the value of Muslims in children because from what I learned in the parent education program, there are three things that affect children, namely, family, school and environment. The spread of liberalization, secularisation is getting stronger and considered threatening, for example, with the existence of Lesbian, Gay, Bisexual and Transgender (LGBT) groups, I was worried about looking for schools and thinking about homeschooling for children. (interview with Ozy, 2016) Muslim housing is supposed to be a stronghold to maintain the quality of Muslims as people who believe and are devoted to Allah, as SWT refers to the Qur'an and hadith. The fort was realized in a physical form that distinguishes us and them, which was also limited by different patterns of time use and different preferences so that it was not possible to meet with the surrounding community groups outside Muslim housing.
One of the reasons why this housing is chosen compared with the public housing located next door (whose conditions are still offered in 2017) was because it would be easier to educate children in a Muslim environment. Although several amily has the experience of living in similar public housing,
Advances in Social Science, Education and Humanities Research, volume 365
for him the offer of the existence of Ustadz and mosques is the main attraction of this housing.
"It's easier to live with the same people, for example, if I live with a non-Muslim, I have to bother explaining what Christmas is." (interview with Iwed, 2016) The majority of the informants had experiences living in GC housing either based on livelihoods or the economy. Perumahan Cahaya Besar, for them, accommodated the need for easy access to the world of work, education, and protection for their families, both physically and mentally.
Muslim housing is part of an effort to form a safe space and needs to be supported by other responses. Muslim housing is considered a safe space because the majority of individuals who live in Muslim housing have the same purpose, marked by an agreement at the time of purchase. Muslim housing is also a safe space for its residents to carry out social religious practices based on only the Qur'an and Sunnah to improve the quality of faith without being considered differently.
The implication is that the community is a safe space because the community reproduces the idea of Muslim way of life into their lives. For this reason, patterns based on rational choices in their daily lives are forms of protection, ranging from children's schools and product selection, that support their status as members of the urban Muslim middle class.
Muslim housing as an enclave has the ability to prevent ideology from being polluted from outside influences considered to endanger the future of their families. In the case of Muslim housing, after getting marriage or having children, there is hope regarding the ideals in the GC about the process of securing a future that involves faith as a Muslim and is related to the finances that must be spent to achieve it. Muslim housing is one part of the process of realizing hope to maintain the purity and piety that is interpreted as an enclave.
Enklaf ini menempatkan masyarakat yang tertindas dan rusak secara moral di bagian luar tembok, yang dipandang sebagai penghuni wilayah yang kotor terpolusi, menular, dan berbahaya bertentangan dengan komunitas penghuni didalam tembok yang bermoral dan bijak. (Noorhaidi, 2008, p. 260) The contribution of Noorhaidi Hasan in this study is that Muslim housing becomes an "enclave" by becoming a sacred safe space where purity is maintained and where each individual conforms to strict general norms. The result of the application of enclave culture in Muslim housing is a process of moving to become a complete Muslim in community based on unity.
One of the informants provided us with a book about how women have the virtue of being at home and the conditions that support that virtue were peeled out by adding a few entertaining jokes. The position of women was as a family pillar, and women played a critical role when carrying out their duties properly.
We also considered this a reflection of the inhabitants of Muslim Cahaya Besar housing, where most of the wives were at home taking care of the children. However, the wives were also active in religious social activities outside the home because they were assisted by a household assistant and usually supervised by close relatives who lived not far from the area. This lifestyle was carried out to maintain parents' obligation to care for their children.
IV. DISCUSSION
The history of GCs in Indonesia was originally part of Dutch colonial influence and created to provide housing needs for its employees based on Freek Colombijn's research. Since then, the emergence of GCs based on the type of work and the military has been considered normal. However, there are also scholars who observed the phenomenon rom GCs, especially those who add Muslim labels, such as Wai Weng Hew. His rgued that Muslim GCs are a spatial formation of the identity and aspirations of the urban middle-class Muslim class. He also stated how the people who lived in Muslim GCs distinguished themselves from non-Muslim GCs through the appearance of Islamic symbols.
The study of Muslim GCs in Jakarta from Weng (2014) demonstrated that the emergence of Muslim housing was part of a religious goal to obtain a balance in daily life in the capital city. The purpose of choosing this type of settlement is related to religion and creating a modern and socially safe community, especially concerning the aspects of crime and floods, which are daily concerns in this capital city. His study also emphasized the discourse of "fear of urban diversity" as part of the strategy developed by certain groups when managing the style of "secular life" and "Christian missionaries." The thoughts of Weng (2014) regarding the relevance of developers in the strength of housing capital, in our opinion, has limitations because it has not arrived at the stage of "sense of community, lifestyle, and security" as Foster conceptualized in Barnes. Involving developers in the discussion of GCs, according to preliminary reasearch, is related to ideological safeguards based on religion and is the marketing language that has been used since the development stage of the planning of GCs.
However, a number of criticisms of the concept of Muslim housing can be debated considering how the conditions of each Muslim housing differ. Some think that Muslim housing is sufficient as "the origin of Muslim residents," and others develop Muslim environments starting with the facilities as part of values education as a Muslim. Additionally, there is the issuance of a number of rules that filter prospective buyers, ranging from religion and manhaj (methods), that include obligations that must be obeyed by residents and households. Housing or settlements have an essential function to create a sense of security from the selection of residences and even the development of physical barriers to create defensible space.
Referring to the aforementioned findings, the source of the occupant's control is initiated by an agreement between the developer and the prospective occupant of the house. Power is in the owner of the resource to limit open access to certain people. This concept by Amos Rapoport, in the "House Form and Culture," considered that for traditional communities, the house is a mechanism of social control that will even affect the way of life of the surrounding community.
"The idea of the house as a social control mechanism, so strong in traditional cultures at least Advances in Social Science, Education and Humanities Research, volume 365 may no longer apply with as much force in a society with the formalized and institutionalized control systems of today." (Rapoport, 1969:49) By contrast with the Muslim housing studied, the mention of traditional society by Amos Rapoport actually occurred in the current era and is expanding, and the target market is young families with high education levels. However, we agree with Rapoport's understanding of the concept of behavioral architecture, which indicates that a place of residence is a reflection of the view of life and values held by someone.
A. Muslim Housing as a Safe Space
The latest Low writings (2017), about ethnography "space" (space) and "place" (place), have greatly contributed to understanding the study of GC. The concept of spatializing culture, according to Setha Low, that we use to analyze a group of people who choose to live in Muslim housing as a place cannot be separated from the processes of social production, social construction, embodied, discourse, emotion, and affection.
The studies of Setha Low have become the framework in the discussion of this chapter is Muslim housing interpreted as a social construction space. The role of power relations in interpreting Muslim housing space is related to class and gender. Although the context is about social segregatiton based onrace, Low (2017) also states how GCs become a social construction space formed by the purification and privilege of certain groups.
The social construction of "whiteness" in Gated Communities provides another example of how intersecting constructions of race and class permeate the suburbs to constitute a purified and privileged social space (Low 2009(Low , 2003. Whiteness is not only about race, but it is also a historical and cultural construct actively produced and reproduced to further and/or improve an individual or social group's position.
This meaning entered into the domestic space where the concept of social construction emerged about families, children, households, although sometimes it was also closely related to the economic aspects as one of the drivers of the formation of the GC community. We find that product choices or activities are part of the social construction to maintain the spaces they think are safe.
"Ethnographies of the social construction of space decode and deconstruct the struggles, contestations and power dynamics underlying existing social and spatial relations." (Low, 2017, p. 69) There are at least two perceptions that we consider important in placing a place and space in the housing of a Muslim GC. Perceptions of residents and developers at the beginning of the purchase of a house are that Muslim housing will be a safe space for themselves and their families when carrying out their life processes as Muslims. Second, the perception of residents after living their lives in Muslim housing is that this Muslim housing is not a sufficiently safe space for themselves and their families and their fear is increasing.
Muslim housing is part of the space that involves the power relations of knowledge about how the social construction of Muslims is shaped. The hope is that Cahaya Besar Muslim housing will be an accommodating place that will be a safe space for those constructing their identity. Muslim housing then becomes a "liaison" to a larger space because it is considered insufficient to accommodate the emotions of its inhabitants.
Referring to the two aforementioned ideas, Muslim housing as a safe space for the process of transformation of the secular community is expected to be religious. Muslim housing becomes a place in space because of one concern, namely, crisis. Crisis becomes a concern because of the shadow of an uncertain future, and a number of strategies are required to minimize risks. Knowledge and experience between those who provide and seek safe space are then realized in the form of a built environment that is the lifestyle of the Indonesian Muslim middle class today.
The concept of Muslim housing is interpreted as part of a journey to become a Muslim who, in the developing language of marketing, is now called "hijrah." This term became popular with the advent of Ust. Arifin Ilham owns a boarding school in a number of communities in Sentul, West Java, and establishes Muslim housing in Bukit Az-Zikra. The concept of the founder of Cahaya Besar Muslim housing also realizes that the meeting of interests between developers and the interests of Muslim housing residents can further explore the Muslim life because they imagine the results of their experience and knowledge. Some of them have been students of religious institutions since childhood, but most of the residents are the general public not actively involved in Islamic religious activities. Through routine activities and the repetitive dissemination of knowledge and the strong role of Indonesia's current social, political, and economic context, the process of migration is expected to be realized. Geertz also indicated that the Javanese Islamic groups, which he made, were not rigid and allowed the existence of abangan priyayi, and priyayi santri.
"Priyayi Santri adalah sebutan untuk priyayi yang secara aktif melibatkan diri dalam agama Islam" (Bachtiar, Harsja W.,2013, p. 598) Geertz's distinctive priyayi designation with Javanese culture is one of the largest ethnic groups in Indonesia in his Modjokuto research area. Despite the many debates, the emergence of the concept of the middle class, which is considered to have a distinctive character within the city community, has not experienced much development since the introduction of Tanter and Young.
"Para kelas menengah ini telah mengenali basis kekuasaan baru, kepentingan material tersendiri, dan membayangkan sebuah struktur produksi baru yang mengandalkan kemampuan organisatoris (manajer), pengetahuan, ijazah, informasi, wacana, dan kewenangan (profesional dan intelektual), atau politik birokrasi (pejabat negara) sebagai aset utama." (Tanter dan Young, 1993) This middle class, for Muslim housing developers, is getting bigger and then becomes the target of the sharia industry market in Indonesia. Individuals in this group have a good level of education and job positions that have authority. Basic needs that have been fulfilled for some people enable them to increase their awareness regarding obtaining spiritual needs. The transformation of these individuals leads to the search for identity as a devout Muslim. This new awareness that emerges then becomes the basis for educating their children so they gain an understanding of their religion better than their parents because of the earlier acceptance of their teachings.
The responsibility of building the quality of children is accommodated by the emergence of activities for the limited community of the Cahaya Besar inhabitants. Pak SD acknowledged that this thought that he obtained during his education became a religious party-based cadre, the majority Islamic party. Pak SD's activities, other than owning a business, include active involvement in study activities inside and outside Cahaya Besar housing. Although Pak SD did not want to be called a businessman, he was a man who with one restaurant in Matraman and several educational foundations; additionally, he often actively participates in prayers at the Cahaya Besar mosque by bringing along his male children.
"I am a member of party with a majority of Islam, we cannot do anything without being social if we live in a good environment, good education, good social will make us good people, if not then we have tried." (interview with Pak SD, 2016) The target of Muslim housing that was initiated was not for the present, that is, parents are buyers but more emphasis is on their children. Creating a good Muslim environment will result in good Muslim people. Children's education is not just a diploma but also represents that the children have acquired quality Muslim knowledge and virtues.
Closed or gated minds'' would have ''channels'' identifying those we wish to be physically separate from, ''ridges'' for those we accept, ''trenches'' for those we hate and fear, and impervious ''barriers'' for those with whom we do not wish to communicate. In summary, human behaviours, which emanate from the brain, have their own ''geographies,'' which are reflected in ''human spatial behaviours.'' (Brunn, 2006:6) Related to this findings is the concept of "Gated Minds," as stated in the writings of Brunn (2006), that is, the process of learning the thoughts and behavior of residents of GCs is instilled in everyday life early. Education is an effort to increase knowledge and skills and is also interpreted as a capital of social control in the context of external conditions considered dangerous.
V. CONCLUSION
This paper provides an understanding of space as a housing-based social construction. The context of religious revival and some events in Jakarta contribute to the political shift in urban space. Referring to Setha Low's work regarding understanding the "Gated Community," which is interpreted as "Fear of the City" (Fear of a Big City), this thesis agrees with the argument that the physical phenomena of the walls and gates of housing are a form of fear of repression and a rejection of the choice to live in a big city. This study then observed that the GC was interpreted as one of the safe spaces in the undergoing the social construction process in an urban Muslim community.
The context of the strengthening of the Islamic revival is lately supported by the existence of a growing Muslim middle class, especially in Jakarta and its surrounding areas such as Depok and Bogor. Such conditions are of course contradictory to the findings of previous GC studies of marginal, marginalized, and minority groups, that is, fear becomes normal. The phenomenon of GCs in Muslim housing has emerged in the midst of a Muslim-majority society, and the community that lives in it also has social capital such as offices, wealth, figurehead positions, and high education, which should not be a reason for the fear experienced elsewhere.
The decision to choose Muslim housing is in line with the process of building a household and is interpreted as an effort to "migrate." Muslim housing is one of the enclaves that supposedly protects individuals from outside influences. Muslim housing as a space is a social construction surrounded by values, lifestyles, systems of activity, meanings, hopes, statuses, and identities of Muslims. One key example of fulfilling one of these basic needs is the mothers who have concerns because they have the role of the first madrasa for their children. This Muslim housing is perceived as the appropriate environment because it fits with the aspirations of Muslims.
Muslim housing is also considered one part of the phenomenon called "Islam Resurgence," including the emergence of symbols of Islamic values in public spaces. This Muslim housing community forms a culture in the packaging of "hijrah" because it came from ordinary people (non-santri) who had the desire to be pious Muslims. The Muslim-majority group, which is still young, is attempting to build a household in an environment that is considered sacred, clean, safe, and based on religion. Muslim housing is also a means to control the fear of crimes such as kidnapping, robbery, or theft and things that pollute their faith aspirations. We find that Muslim Cahaya Besar housing is an interconnection of other social spaces also formed from social, political, and economic networks, such as Muslim families securing a future in the context of fearing of the contamination of belief ideology.
Geertz's contribution to the Javanese religion increased the understanding of the identity of the inhabitants of Muslim housing, which, according to their characteristics, were priyayi. Most of the inhabitants are ordinary people who are not in a strong religious environment, such as pesantren. This Muslim housing aims to make its inhabitants worship experts who, according to Geertz, are categorized as santri. However, to undergo this process, priyayi identity is also active because it requires a number of sacrifices to be able to enter into the construction of identity.
There are many types of Muslim housing, for example, Muslim-only housing, housing constructed by Muslims, housing with Muslim ornaments, or any combination of the three. The initial idea of Cahaya Besar housing was formed and recognized as a Muslim housing by developers who also agreed on its meaning by residents. This Muslim housing is a boundary for the area that can be controlled by the community.
Referring to the history of Muslim housing in the location studied, housing is offered with activities that refer to Islamic values. Agreements are made between the developer and the buyer through a contract but are not necessarily made by consensus because of the diverse buyer backgrounds. At this stage, the developer has the dominant power in forming Muslim housing because of the selection that considered the buyer's profile.
Advances in Social Science, Education and Humanities Research, volume 365
Housing developers do not offer their products in public housing exhibitions; instead, they offer their products through religious-based communication networks. Institutions involved in this case, especially educational institutions or similar self-development institutions, are one of the actors that construct the perception that such a topic of importance is making a house similar to "jannah" (heaven).
One way developers influence the community is through the physical facilities offered such as places of worship and nonphysical facilities, namely, the form of daily routine activities that we interpreted as a form of maintaining a safe space for a Muslim guided by a teacher. The importance of the existence of a teacher or religious leader does not automatically become the ruling party.
Markets targeting middle-class young Muslims have the power to form the community and run it dynamically because it faces a number of challenges in implementing the initial idea of Muslim housing. This form of power is determined by the developer and a collective agreement is obtained with the aim to create a Muslim housing community. The implementation of the code of conduct, which is the obligation of its inhabitants, is part of the effort to control the fears.
Cahaya Besar Muslim housing is part of the "enclave" space, and the influence of the outside world is considered destructive or dangerous to Islamization. The purpose of choosing Muslim housing is to make their children become Muslims by providing an environment that supports the construction of their Muslim identity. Nevertheless, Muslim housing is not sufficient to overcome their fears; thus, other assembleges are needed in line with the aspirations of these young households, for example, by integrating children and mothers at schools or other educational institutions with the concept of Integrated Islam, using halal goods, and visiting places recommended by Ustadz or friends in the study. Thus, indirectly, the residents of Muslim housing form limited community-based networks or GCs in the true sense.
This Muslim housing group also takes a long time to become a community. Although, since the beginning, Muslim housing has been supported by activities carried out together, the perception of the Islamic values of housing has also colored the dynamics of the acceptance of prevailing rules. Knowledge distribution to safeguard communities formed by housing is also carried out by using existing social media networks. Actors who are active in the community usually have a substantial role in managing the Muslim housing environment. The actors selected to maintain the stability of Muslim housing are usually those who give up most of their time, their funds, and their homes for mutual benefit. Nevertheless, the influence of the initiator of the Muslim housing idea is the basis of consideration in every decision made that determines the standards, rules, and norms that apply.
|
2019-12-05T09:24:46.597Z
|
2019-11-01T00:00:00.000
|
{
"year": 2019,
"sha1": "a24e3b453ba0edb0c13ef45031df2046b8d6e26f",
"oa_license": "CCBYNC",
"oa_url": "https://doi.org/10.2991/icsgs-18.2019.24",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "80827756dc64850506560393e4f10712574f38f9",
"s2fieldsofstudy": [
"Environmental Science",
"Sociology"
],
"extfieldsofstudy": [
"Political Science"
]
}
|
247430506
|
pes2o/s2orc
|
v3-fos-license
|
Preparation, characterization, and antibacterial competence of silymarin and its nano-formulation
Abstract Nowadays, antibiotic resistance is a serious problem, especially in countries that have unregulated use of antibiotics. We aimed to prepare and characterize Sily-Nano from silymarin to present its antimicrobial activities in comparison to silymarin and gentamycin and provide natural antibacterial from natural sources. Silymarin NPs (Sily-Nano) were prepared using a high-pressure homogenization and characterized using X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR), ultraviolet-visible (UV-visible) spectrum, and scanning electron microscopy (SEM). The antibacterial monitoring of Sily-Nano and silymarin was compared against Escherichia coli UN75 (E.coli-UN75), Enterobacter aerogenes SL3 (E.aerogenes-SL3), Staphylococcus aureus-PS41 (S.aureus-PS41), and Klebsiella pneumoniae-BD29 (K.pneumoniae-BD29). Sily-Nano was effectively prepared and confirmed using FTIR. SEM, with even distribution, and significant bacterial zone inhibition, compared with silymarin. In conclusion, Sily-Nano preparation is an efficient agent, and effective against certain clinical antibiotic-resistant bacterial strains. Sily-Nano has a novel significant inhibitory effect against the antibiotic-resistant strain of K. pneumonia-BD29, Ent. Aerogenes-SL3 and E. coli UN75 when compared with silymarin and the common antibiotic gentamycin. Therefore, sily-Nano is potentially used as an antibacterial and natural sanitizer agent.
Introduction
Antibiotic resistance and medicinal efficiency is significant problem, challenging healthcare in several countries. Recently, novelties in nanotechnological strategies offer a great opinion to enhance the assimilation, solubility, and bioavailability within the cells so, provide a method to potentiate the therapeutic action and provide sustained silymarin release of the active herbal extract at the absorption sites via nanoencapsulation [1] The number of scientific literature demonstrating silymarin-based formulations with improved solubility, stability, enhanced bioavailability, and effective medicinal roles is continually expanding. The formulations schemes rationally followed stage by stage the development of the nanosystems and nanotechnologies [2].
Antibacterial activity is one of the vital applications in the current studies in biomedical specialties. Numerous infectious species as S.aureus, E. coli, Staphylococcus aureus, P. aeruginosa, and K. pneumoniae induce several serious diseases in humans, so, we need to find out a prospective antibiotic to treat these infections. Sometimes, the antibiotics cannot inhibit the harmful bacterial progression.
Natural and synthetic antibiotics that are leaked in water resources or utilized in food preservation accelerate the development of bacterial resistance and reduce the medication efficiency which, in the judgment of specialists, will have permanent consequences and may become fatal. Antibiotics in uncontrolled amounts not only induce bacterial resistance but also may cause an ecological danger. The expansion of antibacterial resistance for several pathogenic bacteria has one of the riskiest issues in the management of contagious diseases (WHO) [3].
Antibiotic resistance, especially for the latest generation of antibiotics developed by a pharmacist, is one of the main challenges in the recent therapeutic approaches. Therefore, the discovery of innovative medication with a new mode of action is a great task for biomedical investigators and industries. The blend of biomolecules derived from medicinal plants and formulating into nanomaterial offer major possible sources of novel antibacterial agents to regulate bacterial diseases and their resistance.
Silymarin is virtually water-insoluble, upon oral administration, because of its remarkably non-ionizable and hydrophobic structure, so represents minimal biological accessibility where their dissolution is a rate-limiting stage for the expression of bioactivity [4]. Its rate of absorption in the digestive tract is small, offering only about 20-40% bioavailability [5]. To improve silymarin bioavailability, complexation with a phospholipid [6,7], bcyclodextrin, and hydroxypropyl-b-cyclodextrin; solid dispersion with a hydrophilic polymer, such as PEG 6000; self-micro emulsifying systems; co-precipitates; and a complex with crosslinked polymers are required. The antimicrobial activity and control of the growth of food microorganisms of silymarin may be effectively improved by nanoformulation, therefore, we formulated Sily-Nano to challenge these issues.
The recognized effects of silymarin are anti-lipid peroxidation; anti-inflammatory, antiangiogenic, and anticarcinogenic, and the ability to stimulate hepatic regeneration. Though, there are inadequate records about its antibacterial activity [8][9][10]. Natural extracts have large quantities of bioactive substances, mostly polyphenols which prevent the microorganisms' growth, particularly bacteria. Their actions are not completely understood however, they may be linked to their biochemical structures. Silymarin revealed antibacterial action versus the gram-positive bacteria Bacillus subtilis and Staph. epidermidis. This was mainly due to the inhibition of RNA expression and the synthesis of protein. Also, its antiviral action is played through the inhibition of viral transmission and diffusion [11].
There is a shortage in the research dealing with natural antibacterial as a sanitizer and its use as nanoformulation, that consider a research gap that leads to our study. Because of the antioxidant and antibacterial activity of phenolic composites, plant extracts offer a novel substitution to the chemical additives used in the industry of meat, particularly nitrates. They can prevent the development of spoilage and pathogenic microflora, inhibit oxidation of meat components, like carbohydrates, proteins, and lipids, and inhibit staining and discoloration. Therefore, the current study aimed to prepare and characterize Sily-Nano from silymarin to present its antimicrobial activities in comparison to silymarin and gentamycin.
Materials
Silymarin powder was attained commercially from MADAUS GmbH Cologne (51101), Germany (reg no. 4-1089). Silymarin ( Figure 1 derived from pubchem.ncbi.nlm.nih.gov/ compound/Silymarin) is a combination of flavonoids obtained from seeds of the MILK THISTLE, Silybum marianum. It is composed of silybin and its isomers, silichristin and silidianin. Silymarin shows antioxidant and protects several tissues against biochemical damage, and act as a hepatoprotective agent.
Silymarin and Sily-Nano as antibacterial agents
The initial silymarin stock solution was prepared at 10 mg/mL concentration in one mL of dimethyl-sulfoxide (DMSO). The concentration of DMSO didn't exceed the value of 2% to avoid its detrimental effects on the tested bacteria. This concentration was used as follows: the two tested composites were diluted in 1 mL of sterile distilled H 2 O to attain concentrations of 8,12,16,24,32,48,64,96,128,192, and 256 lg/mL.
The antibacterial activity of silymarin and its derivative was evaluated at the concentration of the initial stocks by measuring the resulting inhibition zones (mm) in seeded Muller-Hington agar by a formerly adapted method of Beecher and Wong [12] against selected clinical antibiotic-resistant bacterial strains of E. coli UN75, Staph. aureus PS41, Ent. aerogenes SL3, and K. pneumoniae BD29. Those strains were obtained from the culture collection of the microbiology lab of Basic and Applied Scientific Research Center, Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia. The M€ uller-Hington agar consists of beef extract (30%), casein hydrolysate (1.75%), starch (0.15%), and agar (1.7%). It was prepared then autoclaved at 121 C for 15 min thereafter, cooled to 47 C, and sowed with the examined bacteria under aseptic conditions. Subsequent to solidification, 5 mm diameter holes were pressed by a sterilized cork-borer. The examined compounds with reference to the antibiotic gentamycin, were then added in the holes (only 100 mL) after being dissolved in the DMSO at 10 À4 M. Then, these culture plates were incubated for 18 h at 37 C. The diameter of the inhibition zones in mm was measured to determine the antibacterial activity.
Moreover, the action of silymarin and Sily-Nano was tested in a plate culture of the bacterium Staph. aureus PU41 growing on Baird-Parker specific medium. This medium consists of (g/l), the pancreatic digest of casein (10), glycine (12), sodium pyruvate (10), lithium chloride (5), beef extract (5), yeast extract (1), egg yolk tellurite enrichment, 60 mL, and agar (20) with pH adjusted to 7.0 and incubation was done at 37 C. The upper wells of the plate culture were inoculated with 40 ml of Sily-nano at a concentration of 40 mg/mL, while the lower wells were inoculated with silymarin.
MIC preparation
The minimal inhibitory concentrations (MICs) were determined for both compounds. The following culture media were utilized for the bacteria; brain heart infusion broth (BHI) at 10% as showed by the constructor; Acumedia Manufacturers Inc.) and heart infusion agar (HIA) Difco Laboratories Ltda). All culture media were equipped following the producer's guidelines. Cultures of bacteria were kept at 4 C in HIA. The strains were passaged, before the assessments, utilizing the aforementioned media then incubated for 24 h at 37 C. The strains that plated were injected into BHI broth and once more incubated for 24 h at 37 C. A minor aliquot part of the cultured inoculum was detached and diluted in sanitary saline to provide turbidity equal to 0.5 on the McFarland scale, equivalent to 10 5 CFU/mL [13]. The indicator for the bacterial growth was sodium resazurin reagent. It was attained from Sigma Company, and kept at 4 C far away from the daylight. Concerning the interpretation of the test, bacterial growth was indicated by changing the color from blue to pink owing to the reduction of resazurin [14].
The MICs of silymarin and its prepared derivative regarding the antibiotic gentamycin were measured by the microdilution analysis. 100 lL of 10% BHI medium was tallied to every well of the microplate then 100 lL of the bacterial suspension was inoculated to all wells, thence, 100 lL of the examined product was used to achieve concentrations in the average of 2:512 lg/mL. The negative control consists of 100 lL (10% BHI medium) and 100 lL (test product without inoculation of bacteria, while adding sterile distilled water instead). In The Meantime, the positive control only had the bacterial suspension and 10% BHI.
The microplates were reared at 37 C for 18 h [15]. Bacterial growth was measured utilizing resazurin. The analyses were applied in triplicate. MIC was described as the smallest concentration upon which no bacterial growth was detected in agreement with NCCLS [9]. Ultracentrifugation for the preparation and characterize the particles with SEM.
Silymarin NPs preparation
Three capsules of Silymarin (Legalon forte) were ruptured and the obtained powder was transferred into a separating funnel followed by the addition of methanol (100 mL). The mixture was shaken very well and filtered. The obtained filtrate was evaporated under reduced pressure and obtained powder was stored at 4 C. To prepare the Sily-Nano, 50 mg of Silymarin powder was dispersed in 25 mL of methanol in a beaker, the solution was transferred into a high-pressure homogenizer and treated at 1200-1500 psi using 3 cycles. The solvent was evaporated using a rotary evaporator and the obtained powder was stored at 4 C for additional studies.
Morphology and characterization of the Sily-Nano
Sily-Nano with diverse amounts was equipped by solvent evaporation and nano-precipitation methods. X-ray diffractometer (XRD, Rigaku, Japan) ( Figure 2), was employed to study the phases of Sily-Nano in the range of 10 -80 with 0.9 /minute scanning speed. The morphology and size of Sily-Nano were determined by a scanning electron microscope (FEI, Czech Republic). The sample was spread uniformly on the sample holder having carbon tape on it and images were captured using different magnifications. FTIR (PerkinElmer) was used to evaluate the functional groups of Sily-Nano using ATR accessory. The background was recorded with an empty ATR accessory and then a small amount of sample was put on the ATR crystal and a spectrum was recorded using 4 cm À1 resolution and 32 scans in the range 500-4000 cm À1 . UV-visible spectrophotometer (UV-Vis, JASCO V-750) was used to record the UV-visible spectra of Sily-Nano in methanol in the range 200-800 nm, before the analysis baseline was recorded with blank methanol.
The scanning electron microphotograph of Sily-Nano was displayed in Figure 3. It indicated that Sily-Nano has a distinct circular structure without accumulation. There were no drug crystals noticed on the exterior of the NPs.
Statistical analysis
Data were statistically assessed by one-way analysis of variance (ANOVA), followed by the Tukey-Kramer method for post-hoc analysis. Results showed as mean ± SEM, and the values were considered significant when p < 0.05 and very highly significant when p < 0.001 in the tables. The dissimilar superscript letters (a, b, c) considered significant changes at p < 0.05 in the tables. The statistical examination was achieved by the GraphPad Prism 6 software (San Diego, CA, USA). The statistical examination was achieved by the GraphPad Prism 6 software (San Diego, CA, USA). Table 1 showed that the tested bacteria including, E. coli UN75, Staph. aureus PU41, Ent. aerogenes ST3 has a statistically significant difference between the inhibition zones in silymarin, Sily-Nano, and the antibiotic gentamycin groups (p 0.05). Comparative data of MIC of silymarin, its nano-derivative, and the antibiotic gentamycin, where the nano-derivative induced 15.25 mm inhibition zone diameter with MIC of 48 mg/mL with Ent. Aerogenes-SL3. While the maximum resistance was exerted by K. pneumoniae-BD29, where the nano-derivative just induced 10.25 mm inhibition halos with MIC of 96, 48 mg/mL (Tables 1 and 2).
Results and discussion
XRD pattern of Sily-NPs indicates amorphous nature of Sily-NPs ( Figure 2). SEM image of Sily-NPs, showed that, Sily-Nano particles are discrete spherical structures without aggregation with an average size of 500 nm. Furthermore, there were no crystals of drug observed on the surface of NPs (Figure 3). Figure 4 presented the FTIR spectra of Sily-NPs. Figure 4 displayed the UV absorption spectrum (288 nm) and stability of silymarin-NPs. The antibacterial activity of silymarin and its nano derivative with reference to the antibiotic gentamycin were presented in Table 1. 24 hr plate culture showing the characteristic black growth of the bacterium Staph. aureus PU41 grown on Baird-Parker agar specific medium ( Figure 6). The upper plate wells that inoculated with Sily-nano and showed high inhibition zone (Figure 6a). Moreover, the lower wells inoculated with silymarin and demonstrated lower inhibition zone compare with Sily-nano (Figure 6b). The characteristic black growth of the bacterium Staph. aureus PU41 in the absence of tested compounds shown in (Figure 6c). This type of bacteria that grows on Baird-Parker agar specific medium is distinguished from all known bacteria by its black color showed that Sily-Nano inhibits the growth of bacteria compared to silymarin.
For all the tested bacteria, there was a statistically significant difference between the inhibition zones due to the silymarin, Sily-Nano, and the antibiotic gentamycin as established by one-way analysis of variance (ANOVA).
The test of homogeneity of variances (p ¼ 0.31, 0.87, 0.28, 0.45 > 0.05 respectively cleared that data have equal population variances then the Tukey test indicated that the original and Sily-Nano are the most significantly effective on inhibition zone for E. coli-UN75, Ent. aerogenes-ST3, and K. pneumoniae-L29, while the ANOVA test (p ¼ 0.0002, 0.001, 0.001 < 0.05 respectively (assure that the Sily-Nano scored higher inhibition zone than other groups. Thus, the Sily-Nano is the most significantly effective on inhibition zone for E. coli-UN75, Ent. aerogenes-ST3, and K. pneumoniae-L29.
On the other hand for Staph.aureus-PU41, Tukey test indicated that the antibiotic gentamycin is the most significantly effective on inhibition zone but Sily-Nano is still better than silymarin according to the t-test (p ¼ 0.00 < 0.05 .( Bacteria as E. coli-UN75, Staph. aureus-PS41, Ent. aerogenes-SL3, and K. pneumoniae-BD29 causing infections have a high incidence and are accountable for the increase in worldwide morbidity and mortality of infections [16]. Reasons included in this increase differ from inadequate sources of antibacterial activities, particularly in poor countries, to the rate of antibiotic resistance. Therefore, in the past eras, the common utilization of medicinal plants and their byproducts has been improved for the treatment of diseases caused by bacteria [17]. Many research on the estimation of the antibacterial activities of natural medicinal plants has been directed to increase the range of antimicrobial treatment. Nevertheless, it was significant to indicate that the technique of microdilution, working in the present study, presently signifies the method most frequently used for this bioassay [18]. Medicinal plants are widely used because of their safety, are well tolerated with minimal side effects, and could be used as dietary supplements. Silybum marianum (L.) Gaertn known as the milk thistle has been applied for a long time ago to treat hepatic and gallbladder illnesses, comprising hepatic inflammation, cirrhosis, jaundice, and to defend the hepatic tissues against chemical poisons and environmental toxins [19]. The active component obtained from a standardized extract of S. marianum seeds is silymarin, it includes about 70:80% flavonolignans of silymarin and about 20:30% is a chemically indefinite fraction, including frequently oxidized polyphenolics and polymeric compounds [20]. Studies exploring the biological activity of silymarin has improved, including several pharmacological roles accompanied with these substances, in addition to the safe use of silymarin, while there were few records of the contrary effects, the contrary effect is few side effect which may be, diarrhea, nausea, heartburn and itching, the possible reason is that milk thistle can trigger allergic reactions [21,22]. Various methods have been used to improve the efficacy and solubility of medicines/drugs. Herein, we prepare the Sily-Nano to improve its antibacterial activities in comparison to silymarin.
XRD pattern of Sily-NPs is presented in Figure 2, a wide peak between 10-35 degrees was observed indicating the amorphous phase of Sily-NPs.
SEM image of Sily-NPs, specified that, Sily-Nanoparticles are distinct spherical structures without accumulation with 500 nm size ( Figure 3). Also, there were no crystals of drug observed on the surface of NPs. This data confirms the validity and homogeneity of Sily-Nano for the antibacterial agent.
FTIR spectra of Sily-NPs indicated that the peak was observed at 3300-3400 cm À1 which is assigned to the OH (hydroxyl) group of silymarin while the peak at 2900 cm À1 is due to the presence of CH group. The carbonyl group (C ¼ O) of silymarin showed a peak at 1638 cm 1 . Almost, similar FTIR pattern was observed in the case of silymarin (Figure 4). The UV absorption spectrum of silymarin-Nano was reordered in methanol and it showed an absorption spectrum around 288 nm ( Figure 5). To study the stability of Sily-NPs, spectra of Sily-NPs were recorded in methanol for three consecutive days, as it is evident from the spectra ( Figure 5), there is no degradation of Sily-NPs and in all spectra, peak was observed around 288 nm. Moreover, no extra peak was observed in all spectra indicating that Sily-NPs are stable.
The results in this study demonstrate a comparative antibacterial activity of silymarin and its nano-derivative in comparison with the antibiotic gentamycin. Where the activities of the nano-derivative against all the tested bacteria were better than the original silymarin ( Table 1). The best results were obtained against Ent. Aerogenes-SL3 as the nanoderivative induced 15.25 mm inhibition zone diameter with MIC of 48 mg/mL. While the maximum resistance was exerted by K. pneumoniae-BD29, where the nano-derivative just induced 10.25 mm inhibition halos with MIC of 96 mg/mL and 48 mg/mL (Tables 1 and 2).
The highest significant inhibition zone for antibiotic resistance strain of K. pneumonia, Ent. aerogenes-SL3 and E. coli-UN75, was recorded with Sily-Nano compared with gentamycin. These results indicated the best effect of Sily-Nano on the inhibition zone. MIC showed that Sily-Nano's inhibitory concentration is minimal compared with both silymarin and gentamycin signifying its efficiency in the inhibition of bacterial growth. This may be due to the large surface area of NPs to provide its effect on bacterial growth. The antibacterial action of silymarin as a natural medicinal plant extract is based on the presence of flavonoids.
In addition, for most bacteria, the antibacterial activity of gentamycin was inferior in comparison with the modified drug except for the bacterium Staphylococcus aureus PU41. This may be attributed to the lower amounts of phospholipids in the cell wall of this gram-positive bacterium. Whereas, the cell walls of other tested bacteria are rich in lipids.
The plate culture that inoculated with Sily-nano and silymarin (Figure 6a and 6b) showed a high inhibition zone in Sily-nano compared with original silymarin confirming the efficiency of Sily-nano as an antibacterial agent. Figure 6(c) displayed the characteristic black growth of the bacterium Staph. aureus-PU41 on Baird-Parker agar, declaring that Sily-Nano inhibits the growth of bacteria compared to silymarin, indicating its efficiency as a specific antibacterial agent against Staph. aureus.
With reference to literature, studies have revealed that several natural compounds including silymarin exert their antibacterial activity by changing the permeability of the cell membrane [23] and can produce morphological alterations in the bacteria, destruct bacterial cell membrane, or interact with DNA topoisomerase that modifies the enzyme binding activity [24]. Moreover, polyphenols affect bacterial protein synthesis, modify the metabolic processes in bacterial cells, and suppress ATP and DNA biosynthesis. In this perception, the antimicrobial action of these substances might also be connected to the existence of hydroxyl phenolic groups that inhibit the enzymatic activities in the bacterial synthetic processes [25][26][27]. For this, medicinal plant types rich in active ingredients as flavonoids merit consideration [28,29].
The superiority of NPs bulk counterparts may be explained based on (1) their capacity to hook up thiol groups (-SH groups) located on the bacterial cell membranes, leading to cell cracking and lysis, (2) their ability to induce the production of high amounts of reactive oxygen species, and (3) the release of NPs tracked by the production of highly reactive oxygen species (OH À , H 2 O 2 and O 2 2À ). Thereafter, holes split up water molecules into hydroxyl ions and proton ions. Then dissolved O 2 is converted to superoxide radical anions ( O 2-), that react with protons to generate HO 2 radicals, which strike electrons to form hydrogen peroxide anions (HO 2 À ). Therefore, they react with protons to form H 2 O 2 that penetrates the bacterial cell membrane and suppresses the intracellular metabolic pathways of bacteria [30].
As a conclusion, this study specifies the possibility of using Sily-Nano as a basis of a new medication as adjuvants in the antibiotic treatment versus several antibiotic-resistant bacteria. The mechanism of Sily-Nano as an antibacterial agent is due to several bioactive components, Silybin, a flavonoid with a smaller surface area [31,32]. Mostly NPs, due to smaller particle sizes, cause an increase in adhesiveness and actively penetrate the cell membrane through small pores in the microbial cell, leading to disrupting the bacterial membrane and exhibiting higher antibacterial activity. Moreover, Sil-Nano may cause the imbalance of minerals and leakage of intracellular proteins and enzymes, resulting in cell growth inhibition and death. It also prevents the accumulation of the bacterial strain that causes antibiotic resistance and failure in treatments.
Conclusion
Sily-Nano as an organic compound could be prepared and characterized with stability. Both Sily-Nano and silymarin displays antibacterial activities against particular typical strains of bacteria which could be valued as a nutritional supplement or a medication. The use of Sily-Nano could be effective as an antibacterial agent via a significant increase in inhibition zone and decrease in MIC against the antibiotic-resistant strain of Staph. aureus PU41, Ent. Aerogenes-SL3 and E. coli UN75 when compared with silymarin, so prevent the development of antibiotic resistance for bacteria with suitable applied methods. Moreover, Silymarin and Sily-Nano could be used as natural sanitizers without bleaching and in the fresh-cut food industry to certify its microbiological safety. The outcomes of this work may be used in the novel industrial field to improve the nanoformula for topical application and to deliberate encapsulation of conservative antibiotics to improve the germicide activity. Further research may be applied to explore the bioavailability improvement via nanoformulation of silymarin.
|
2022-03-14T15:04:53.712Z
|
2022-03-11T00:00:00.000
|
{
"year": 2022,
"sha1": "9ee9750a206d4a9cf89a606b744156f18bb860ee",
"oa_license": "CCBY",
"oa_url": "https://www.tandfonline.com/doi/pdf/10.1080/17458080.2022.2041192?needAccess=true",
"oa_status": "GOLD",
"pdf_src": "TaylorAndFrancis",
"pdf_hash": "cb4d0ecd92527ea62c28ad631e88e9414840ae46",
"s2fieldsofstudy": [
"Engineering"
],
"extfieldsofstudy": []
}
|
270470411
|
pes2o/s2orc
|
v3-fos-license
|
Enabling action: Re-envisaging education of health professionals in Aotearoa New Zealand
Healthcare in Aotearoa New Zealand is changing with the aim of becoming truly universal. Development of a new curriculum model in the education of health professionals can aid this goal through increased focus on community needs and flexibility for multiple health professional registrations. Universal healthcare and disability support are promoted as a defining feature of Aotearoa New Zealand, yet inequities are blatantly evident, with increasing calls for equity. Complexity of the issue and the multitude of stakeholders favour action research founded on Vision Matāuranga, a problem-solving philosophy which embraces the potential for innovation of Māori – people, knowledge and resources. Action research embraces collaboration with stakeholders to identify and implement solutions. Government ministries determine policy, with Responsible Authorities accrediting educational institutions, which in turn provide educational programs. Changing what and how students learn can increase their understanding of equity and community needs when they become practitioners, with the voice of service users being paramount.
Introduction
E koekoe te t ui, e ketekete te k ak a, e k uk u te kerer u The tui chatters, the k ak a cackles, the kerer u coos professional regulation and educational development.Universal access is a key tenet of the philosophical approach that underpinned the establishment of the system (Goodyear-Smith and Ashton, 2019).A sense of pride is attached to universal systems of healthcare which then contribute to a sense of national identity (Garrod, 2019;Hiam et al., 2019).Aotearoa New Zealand is noted for its universal model of healthcare and for being one of the first nations to establish such a system, but inequities have become obvious (Goodyear-Smith and Ashton, 2019).Increasing evidence and the voice of health service users illustrate inequitable access and health outcomes, especially for M aori.This has been recognised in part by the New Zealand government announcing the establishment of a M aori Health Authority (Broadbent, 2022;Taylor, 2022).
The education of health professionals is a key element in the healthcare and disability support sector.Changing what and how health students learn has the potential to ensure they genuinely understand and address inequities when they graduate and become practitioners.Qualifications are founded on competencies required by healthcare Regulatory Authorities.Inequities in competencies with respect to Te Tiriti o Waitangi were identified by Came et al. (2021).Te Tiriti o Waitangi is both a founding and a living document, signed in 1840 between the British Crown and M aori as the indigenous people of the land (Tangata Whenua).Came et al. recognised considerable variation in the reviewed competency documents and suggested 'alternative competencies that could strengthen Te Tiriti engagement ' (p.35).These additional competencies can contribute significantly to improving healthcare practice and are worthy of consideration in any graduate qualification.
Creating more expeditious routes through qualifications also has the potential to enhance workforce diversity and flexibility.Currently the flexibility of the workforce is significantly reduced by the design of qualifications, in particular the time-frame required for a health professional to train (or re-train) for a related field.For example, a nurse wishing to train as a paramedic currently undertakes most of the training for the second qualification, despite considerable common knowledge and skills shared between the two disciplines.A less complex career path can also enable a health professional to practice in two fields concurrently thus potentially creating greater flexibility, especially in smaller and rural communities.It can also retain health professionals within the health workforce rather than move to an entirely different sector, if seeking a change in their work.
The Faculty of Health and Environmental Science is ideally placed to undertake this study.It provides a wider range of health professional education than any other institution in New Zealand in terms of the number of students and the range of professions offered.This experience gives a broad perspective from which to reflect on and reconsider how we educate health professionals and to enact change.Eleven of our programs prepare students for registration under.The project's specific aims are to help to address inequitable health outcomes and complex pathways that reduce flexibility of the workforce.
The complexity of the problem cannot be over-stated.A curriculum structure will be reenvisaged as part of this project.The content and the learning journey of students will be considered, with the intention to bring about change that enables improved access to care and support, thereby creating optimal health outcomes and experiences for users of health services.
Frameworks
The complexity of this issue and the multitude of important stakeholders make action research founded on Vision Mat auranga an ideal process for enabling action.
Complexity of the problem
This issue can be described as a complex problem without a 'definitive solution … because different stakeholders bring different perspectives to the problem' (Waddock et al., 2015(Waddock et al., : p.1000).Using the complex problem typology of Alford and Head (2017), change in health education is 'a politically turbulent problem' (p.402),the problem is clear, but the solution is not, 'with information fragmented and a power imbalance amongst stakeholders who have conflicting interests'.
The two dimensions in this typology are: (a) the nature of the problem and solution and (b) the people involved in problem-solving.Mid-range complexity on the first dimension arises when the nature and cause of the problem is known but it is difficult to find a sound and universally accepted solution.The second dimension significantly impacts on the ability of key people to address the problem.Three factors influence the second dimension: (a) relevant knowledge is fragmented amongst stakeholders, (b) stakeholders have conflicting interests and (c) an imbalance of power amongst stakeholders.With regard to the education of healthcare professionals, all of these factors place the issue at the most complex end of the second dimension and demonstrate why the involvement of a full range of stakeholders in developing a solution is so important.
Vision M atauranga
Vision M atauranga aims 'to envision knowledge, to think about new ways of doing things, to find answers, to solve problems' through 'the innovation potential of M aori knowledge, resources and people to assist New Zealanders to create a better future' (Ministry of Research Science and Technology, 2007: p.2).It was developed as policy by the Ministry of Research, Science and Technology and was incorporated as a pivotal strategy and policy for Crown research institutes in 2011.One of four key research themes of Vision M atauranga is Hauora/Oranga: Improving Health and Social Wellbeing.This theme draws on the Health Research Council Strategy to improve M aori health and wellbeing, including whenua (land).It aims to support research activities to 'improve tangata whenua access to quality health services' and to 'improve health service provisions to tangata whenua'.For the Hauora/Oranga theme (improving health and social wellbeing), the Vision M atauranga framework embraces both M aori and non-M aori researchers contributing to the delivery of the desired outcomes (Ministry of Research Science and Technology, 2007: p.13).
Action research
An action research approach founded on stakeholder involvement and partnership is the methodology being used for this project.Bradbury et al., 2019a, p.1) describe educational action research as 'transformative social learning with a change agenda.It shapes the world with others in a more desired direction'.Therefore, a key principle is that stakeholders in a system are actively involved in 'transforming structural forces that inhibit thriving' (Bradbury et al., 2019b: p.3.).
With an aim of working towards the 'flourishing' of people and their communities, action research is a natural fit with Vision M atauranga.It brings together 'action and reflection, theory and practice, in participation with others, in the pursuit of practical solutions to issues of pressing concern to people' (Bradbury, 2015, p. ix).Broad characteristics of action research are a collaborative and democratic partnership, an approach to problem-solving and a focus on research in action (Coghlan, 2019).Bradbury et al., 2019a highlight the importance of 'shared learning amongst people with a stake in transforming structural forces that inhibit thriving' (p.9).One of the examples discussed by Bradbury et al., 2019a is a change in a healthcare setting that needed transformation.
The approach is immensely valuable for problems that are seemingly too difficult to solve.It is a dynamic process that connects change agents and the community and also action and reflection.The action-oriented nature of the methodology results in iterative cycles of collective planning, taking action, evaluating the action, leading into further planning and action.
Stakeholders
Critical to the action research philosophy is the collaborative involvement of all members of a system.In the case of health education, overall competencies for registration of health professionals are determined by Responsible Authorities (RAs) in Aotearoa New Zealand.Regulatory Authorities are healthcare occupational organisations that register practitioners for professional practice, bound by the Health Practitioners Competence Assurance Act (2003).These RAs also accredit educational institutions, which in turn design and deliver educational programs.The graduates from these programs belong health professionals providing healthcare to the community.Each of these stakeholder groups are now discussed.
M aori community and wellbeing
Te Tiriti o Waitangi included assurances that M aori would have the same rights as British subjects along with protection of their chieftainship and treasures.Despite these undertakings, M aori have significantly poorer health outcomes and experiences in seeking care and support (Goodyear-Smith and Ashton, 2019;Palmer et al., 2019).M aori adults report high rates of most health conditions and until recently relied on these being addressed through the current healthcare provisions in Aotearoa New Zealand (Goodyear-Smith and Ashton, 2019).The new M aori Health Authority which 'will work with Iwi-M aori Partnership Boards, M aori health providers, iwi, hapu and M aori communities to understand M aori health needs across New Zealand' (New Zealand Government, 2022).The Health Futures project will be informed by directions arising from this fundamental change to the health service.
Decades ago, nursing educators in Aotearoa New Zealand embarked on a journey which resulted in the wide acceptance of 'cultural safety' into curricula (Ramsden, 1993).Cultural safety is clearly distinguishable from more generic concepts such as cultural competence (Curtis et al., 2019) and is very specific to Aotearoa.Deliberate attempts to reference M aori concepts, experiences and values into health professional education and practice have evolved in a range of disciplines including medicine (Al-Busaidi et al., 2018;Pitama et al., 2018), health promotion (Ahuriri-Driscoll et al., 2021) and others.Cultural safety is considered highly significant across professions; it is embedded in education and clearly articulated in requirements for professional registration, scopes of practice, professional competencies and expectations of employees (McKenna, 2020).
Public agencies, including the Ministry of Health, publish aspirations and plans to address inequity such as Whakamaua, the M aori Health Action Plan 2020-2025 (Ministry of Health, 2020).This plan emphasises the need for a comprehensive approach that sets goals, measures outcomes, and actively addresses issues with access, service delivery, resources and workforce.While policy and leadership are important, every element of the healthcare and disability support experience, including the physical environment (Ku Leuven, 2019), also matters to M aori.
Despite these obvious steps to ensure the M aori worldview (M atauranga) is acknowledged, and entitlement to services and equitable outcomes are addressed, there remains considerable inequality in wellbeing and life expectancy.Such systemic inequalities result from colonialism (Hobbs et al., 2019;Reid et al., 2019) and represent a failure to honour Te Tiriti o Waitangi (Came et al., 2021) in ways that are truly meaningful.The problem is less one of inaction but rather of a lack of understanding of M aori, their contexts and wellbeing, and as Wepa and Wilson (2019., p.4) have argued, 'the worldview of the New Zealand health system is predominately a Western biomedical approach that for the most part focuses on individuals and their health experiences'.
Addressing inequity in access to health and support and the resulting experiences and outcomes, requires fundamental changes to the prevailing narrow understandings of how health is defined, and to broaden this to embrace connectedness to whenua (land) and wh anau (family), spirituality and the centrality of these in wellness and wellbeing.However, the existence of policies and structures in themselves does not necessarily result in positive change (Ferdinand et al., 2020) and are insufficient in isolation from the wider context.
Lived experience of the users of health services
The voice of the health service users is formally acknowledged in the work and oversight of government agencies including the Health and Disability Commissioner, the Health Quality and Safety Commission.It is also evident in the certification of health services by the Ministry of Health.Other interest groups and agencies such as Te Pou (a national workforce centre for mental health, addiction and disability in New Zealand), CCS Disability Action, the Mental Health Foundation and Patient Voice Aotearoa also actively present the experiences of patients, caregivers and whanau in relation to service access and provision.Patient voice initiatives can serve a valuable purpose in undergraduate education (Dijk et al., 2020) and the inclusion of the service user voice in governance is an essential requirement for true stakeholder engagement (Akmal and Gauld, 2020).
Despite this recognition, people continue to experience services in ways that require attention, and their feedback about service provision does not lead to sufficient change.Several initiatives have publicly articulated the voices of service users and their call for change.The People's Mental Health Report (Elliot and Cloet, 2017) presented the perspectives of those with expertise as service users.Recommendations included an inquiry into the structure and provision of services, urgent increases to funding, and independent oversight of the system.A recent report into the experience of people living with harm caused by surgical mesh was prepared for the Ministry of Health (Wailling et al., 2019).This report illustrates how service user voices can be valued and used to inform government agencies.
Increasing representation of people with first-hand experience of health challenges is not new, in particular, people with lived experience of mental health and addiction services have been actively involved in academic settings for some time (Classen et al., 2021;Happell et al., 2014).Classen et al. (2021) further argue that increasing representation of people with first-hand experience of mental health and addiction improves student's preparedness for practice, empowers mental health consumers and has the potential to reduce stigma and improve outcomes for such marginalized groups.In summary, including the service user voice in the curriculum and learning journey of health professional students has the potential to establish relevant and meaningful understanding of issues and therefore the ability to transform their practice.
Education and health sector: current curricula
The education of health professionals and regulation of their practice is highly controlled in Aotearoa New Zealand.Individual regulators are in place for a number of professions and established as RAs under the Health Practitioners Competence Assurance Act (2003).Analyses of the requirements for educational programs and the competencies of practitioners (Shaw and Donaldson, 2020) indicate that knowledge of Te Tiriti and cultural safety in the context of Aotearoa New Zealand are expected.While these concepts are present, they tend to lack depth with an emphasis on knowledge and personal reflection (Heke et al., 2018).
A starting point is the critical analysis undertaken by Came et al. (2021) of competencies of healthcare professionals set out by the Regulatory Authorities which issue annual practising certificates to ensure practitioners work within their professional scope of practice.This critical analysis aligns to the four articles of Te Tiriti o Waitangi to assess te Tiriti compliance of the regulated competency documents.Overwhelmingly the study found that existing professional competency documents were 'not yet fit for purpose as frameworks for upholding te Tiriti' (Came et al., 2021: p.38).Therefore, the future design and delivery of health professional education programs in Aotearoa New Zealand must incorporate Te Tiriti o Waitangi as a living and foundational document and extend this to the lived experience of M aori.Paying genuine attention requires all health professionals to step outside of the prevailing western-based understandings of health and approaches to education.
Curricula changes are apparent in the education of health professionals in Aotearoa New Zealand including deliberate attempts to integrate service-user perspectives and te Tiriti, indigenous and M aori perspectives.Analyses of the success of these suggest that they remain inadequately addressed (Pitama et al., 2018).Internationally, the need to acknowledge indigenous ways of knowing and include them in curricula to address inequity and disparity has been realised (Fildes et al., 2021;Jones et al., 2019).
Current challenges to enable this include the requirement of clinical practices hours by some RAs and how these are defined in the development of the individual discipline.Restrictions on student progress through programs, such as requirements that education providers seek permission of an RA for a student to repeat a course component, are another example of the inflexibility and the degree of control within the current approach to course accreditation and monitoring.
An educationally informed evaluation of curricula, clinical learning and student pathways would focus on the learning and competence of students.The opportunity to create more flexible routes to professional registration and practice could result in a graduate from one discipline completing an additional and shorter period of learning and assessment in another related field to gain an additional registration.This would have the added benefit of valuing extended and advanced practice as a career pathway.Some 'double degrees' such as paramedicine and nursing have evolved in response to the need for a more flexible workforce that provides a wider range of care (Plummer et al., 2017), including their practical skills, and ability for creative planning and communication in a variety of clinical situations (Lee, 2020).
Redesigning curricula so that the final year of study is a 'capstone', would incorporate the highest level of skill and practice development for the beginning practitioner, and the associated teachers' strengths required for capstone design, such as assessment skills, experience and their continued connection to the health industry, and ability to work across faculties (Howe and Rosenbauer, 2017).The current graduate entry masters-level qualifications into nursing (Macdiarmid et al., 2021) and physiotherapy are good examples of how traditional approaches that individual professional registrations require separate 3 or 4-year undergraduate degrees can be re-envisaged.
Health professionals and graduates
The need for flexibility within the health and disability workforce is referenced by the World Health Organization (WHO) in its work on interprofessional learning and collaborative practice more than a decade ago (World Health Organisation, 2010).Recognition of similarities across professions (knowledge, skills, cultural safety, professional practice) and the benefits of sharing expertise and care to provide better access and outcomes (Jones et al., 2014) formed the basis of shared and common curricula elements at the Auckland University of Technology.It is common for students in health professional programs to spend time learning with, and about those enrolled in other courses of study.
The increasing recognition of the need for potential new and blended scopes of practice has not been served well by traditional approaches to workforce planning.In recent times, workforce planning in the health sector has focussed on projects relating to resources and service delivery requested by the employers (Rees, 2019).Quantitative approaches to analysing services and predicting staffing numbers are more likely to replicate existing roles and skillsets than enable new and innovative opportunities for defining services and practice.Simple approaches to workforce planning consider historical numbers within individual groups of practitioners without necessarily referencing future needs (Rees et al., 2020).
The regulation of professions through distinct RAs increases a sense of individualism and effectively creates silos which potentially work against the benefits of shared learning and practice (Shaw and Tudor, 2021).To date the approach to pre-qualifying curricula incorporates limited interprofessional learning.However, the most recent amendment to the Health Professionals Competence Assurance Act (2019) specifically references opportunities to learn and practice across professional boundaries by adding a function for RAs: 'to promote and facilitate inter-disciplinary collaboration and co-operation in the delivery of health services' (Health Professionals Competence Assurance Act, 2019, §37(3)).
Regulatory frameworks are formal and bureaucratic, arguably affording professions more power than service users (Fraher and Brandt, 2019), nevertheless interprofessional learning and service user-centred approaches to curricula development and design have the potential to emphasise service users, their experiences and outcomes.Other jurisdictions have established models that enable collaboration between regulatory agencies (Bogossian and Craven, 2020;Lahey and Fierlbeck, 2016;Leslie et al., 2018).
Additional scopes of practice have the potential to extend and enhance the service that can be delivered to communities.Postgraduate study currently emphasises research-based learning outcomes and the production of research (Bosch and Casadevall, 2017) but there is no reason why professional development could not include an extension of skillsets and recognition across professional boundaries.
Discussion
This section problematises the impact on the existing design of health professional curricula (by educational institutions) of the requirements for healthcare professional registration (by Responsible Authorities).Discussion focuses on the tensions between attesting competencies including soft skills which can arise from attitudes, beliefs and insight (Vernon et al., 2018) and mandatory technical skills, both of which are required to address improved healthcare provision within Aotearoa New Zealand.A process to inform curriculum change is therefore outlined.
Aotearoa New Zealand has a diverse and dynamic community of healthcare professionals whose education needs to genuinely reflect Te Tiriti o Waitangi and demonstrate an understanding of cultural safety as well as an appreciation of the issues our multi-cultural population face.In order to accomplish this outcome, we need people (we suggest graduates) who are working in the health and disability sector to engage in inter-disciplinary settings and across traditional boundaries (Health and Disability System Review, 2020).We need healthcare professionals to be multiskilled, adaptable and able to innovate and respond.Service user-centred practice should be at the heart of the services provided by health service professionals.Therefore, hearing the voices of communities and the people using services is essential to making meaningful progress.The Health Futures project envisages that by changing the content and process of curricula these issues could be addressed.
We believe there are opportunities to redesign curricula to enable graduates from one profession to complete the final capstone year in another related program and gain additional professional recognition (and registration).More expeditious routes through programs to registered practice can be identified by mapping competencies across programs and resolving critical questions about the detailed requirements of regulators.Changes to the structure of curricula could enable a more multiskilled and flexible workforce that could be responsive to society's needs.In order for this to occur, Regulatory Authorities would need to reconsider their policies that result in highly prescriptive compliance required in each health professional discipline, including clinical practice hours.
The existing curricula and requirements for practitioners by Regulatory Authorities already appear to reference service user needs and expectations under Te Tiriti o Waitangi.As previously discussed, the majority of frameworks from the Regulatory Authorities with respect to the articles in Te Tiriti were not compliant (Came et al., 2020).This and other issues addressed in the Health Futures project are most easily seen at the point of inequitable experiences and outcomes.Shaw and Tudor (2021) suggest that a lack of educational expertise on the part of the Regulatory Authorities has contributed to a focus primarily on distinct skills and competencies required for each profession and not on complementary skills and competencies that could be utilised across professions.Measuring what we cannot see is difficult.Meaningful elements of interactions with service users, that make a difference for them, such as cultural skills (Jongen et al., 2018), are very difficult to quantify.However, good educational design and practice at undergraduate level could alleviate this concern.
Listening and responding to the community
The action research project involves at least three iterations of conversations with community groups, specifically M aori and service users.Dialogue with the other stakeholders also informs the project.The process is iterative, consultative and facilitative.We value expertise, engagement and honesty.We also take an investigative approach because we want to learn and contribute to the wider community of health professional education developers, teachers and practitioners.For this reason, we will seek consent from those interacting with us to potentially use unidentifiable information as we document and publish information about the process, our learnings and results.
Sharing and evolving
The project team is committed to learning from existing knowledge and expertise, reflecting on the journey itself and sharing what we learn with others.An action research methodology enables participatory research, emphasising reflection and reflexivity.Action research provides an iterative framework and has previously been used to consider indigenous curricula in health professional education (Wilson et al., 2020).The project aims to create a pilot curriculum which will provide an opportunity to embed an evaluative approach, focussing on collaboration and real-world context.It is envisaged that this pilot curriculum will ensure that the competencies of graduating students will meet the requirements of both the educational institution and the Regulatory Authorities, demonstrating that collaborative practice is both iterative and organic.
Process considerations
With its strength in health professional education in Aotearoa New Zealand, including a team of health professional education experts, a range of active curricula initiatives and an extensive curriculum database (Shaw and Donaldson, 2020), Auckland University of Technology is well positioned to consider workforce education and community voice in relation to a range of regulated health professional disciplines.We acknowledge along with Wellings et al. ( 2017) that it is much more difficult to measure the result of an educational intervention on service user experience or outcome, hence the importance of embedding te Tiriti and generic competencies within an undergraduate program.We intend to listen and respond to all our identified communities as we share and evolve our project.
Re-envisaging the content, process and structure of curricula (an aim of the Health Futures project) has the potential to bring about enduring and meaningful change in the workforce and delivery of healthcare and disability support in Aotearoa New Zealand.This project has been developed in consultation with educators, M aori community, sector experts and discipline leaders, alongside conversation with the Tertiary Education Commission and Ministry of Health.
Conclusion
This action research study provides a springboard for all stakeholders to participate in addressing issues (both policy and practical) in the educational preparation of health practitioners.Equity in healthcare in Aotearoa New Zealand is valued but not yet achieved.Hence, a sector review was undertaken and demonstrated the complexity of the problem of inequity amongst service users in access to healthcare and in resultant outcomes.The review highlighted the need for stakeholders to work together to improve healthcare provision.The active involvement of stakeholders in our study (inherent in action research) brings the very real potential to contribute towards meeting the goals of service users, educational providers and the healthcare workforce.Importantly, Vision M atauranga provides a new way of considering equity to create a better future.Uniting the two frameworks of Vision M atauranga and action research can offer a way of reducing the complexity of the problem by creating a more cohesive understanding and solutions to the issues in the education of health professionals.
This paper contributes to policy discussions related to enabling graduating health professionals to be better equipped to implement change in the healthcare system.Our motivation is focussed on re-envisaging the way we educate health professionals to reflect the many voices in Aotearoa New Zealand.The ultimate goals are twofold: (1) to bring about positive change in what is taught and how it is taught, with potential to transform the practice of graduates for the changing environment and (2) to provide flexible routes to qualification and registration, which has the potential to extend the scope of practice for health professionals.This in turn can expand both the range of and access to healthcare services for the whole population.Katharine Hoskyn's family are of Cornish, Scottish and Yorkshire descent.Her ancestors include a Scot who was beheaded for leading an uprising against the King of England and farmers who were displaced during the Scottish Highland clearances when animal farming was introduced to an arable part of Scotland.After 15 years in the market research and food industries, Katharine joined Auckland University of Technology (AUT) as a lecturer and supervisor of work-integrated learning students.She has been passionately involved in Work-Integrated Learning in leadership, co-ordination and supervising roles, with considerable experience in the development, co-ordination and management of WIL programmes at AUT, as well as marketing papers.This included involvement in the organisation of papers for conjoint students taking business in conjunction with arts, sport, social science, psychology, hospitality and environmental science.Her wide-ranging research interests are focussed on not-for-profit communities.
Wellings CA, GendekMA and Gallagher SE (2017)Evaluating continuing nursing education: a qualitative study of intention to change practice and perceived barriers to knowledge translation.Journal for Nurses in Professional Development 33(6): 281-286.Wepa D and Wilson D (2019) Struggling to be involved: an interprofessional approach to examine M aori wh anau engagement with healthcare services.Journal of Nursing Research and Practice 03(3): 1-5.Wilson C, Heinrich L, Heidari P, et al. (2020) Action research to implement an Indigenous health curriculum framework.Nurse Education Today 91: 104464.DOI: 10.1016/j.nedt.2020.104464World Health Organization (2010) Framework for Action on Interprofessional Education and Collaborative Practice.https://www.who.int/hrh/resources/framework_action/en/.Retrieved from 7 February 2021.Susan Shaw has a background in education and health care having originally qualified as a primary school teacher and then a nurse.She has held various academic leadership roles since joining Auckland University of Technology in 1992.Her commitment and contribution to educational practice been recognised with the award of a Principal Fellowship from the Higher Education Academy.She has maintained clinical interest and practice in the areas of disability and palliative care; has co-edited five textbooks with Oxford University Press and continues to teach undergraduate and postgraduate students.Susan has served as the Associate Dean (Academic) in the Faculty of Health and Environmental Sciences and a Director of the National Centre for Interprofessional Education and Collaborative Practice at AUT. Denise Atkins is a Senior Lecturer and Academic Development Advisor in the Faculty of Health and Environmental Sciences at AUT University.She is a Senior Fellow of the Higher Education Academy (UK), a member of the AUT Academy (2019) and a recipient of the Vice Chancellor's Excellence in Teaching award (2016) and undertakes research in the fields of Teaching and Learning, Health and Physical Education, higher education, pedagogy, play and children's voice.As a previous senior advisor with the Ministry of Education, Denise worked in partnership with a range of government agencies with a focus on young people's wellbeing.Denise has had experience leading and managing teams and has considerable governance experience.Her expertise lies in working both with Boards and Management, as well as with teachers and academics in professional learning and teaching using evidence-based research.
|
2024-06-14T14:55:19.463Z
|
2024-05-01T00:00:00.000
|
{
"year": 2024,
"sha1": "76c28b2b9735108965783a0760578aedd77ba3e3",
"oa_license": "CCBY",
"oa_url": "https://journals.sagepub.com/doi/pdf/10.1177/14782103231176363",
"oa_status": "CLOSED",
"pdf_src": "Sage",
"pdf_hash": "76c28b2b9735108965783a0760578aedd77ba3e3",
"s2fieldsofstudy": [
"Education",
"Medicine"
],
"extfieldsofstudy": []
}
|
254663832
|
pes2o/s2orc
|
v3-fos-license
|
Synthesis of Mesoporous Silica Incorporated with Low Iron Concentration and Gelatin Co-Template via The Ultrasonication Method and Its Methylene Blue Photodegradation Performance
In this work, low iron concentration incorporated on mesoporous silica with gelatin co-template (Fe 2 O 3 /GSBA-15) has been successfully synthesized via the ultrasonication method. The physical, chemical, and structural properties of the samples were investigated with X-Ray Diffraction (XRD), Scanning Electron Microscope-Energy Dispersive X-Ray (SEM-EDX), Fourier Transform Infra-Red (FTIR), and N 2 adsorption-desorption. Results showed good distribution of low concentration of iron oxide on the gelatin mesoporous silica GSBA-15. Elemental and surface analysis presented that iron oxide incorporation with higher concentration exhibited lower surface area due to the blocking pore. The highest photocatalytic activity on the methylene blue dye degradation was achieved at 10% Fe 2 O 3 /GSBA-15 with ~80% efficiency. The results revealed that the photocatalytic activity of Fe 2 O 3 /GSBA-15 enhanced with the presence of iron oxide.
Introduction
Methylene blue (MB) is a liquid dye waste massively produced by the textile industry annually. MB waste enters the waters and poses many health risks [1]. MB waste is handled mainly through adsorption and photocatalysis [2,3]. The photocatalysis approach to reducing MB is famous as an environmentally friendly and high-efficiency method [4][5][6]. Many researchers have studied efficient photocatalysts to pro-duce an effective photodegradation process. One of the photocatalysts reviewed is iron oxide [7,8]. However, the easy agglomeration of iron oxide causes high cost and low photocatalytic efficiency [9,10]. Therefore, a suitable support material is considered effective in increasing iron oxide distribution. Besides, the support material is attempted to be synthesized from various materials that have high sustainability by minimizing synthetic materials, for example, with gelatin as the co-template [11][12][13][14]. The main problem is that the normal impregnation process still causes a large metal oxide agglomeration effect, reducing the photocatalytic efficien-cy. Hence, various impregnation modifications have been performed to deal with this problem so that metal oxides can be appropriately distributed, one of which is by the ultrasonication method [15][16][17][18].
The ultrasonication method is a simple synthesis pathway via a high-frequency sound field (>20 kHz). It is widely used in synthesizing organic and inorganic compounds because it is effective for homogenizing, mixing, and reducing particle size [16,19,20]. Ultrasonic waves have been used for the separation of agglomerates to make composites because they exhibit higher selectivity, reactivity, and improved product amount. Additionally, their energy savings are higher and cleaner than photochemical, hydrothermal, and pyrolysis [17,21,22].
However, the ultrasonication method has never been employed in the incorporation of iron oxide in mesoporous silica using gelatin as the co-template, especially at low concentrations of iron oxide. Thus, here, to prevent agglomeration and achieve better dispersion of iron oxide in silica, iron oxide precursors are impregnated on the silica surface via the ultrasonication method. Further, this research focuses on studying the effect of ultrasonication on iron oxide impregnation into mesopore silica with gelatin co-template. The synthesis results are characterized using XRD, SEM-EDS, FTIR, and Nitrogen (N2) adsorption-desorption. The synthesis results are applied in methylene blue photocatalysis to determine information related to degradation efficiency and an appropriate kinetic model.
Synthesis of Fe2O3/GSBA-15
Mesopore silica with gelatin as the cotemplate (GSBA-15) was prepared by following the previous study with a P123:gelatin ratio of 1:0.01 (%w/w) [23]. The incorporation of iron oxide into GSBA-15 was done by mixing H2O, Fe(Cl)6.6H2O, and GSBA-15 with a ratio of 2:10:100 (%w/v/w), respectively. It was then followed by ultrasonication at 100 rpm for 2 hours at a room temperature of 30 o C. The result was dried at 100 o C for 30 min and then activated with 0.5 M HCL, followed by calcination in a furnace at 600 °C for 6 hours. The resulting sample was named 1% Fe2O3/GSBA-15. The synthesis process was repeated for 5% and 10% concentration variations. Thus, there were three samples in this study, namely 1% Fe2O3/GSBA-15, 5% Fe2O3/GSBA-15, and 10% Fe2O3/GSBA-15.
Methylene Blue Photodegradation
The photocatalytic activity of Fe2O3/GSBA-15 was evaluated by methylene blue degradation. First, 200 mL of 5 ppm methylene blue solution was mixed into 50 mg Fe2O3/GSBA-15 photocatalyst. The solution was then irradiated under a UV lamp (300 W Xenon lamp at 365 nm). Samples were taken every 5 minutes of irradiation (until 90 minutes) and then measured by UV-Vis spectrophotometer at 665 nm (Shimadzu UV-3600) to determine the absorbance. Analysis of %Efficiency of methylene blue degradation was done using Equation (1).
where, C0 is the initial methylene blue concentration before irradiation, and Ct is the methylene blue concentration at t (min) after irradiation.
Characterization
Characterization was done using FTIR spectrophotometer (Nicolet 6700-Thermo Fisher Scientific), XRD (Philips X'pert, Cu-Ka 0. (2) where, D is the crystal size, λ is the wavelength used in the XRD test (1.54056 Å), B is the half-peak width in radians, and θ is the angular position of the peak formation. Besides, the crystallinity of the sample [23] can also be calculated using Equation 3.
(3) [11]. The crystal size and crystallinity degree of gelatin mesoporous silica before and after 10% Fe2O3 impregnation change from 14% to 29% (Tabel 1). It indicates that the presence of iron oxide in the silica increases the crystal size and crystallinity. Figure 2 presents the N2 adsorptiondesorption isotherm of mesoporous silica GSBA-15 before and after iron oxide incorporation. It exhibits that all samples have isotherm type IV with a hysteresis loop of type H-1 at a wide relative pressure (P/P0) in the range of 0.5 to 0.8 and with almost vertical adsorption and desorption branches as a regular character of the mesoporous material. This type is in line with the previous report [24].
Results and Discussion
The surface area of the samples decreases from 481.7 m 2 /g to 385.0 m 2 /g after incorporating 10% iron oxide (Table 1). Besides, the pore volume also reduces from 0.62 cm 3 /g to 0.585 cm 3 /g due to pore clogging by iron oxide. The decrease in surface area after the incorporation of iron oxide in this study was only 18% which is much smaller than the previous study (43%). It is considered due to the ultrasonication effect homogenizing the distribution of iron oxide in the mesopore silica. Figure 3 shows the FTIR spectra of gelatin mesoporous silica (GSBA-15) before and after the incorporation of iron oxide. It demonstrates Figure 4 depicts SEM images and particle size distribution of gelatin mesopores silica GSBA-15 before and after iron oxide incorporation. It reveals that the morphology of all mesopore silica samples is rod-shaped, with the particle size distribution between 354-757 nm. The morphology of all samples does not show significant aggregation of iron oxide. It is considered due to the ultrasonication effect, which helps distribute metal oxides on the silica surface. Nevertheless, the 10% iron oxide concentration decreases the particle size distribution. It might be because of silica shrinkage due to the intense attraction between Fe−O−Si on the surface.
Further, EDX analysis denotes that all samples experience an increase in Fe content after incorporating 1-10 % iron oxide. It has a Fe content of 0.97-5.98 % (Table 1). It indicates the success of the metal deposition process in mesopore silica by ultrasonication. Moreover, based on Figure 5, the increase in iron oxide content in mesoporous silica enhances the removal efficiency of methylene blue from 66% to 85% in 90 minutes of irradiation. It is considered due to the increase in well-dispersed catalytic sites, although in small levels, due to the effect of ultrasonication ( Figure 5, Equations (1)(2)(3)(4)(5).
Further, this result can be compared with several previous studies using non-ultrasonic methods (see Table 2), which revealed that the incorporation of low iron content (< 30%) into the support material showed an efficiency of less than 50%, while high iron content between 35-65 % achieved an efficiency of 50-99 %. High iron content is not economical for large-scale photodegradation applications, so this is an important consideration in saving the environment. The research results using ultrasonication with metal oxides, but not iron oxides, reached 99% efficiency with an average irradiation time of 150-700 minutes. From this result, the ultrasonication method on the incorporation of iron oxide into the gelatin mesoporous silica G-SBA-15 provides many advantages in increasing the dispersion of small levels of Fe particles to the silica surface, which enhances the degradation efficiency of methylene blue significantly in the first 90 minutes. For the issue of saving the environment on a large scale, this gives cost and time efficiency advantages.
The mechanism of the degradation reaction of methylene blue in the Fe2O3/GSBA-15 photocatalyst under UV light irradiation can be written as Equations (4 -8).
H2O + h + → OH * + H + (6) Methylene Blue + e − → CO2 + H2O (8) Equations (4)(5)(6)(7)(8) indicate that when the Fe2O3/GSBA-15 photocatalyst absorbs photons with energy equal to or greater than the band gap energy of Fe2O3, the electrons in Fe2O3 are excited from the valence band to the conduction band. When iron is incorporated into the silica surface [31], holes and electron pairs can be generated. Then, e − and h + can migrate to the surface of Fe2O3/GSBA-15 and will enter into a redox reaction with organic pollutants, in this case, methylene blue, on the surface. Holes and electron pairs will react with H2O or OHto produce hydroxyl radicals. These radicals are powerful oxidizing agents and the main oxidizer in the photocatalytic oxidation of methylene blue to carbon dioxide, water, and other mineralized products. Meanwhile, electrons will react with methylene blue to produce reduced products, namely CO2 and H2O. In general, the iron oxide incorporated in the gelatin mesoporous silica G-SBA-15 reveals good photodegradation performance, making this material a future photocatalyst material.
Conclusions
The incorporation of iron oxide onto gelatin mesoporous silica GSBA-15 by the ultrasonication method has been successfully investigated in this work. The uniform dispersion of iron oxide on GSBA-15 as the supporting material was achieved with ultrasonic radiation. FTIR, XRD, and EDX analysis exhibited successful incorporation of 1-10% iron oxide onto GSBA-15 with the distribution particle size ranging from 30 to 70 nm and the iron content of 0.97 -5.98 %wt. The blocking pore of 1-10% iron oxide reduced surface area from 481 to 385 m 2 /g and pore volume by 5%. The study demonstrated an increase in the photocatalytic activity with enhancing methylene blue degradation as the increasing iron oxide concentration. The degradation efficiency reached optimum at ~ 80% on 10% Fe2O3/GSBA-15.
|
2022-12-15T16:01:11.348Z
|
2022-11-28T00:00:00.000
|
{
"year": 2022,
"sha1": "83f52499ab4ad2a76a4587c44e37fe2376ab3509",
"oa_license": "CCBYSA",
"oa_url": "https://ejournal2.undip.ac.id/index.php/bcrec/article/download/16210/8293",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "002a7cf905e13e1d04f7b37bba41360a2ecb1da2",
"s2fieldsofstudy": [
"Materials Science",
"Environmental Science",
"Chemistry"
],
"extfieldsofstudy": []
}
|
229488028
|
pes2o/s2orc
|
v3-fos-license
|
Reduced Graphene Oxide on Screen-Printed Carbon Electrodes as Biosensor for Escherichia coli O157:H7 Detection
: Mixture of drinking-water supplies with sewage discharges poses disease threats in flood-stricken areas. In such exigent conditions, on-site testing of water samples is the only option, as water samples cannot be transported to laboratories owing to severely impacted transportation services. Hence, we developed a low-cost electrochemical biosensor fabricated from a screen-printed carbon electrode (SPCE) to detect E. coli O157:H7, a virulent pathogen often found in sewage discharges. We focused on understanding antigen-antibody interaction when the antibody used is not specific for E. coli O157:H7. We found that antibody immobilized on a reduced graphene oxide (rGO)–modified SPCEs distinguished between E. coli O157:H7 concentrations of 4 × 10 8 and 4 CFU/ml, with lowest current reported for 4 × 10 8 CFU/ml. In contrast, a reduced graphene oxide– modified SPCEs without antibody immobilization does not produce a prominent peak that distinguishes the highest and lowest E. coli concentrations. However, a few E. coli cells were still attached to the rGO/SPCEs in the absence of antibody, as shown in FESEM images. A processing step of differential readings from reference and active electrodes needs to be programmed into an Arduino ® microprocessor to realize a prototype of a bacteria sensor for field use.
Introduction
Rapid development and deployment of electrochemical biosensors for pathogen quantification is needed for use in flood-stricken areas where drinking-water supplies and sanitation systems are disrupted and making possible serious waterborne diseases. These outbreaks of waterborne diseases become more serious when flooding occurs in rural areas with limited resources, such the case of Kuala Krai, Kelantan, East Coast of Malaysia [1]. In the resource-constrained setting of Kuala Krai, water samples are filtered through microfiltration techniques that remove only large debris and are ineffective for removing bacteria and viruses [2,3]. However, water samples generally cannot be sent to laboratories for biochemical analysis owing to severely impacted transportation facilities. Hence, on-site testing using biosensors may be the only option in such emergency situations. Biosensors for pathogen quantification can be developed by following WHO ASSURED-affordable, sensitive, specific, user-friendly, rapid and robust, equipment-free, and deliverable to end-users-guidelines for diagnostic devices for use in resource-constrained settings [4]. In this paper, we describe a low-cost electrochemical biosensor that can be integrated into an Arduino ® microprocessor encased in a portable platform for Escherichia coli (E. coli) O157:H7 detection. To fulfil ASSURED criteria, antibodies used in the biosensors require a very specific epitope to the E. coli strain because surface proteins with similar homologs across bacterial species can lead to nonspecific binding [5]. We aim to understand the effect of E. coli O157:H7 binding to an antibody (Immunoglobulin G, IgG) not specific for E. coli O157:H7 on the measurements made by electrochemical biosensors fabricated from screen-printed carbon electrodes (SPCEs) and their suitability for use in flood-stricken areas.
Instrumentation
SPCE activation, graphene oxide reduction, and electrochemical measurements were performed using a portable pocketSTAT (IVIUM Technologies, Eindhoven, The Netherlands) at ambient temperature. The results were analysed using IviumSoft software. Screen-printed carbon electrodes (SPCEs) with Ø = 2 mm-diameter working electrodes (WE) were purchased from Pine Instrument, Grove City, Pennsylvania, USA. The reference electrode (RE) is Ag/AgCl, and the counter electrode (CE) is made of carbon. Morphological change in the modified SPCEs incubated with different E. coli concentrations was observed via field emission scanning electron microscopy (FESEM) using a Hitachi SU8020 UHR Cold-Emission FESEM available at MIMOS Technology Solutions, Sdn. Bhd., Seri Kembangan, Selangor, Malaysia.
Preparation of GO, rGO, and IgG/rGO Electrodes
All SPCEs were activated using repetitive cyclic voltammetry for 3 cycles at a scanning potential of −2.5 to 2.5 V and scan rate of 100 mV/s in 0.1 M H2SO4. GO/SPCEs were prepared by drop-casting 3 µl GO onto the WE of an activated SPCE and dried at ambient temperature for 2 h. To prepare rGO/SPCEs, GO/SPCEs were electrochemically reduced using repetitive CV for 5 cycles at a scanning potential of 0 to −1.5 V, and scan rate of 100 mV/s in PBS, pH 5.0. Then, 3 µl 1-ethyl-3-(3dimethylaminopropyl) carbodiimide (EDC; 0.5 M) and N-hydroxysulfosuccinimide (NHS; 0.1 M) (ratio: 1:1) composite was drop-cast onto the WE of a rGO/SPCE, and dried at 37°C for 20 min. Three (3) µl IgG (2 µg/ml) was drop-cast onto the WE of a rGO/SPCE with EDC-NHS and incubated at 37 °C for 2 h. The electrode was rinsed with PBS, pH 7.1, to remove unbound IgG. All modified electrodes were stored in a 90 × 15-mm (diameter × height) Petri dish and kept at ambient temperature until needed for further use.
Linear Sweep Voltammetry (LSV) Measurements
LSV was performed at a scanning potential of −0.5 to 1 V, and scan rates of 25, 50, 100, 125, 150, and 200 mV/s in 0.01 M PBS, pH 7.1, containing 5 mM K3Fe(CN)6: K4Fe(CN)6 to determine the current flow in SPCEs, GO/SPCEs, rGO/SPCEs, and IgG/rGO/SPCEs. The half reaction that occurs at the electrode can be described by Equation (1) [6]: From the peak current obtained from LSV measurement, the graph of peak current (Ip) vs. square root of scan rate (v 1/2 ) was plotted, and effective surface area was calculated from only the slopes of linear curves using the Randles Sevcik Equations (2) and (3) where the diffusion coefficient D for the ferricyanide/ferrocyanide redox couple is typically 6.70 × 10 −6 cm 2 /s, C is the molar concentration of ferricyanide-ferrocyanide solution (5 mM), A is the effective surface area (cm 2 ), v is the scan rate (mV/s), and is the number of moles of electrons transferred per mole of electroactive species, which in this case is one (1). The nonspecific binding of E. coli O157:H7 to rGO/SPCEs and IgG/rGO/SPCEs was tested using LSV. To prepare E. coli solution (4 × 10 8 , 4 × 10 7 , 4 × 10 6 , 4 × 10 5 , 4 × 10 4 , 4 × 10 3 , 4 × 10 2 , 4 × 10 1 , and 4 CFU/ml), the stock solution (3.98 × 10 9 CFU/ml in dextran solution) was diluted with PBS, pH 7.1. Then, 3 µl diluted E. coli solution was drop-cast onto the WEs of IgG/rGO/SPCEs and rGO/SPCEs and incubated at 37 °C for 1 h. Both SPCEs were washed with PBS, pH 7.1, to remove unbound E. coli. For SPCEs incubated with E. coli, LSV was performed at a scanning potential of −0.5 to 1 V and a scan rate of 100 mV/s in tap water.
Field Emission Scanning Electron Microscopy
The morphology of SPCEs incubated with E. coli was observed using a FESEM. The SPCEs were coated with platinum to avoid charging effects during FESEM imaging. FESEM images were obtained at a magnification of 5K at all four quadrants of the WE to provide an overall overage of sensing area. 6. We compared the current values obtained at 100 mV/s to understand electron-transfer behavior on the bare SPCEs, GO/SPCEs, rGO/SPCEs, and IgG/rGO/SPCEs. At 0.7 V, the bare SPCEs recorded higher current value of 14 µA compared to the GO/SPCEs (11 µA), because of the oxygenated functional groups that impair electron flow at the edges of GO [9,10]. The current value was improved to 23 µA for rGO/SPCEs owing to the electrochemical reduction process that removed a considerable number of oxygenated functional groups. Studies have shown that electrochemical reduction with precise control over the reduction potential and time [11,12] allows reproducibility of rGO for use in electrochemical devices [13]. We found that IgG/rGO/SPCEs achieved the highest current value of 73 µA, which could be due to the semiconductive nature of protein in dry and wet states [14]. In addition, the conductivity can be influenced by the way in which IgG is immobilized on the electrode surface such that the alpha domain is more conductive than the beta domain of a protein [15]. Nanomaterials increase the effective surface area (Aeff); higher Aeff leads to improved sensitivity of biosensors. We plotted a graph of current (Ip) vs. square root of scan rate (v 1/2 ) to determine Aeff. The Aeff of bare SPCEs and GO/SPCEs were 21.7 mm 2 and 4.06 × 10 -9 mm 2 ; the graphs of current (Ip) vs. square root of scan rate (v 1/2 ) were linear. However, the graphs of current (Ip) vs. square root of scan rate (v 1/2 ) for rGO/SPCEs and IgG/rGO/SPCEs were nonlinear, and hence we did not calculate Aeff. This nonlinearity can be the result of a heterogenous electron-transfer process [16]. Figure 2 shows LSVs over a potential range of −0.5 to 1 V and scan rate of 100 mV/s for IgG/rGO/SPCEs and rGO/SPCEs incubated with E. coli concentrations that ranged from 4 × 10 8 to 4 CFU/ml. We compared the current values obtained for the above-mentioned SPCEs; high current value indicates lower binding of E. coli cells to the IgG or the rGO surface. We performed preprocessing of the signal by subtracting the current value of zero E. coli from that of each E. coli concentrations.
LSVs at Different E. coli Concentrations
In Figure 2a, a prominent peak was observed in the region of −0.35 to 0.07 V, and the graphs for E. coli concentrations of 4 × 10 8 and 4 CFU/ml have a distinct gap in current values that enables us to differentiate between the highest and lowest concentrations. Incubation with a higher concentration of E. coli reported lower current value, implying a higher number of E. coli cells attached to IgG/rGO/SPCEs. We found that current value does not increase linearly as the concentration of E. coli cells is reduced, which could be due to the random attachment of E. coli to IgG with low affinity for E. coli O157:H7. In addition to affinity, several IgGs could also potentially bind to a bacterium, which might lead to random orientation of the E. coli cells on the surface of the electrode. IgG could bind to several outer membrane constituents of the E. coli, including its lipopolysaccharides and porins. Since the IgG used in this study is not specific to E. coli O157:H7, a condition of nonspecific binding can be expected. However, we found that binding of E. coli to IgG occurred even at E. coli concentrations of 4 CFU/ml. Therefore, a nonspecific IgG can be used to detect but not quantify E. coli O157:H7 in the event of unavailability of antibody specific for E. coli O157:H7.
On the other hand, we found rGO/SPCEs do not produce any distinctive peak between E. coli concentrations of 4 × 10 8 and 4 CFU/ml (Figure 2c). The LSV graphs overlapped for most of the E. coli concentration range, implying a random attachment of E. coli to the rGO surface. Studies have proven that rGO surfaces in general repel E. coli attachment [17], which could result in the random attachment of E. coli to the rGO surfaces. Figure 3 shows FESEM imaging of IgG/rGO/SPCEs incubated with (a) 4 × 10 7 and (b) 4 × 10 5 CFU/ml E. coli. Higher amount of E. coli was attached to IgG/rGO/SPCE for 4 × 10 7 CFU/ml than for 4 × 10 5 CFU/ml. The attachment of E. coli influences current value obtained during LSV; higher rate of attachment results in lower current value. We found that IgG/rGO/SPCEs incubated with 4 × 10 7 CFU/mL reported lower current value than did those with 4 × 10 5 CFU/mL, confirming that E. coli attachment blocks electron flow to the electrode surface. However, IgG/rGO/SPCEs incubated with 4 × 10 3 CFU/ml reported higher current value than 4 CFU/ml, implying E. coli attaches randomly to IgG and making the biosensor suitable only for E. coli O157:H7 detection and not quantification. We observed nonspecific binding of E. coli O157:H7 to rGO/SPCEs (Figure 2 c). Studies have shown that E. coli O157:H7 attachment is dependent on orientation of the graphene structure; GO extracts phospholipids from the E. coli cell membrane onto the basal planes, which leads to bacterial cell death [18]. However, the basal plane of GO readily absorbs a variety of molecules via noncovalent interactions that quench the antibacterial property of GO [19]. In this study, we presume that molecules from the dextran solution (a medium used to store E. coli O157:H7) could possibly mask the basal plane, and thus E. coli are not destroyed when they adhere to rGO/SPCEs.
Conclusions
We developed a low-cost electrochemical biosensor using antibody not specific towards E. coli O157:H7 immobilized on a screen-printed carbon electrode. The biosensor was able to distinguish between E. coli O157:H7 concentrations of 4 × 10 8 and 4 CFU/ml, and hence allows the use of nonspecific IgG for E. coli O157:H7 detection but not quantification in the event of unavailability of antibody specific for E. coli O157:H7. Since E. coli O157:H7 attaches to rGO/SPCEs, an additional processing step is required wherein readings from reference electrode (without antibody) is subtracted from the working electrodes (with antibody). Such a processing step can be programmed in an Arduino ® microprocessor and the biosensor can be made portable for use in flood-stricken areas.
Author Contributions: N.A.A. planned and performed the experiments, P.R. Bdrafted,wrote the manuscript, and validated the experimental results. W.W.A., W.S., the Principal Investigator, acquired funding for the work, provided the initial idea, contributed intellectually to the planning of the experimental design, validated the experimental results, edited and revised the manuscript content and writing, and approved the submitted version.
Funding: This research was funded by nanoSkunkWorkX Sdn. Bhd and the The Royal Society Travel Grant.
Acknowledgments:
The authors would like to thank members of Amani Research Group for their assistance during preparation of electrodes.
Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analysis, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. The preprints for this research is available for public viewing, and can be assessed at doi: 10.20944/preprints201810.0631.v1.
|
2020-11-26T09:06:52.016Z
|
2020-11-02T00:00:00.000
|
{
"year": 2020,
"sha1": "593c0cb6ab9efbd03f09cf406608a37f6c852ab4",
"oa_license": "CCBY",
"oa_url": "https://sciforum.net/manuscripts/7056/manuscript.pdf",
"oa_status": "HYBRID",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "d4688b9d0ddbd80158539bfbbf581044225af061",
"s2fieldsofstudy": [
"Environmental Science",
"Engineering",
"Biology",
"Chemistry"
],
"extfieldsofstudy": [
"Chemistry"
]
}
|
268265100
|
pes2o/s2orc
|
v3-fos-license
|
Burnout and depression in college students
of
Introduction
Burnout has long been recognized as an occupational hazard in various human service occupations (Lacy and Chan, 2018; Maslach and Leiter, 2016;Sullivan et al., 2022;Weigl, 2022).Since its emergence, the concept of burnout has attracted increasing attention worldwide and has gradually expanded to include a more general range of occupations (Aronsson et al., 2017;Embriaco et al., 2007;Shanafelt et al., 2019).Burnout is now defined as a psychological syndrome, caused by prolonged occupational stressors.Additionally, personality traits, particularly neuroticism, play an important role in the development of burnout (Angelini, 2023).People with high levels of neuroticism may adopt maladaptive coping strategies that magnify the impact of adverse events in the workplace, leading to greater anxiety and exhaustion from work.In summary, both external and internal factors influence the development of burnout.
The widely accepted conceptual model of burnout is a multidimensional model that stratifies burnout into three dimensions: exhaustion, cynicism or depersonalization, and a lack of professional efficacy (Maslach and Leiter, 2016).However, in contrast to the increasing research on burnout, minimal advances have been achieved in the establishment of clinical diagnoses.An undetermined definition of burnout is reportedly the primary issue for a diagnostic consensus (Rotenstein et al., 2018).The evolving definitions of burnout contribute to variability in estimating its prevalence, posing challenges in policy development.Initially, exhaustion was identified as the most central and obvious manifestation of burnout (Maslach et al., 2001).Accordingly, in a few northern European countries, the diagnosis of burnout has been simplified to exhaustion; consequently, the focus of the public and policy has been limited to exhaustion alone.Meanwhile, some argued that despite the importance of exhaustion, it was not the most central component of burnout (Maslach and Leiter, 2008) and that focusing on exhaustion devalued the significance of burnout as a distinct construct.A recent study on the relationship between the three dimensions of burnout suggested that exhaustion is not the most closely associated dimension (Wu et al., 2021).Measurements using an alternative burnout structure exhibit better validity (Shoman et al., 2021).To date, there has been minimal consensus on the definition or dimensional structure of burnout, as well as the most important symptoms of burnout (Tavella et al., 2021).
The association and distinction between burnout and mental illness have been another issue since burnout was first proposed.Much of the related work has focused on depression (Meier and Kim, 2022;Schonfeld and Bianchi, 2021;Schonfeld et al., 2019;Verkuilen et al., 2021).On the one hand, burnout is considered as a gradation or a prodromal stage of depression under the same construct (Bianchi, 2020;Bianchi et al., 2015).Supporting this hypothesis, evidence has indicated a high rate of comorbidity between burnout and depression (Youssef, 2016), particularly demonstrating a positive correlation, especially between the exhaustion dimension and depression (Chen and Meier, 2021;Dyrbye et al., 2014;Ma et al., 2022;Sun et al., 2022).Some studies have suggested that burnout and depression influence and predict each other through circular effects (Ahola and Hakanen, 2007;Toker and Biron, 2012).A growing body of empirical evidence recognizes burnout as a concept that largely overlaps with depression (Schonfeld and Bianchi, 2021).On the other hand, burnout represents a quality distinct from depression (Koutsimani et al., 2019).Studies supporting this idea suggest that exhaustion-related items inflate the association between burnout and depression (Koutsimani et al., 2019;Maslach and Leiter, 2016) and that the high correspondence between burnout and depression reflects only the conceptual redundancy of the measures (Maslach and Leiter, 2016).In fact, the cynicism and professional efficacy dimensions of burnout do not show a strong link with depression (Hakanen and Schaufeli, 2012;Steinhardt et al., 2011).Therefore, as a distinct variable, burnout involves content that does not overlap with depression.Burnout is thought to be job-and situation-specific, whereas depression is more general and independent of context (Maslach and Leiter, 2016).In conclusion, the debate on whether burnout is theoretically and empirically distinct from depression has gained prominence with evidence from both sides.
Previous research on burnout-depression relationships has been based mainly on the latent variable theory, which assumes that a genuine cause of a mental disorder and its symptoms exist.Network analysis has recently emerged as an alternative way of conceptualization (Borsboom, 2017).In the network model, a mental disorder is characterized as a cluster of symptoms that directly relate to each other, with no prior assumption of a common cause.The network comprises symptoms (nodes) and their causal relationships (edges).Given that mental disorders are empirically diagnosed without hard boundaries, network analysis allows for a more comprehensive conceptualization of mental disorders and their symptoms.Thus, we can observe the symptoms with the highest centrality in one disorder and verify the dimensional model of one concept (Christensen et al., 2023).More importantly, in the study of comorbidities, we can identify the symptoms that bridge the two disorders and maintain the comorbidity (Cramer et al., 2010).Thus, network analysis provides valid supporting evidence for the latent variable model.Therefore, to identify the central symptoms of burnout, provide evidence for a conceptual model of burnout, and distinguish burnout from depression, it is necessary to establish a network structure at the symptom level.
Burnout has been studied mainly in health-service occupations; however, academic burnout among college students has been a growing concern (Frajerman et al., 2019;Rotenstein et al., 2016).Although students are not employed in a traditional work setting, the structured and mandatory activities they engage in, such as attending classes and completing assignments, can be considered as a form of "work".Moreover, the academic pressure associated with passing exams can act as a chronic stressor.Academic burnout can persist through college and predict negative outcomes such as alcohol abuse (Jackson et al., 2016) and suicidal ideation (Rotenstein et al., 2016).However, despite the high prevalence of academic burnout among college students and its coexistence with depression (Dyrbye et al., 2014;Paro et al., 2014), limited studies have focused on the burnout-depression relationship using network analysis.
Therefore, to explore the conceptual model of burnout and its overlap with depression in college students, we constructed three networks (burnout, depression, and their co-occurrence) using network analysis.The aims of our study were threefold: 1) to demonstrate the central symptoms of burnout; 2) to identify the association and distinction between burnout and depression; and 3) to provide evidence of a single-or multiple-dimensional structure of burnout.
Participants
This study was conducted among college students in Xi'an City from February 6th to February 20th, 2023.Given that WeChat has the widest range of users in China and is used by almost every college student, we distributed and collected questionnaires online via the WeChat platform.Snowball sampling, a non-probability sampling method, was adopted to optimize the number and variety of study participants.We used a respondent-driven sampling method that relied on initial respondent referrals to generate additional respondents.First, we randomly selected college students to participate in the survey and anticipated that they would forward the questionnaire to as many students as possible.Since our questionnaire evaluation was completed online, we designed and included two attention checking items in the questionnaire to better ensure the quality of the measure (for example, "Please choose the first option in this question").Current college students who had no self-reported history of neurological or mental illness and volunteered to participate were included in this study.Participants were excluded if they failed to provide the required information in the questionnaire or if they failed to correctly answer the checking items.Consequently, 1096 college students participated in this survey, and seven questionnaires were excluded because of incomplete basic information or questionnaire content.Ultimately, 1089 valid questionnaires were collected for the analysis.This study was approved by the Ethics Committee of the First Affiliated Hospital of the Fourth Military Medical University (No.KY20234188-1) and complied with the Declaration of Helsinki.Informed consent was obtained from all the participants at the beginning of the online questionnaire.
Measures
Burnout was assessed using the Maslach Burnout Inventory (MBI)-Student Survey.The MBI was the first standardized tool developed to assess burnout, exerting a pivotal influence on burnout research (Bianchi et al., 2015).Other instruments developed for burnout measures have also shown better validity than the MBI (Shoman et al., 2021).
X. Wang et al.However, because the MBI was based on the dimensional structure proposed by Maslach, it was the most suitable measure for the purpose of our study and for comparing our results from a network perspective with those of other studies using the MBI.For application to a variety of occupations and groups of people, the MBI has been divided into several subtypes, one of which is a Student Survey version for college students.Its validity and reliability were recently verified (Portoghese et al., 2018).Consistent with the MBI, the MBI-Student Survey has a three-dimensional structure that includes exhaustion, cynicism, and professional efficacy, with 15 items scored on a 7-point scale.The reliability and validity of the MBI-Student Survey for Chinese students have also been confirmed (Hu and Schaufeli, 2009;Liu and Cao, 2022).
Depression symptom severity was measured using the Patient Health Questionnaire-9 (PHQ-9).The PHQ-9 is one of the most commonly used questionnaires in depression research and measures the frequency of depression symptoms over the prior two weeks.It contains nine items and adopts a 4-point scoring (from "not at all" to "almost every day").Previous studies have confirmed the reliability and validity of the PHQ-9 in the Chinese population (Du et al., 2017;Wang et al., 2014).Higher scores indicate more severe symptoms of depression.
Network estimation and visualization
The R package "qgraph" was used to compute networks, based on Spearman rho correlations (Epskamp and Fried, 2018;Isvoranu and Epskamp, 2023).The Graphical Least Absolute Shrinkage and Selection Operator technique was used to regularize the partial correlations in the networks (Friedman et al., 2008).By shrinking all edges and penalizing edges with trivially small partial correlation coefficients to zero, this regularization process aided in eliminating spurious edges, resulting in a more stable, sparse, and interpretable network (Epskamp and Fried, 2018;Friedman et al., 2008).In addition, the Extended Bayesian Information Criterion hyperparameter γ was set to 0.5 to balance sensitivity and specificity (Epskamp and Fried, 2018;Foygel and Drton, 2010).The visualization of networks was based on the Fruchterman-Reingold algorithm (Fruchterman and Reingold, 1991).
Central and bridging symptoms
Network centrality is calculated to quantify the extent to which a node is directly connected to other nodes in the network.Expected influence (EI), defined as the sum of all edges extending from one given node to the remaining nodes in the network, is one of the most commonly used indices of centrality.This indicates greater stability, because it considers the presence of negative associations (Robinaugh et al., 2016).Higher EI values indicate greater centrality or importance of the symptoms.Thus, we computed the EI for burnout network and depression network using the R package "qgraph" (Epskamp et al., 2012).
A co-occurrence network was constructed to explore the association between academic burnout and depression.In this network, the bridge expected influence (BEI) was calculated to evaluate the bridging symptoms linking depression and academic burnout.The BEI is the aggregate of edge weights connecting a single node to all nodes in the other community of the co-occurrence network.A higher BEI value indicates a greater likelihood of one symptom activating symptoms in another community.The BEI values were calculated using the R package "networktools" (Jones et al., 2021).
Network stability and accuracy
To determine the robustness of the centrality indices, we applied a case-dropping bootstrap method (with 2000 bootstrap samples) and computed the correlation stability (CS) coefficient.The CS coefficient represents the maximum percentage of sample cases that can be dropped from the original full cases while retaining a correlation of 0.7 in at least 95 % of the samples (Yang et al., 2022).A CS coefficient should not be less than 0.25, and a CS coefficient above 0.5 indicates decent network stability (Epskamp et al., 2018).The accuracy of the edge weights was assessed using a 95 % confidence interval (with 2000 bootstrap samples) for each edge within the network.A narrower 95 % confidence interval indicates a more reliable network.Moreover, we conducted bootstrap difference tests (with 2000 bootstrap samples) on edge weights, EI, and BEI.The aforementioned procedures were conducted using the R package "bootnet" (Epskamp et al., 2018).
Exploratory graph analysis
To explore the dimensional structure of burnout generated by factor analysis and to test for the burnout-depression overlap, we performed exploratory graph analysis (EGA) on the burnout network and burnoutdepression co-occurrence network.The Walktrap algorithm (Pons and Latapy, 2006) was used to detect communities of burnout and depression symptom nodes.It is designed to identify densely connected subgraphs, also called communities, in a graph using random walks.The principal concept is that short random walks tend to remain in the same community.The EGA has proven to be as accurate or even more accurate than factor analysis (Golino and Epskamp, 2017).A bootstrap EGA with 2000 iterations was used to examine the stability of the identified dimensions and items (Christensen and Golino, 2021).Dimensional stability was estimated using structural consistency, which was the proportion of times that the initial EGA-derived dimensions were recovered from the replicate bootstrap samples.Item stability was estimated as a measure complementary to structural consistency, which was the proportion of times each item was placed in each EGA-derived dimension.EGA was performed using the "EGAnet" R package.
Sample description
Descriptive demographic information of the sample population is shown in Table 1.In general, most participants were young adults (aged 17-24 years), and more than half were not the only children in their families.Moreover, most were raised in a non-single-parent family.
Burnout network
The burnout network and EI values for each node are shown in Fig. 1 and Table 2. Symptoms representing higher EI values played a central role in the network.Among all burnout symptoms, the node "I have become less interested in my studies since enrolling at the university" (CY1, EI = 1.255) had the highest EI value within the network, followed by "I have learned many interesting things during the course of my studies" (EF5, EI = 1.101) and "I believe that I contribute effectively in the classes I attend" (EF2, EI = 1.095).While "I doubt the significance of my studies" (CY4, EI = 0.612) and "I can effectively solve study-related problems" (EF1, EI = 0.614) had the lowest EI values.The CS coefficient of the node EI values was 0.75, indicating that the estimated EI values in the network were adequately stable (Figure S1).Additionally, bootstrap difference tests on the EI values showed that the three nodes with the highest EI values were significantly different from approximately 57 %− 100 % of the other symptom nodes in the network (Figure S2).
The strongest edges in the burnout network arose between "I have learned many interesting things during the course of my studies" and "In class, I feel confident in my ability to get things done" (EF5-EF6, weight = 0.514); between "I have become less interested in my studies since enrolling at the university" and "I have become less enthusiastic about my studies" (CY1-CY2, weight = 0.474); and between "I believe that I contribute effectively in the classes that I attend" and "In my opinion, I am a good student" (EF2-EF3, weight = 0.442).The 95 % bootstrap confidence intervals of the edge weights are shown in Figure S3 and indicated that the edge weights were relatively stable and accurate.Moreover, a bootstrap difference test indicated that the three strongest edge weights were significantly different from most other edge weights (Figure S4).
Depression network
The depression community structure and EI values for each node are shown in Fig. 2 and Table 2.The higher the value, the stronger the association of that symptom in the network.Among all depressive symptoms, the symptom node "fatigue" (D4, EI = 1.079) had the highest EI value in the network, followed by "depressed mood" (D2, EI = 0.993) and "hopelessness" (D6, EI = 0.975).In contrast, "suicidal ideation" (D9, EI = 0.441) and "sleep-related problems" (D3, EI = 0.735) had the lowest EI values within the network.The CS coefficient of the node EI values was 0.75, suggesting a sufficiently stable estimation of EI values in this network (Figure S5).Moreover, bootstrap difference tests showed that the three nodes with the highest EI values were significantly different from approximately 62.5 %− 75 % of the other symptom nodes in the network (Figure S6).
The strongest edges in the depression network emerged between "anhedonia" and "depressed mood" (D1-D2, weight = 0.326); between "sleep-related problems" and "fatigue" (D3-D4, weight = 0.262); and between "anhedonia" and "fatigue" (D1-D4, weight = 0.220).The bootstrap 95 % confidence intervals of the edge weights are shown in Figure S7 and indicated that the estimates of the edge weights were stable and accurate.The bootstrap difference test results are shown in Figure S8.
Co-occurrence network of burnout and depression
The structure of the co-occurrence network is shown in Fig. 3.In the co-occurrence network, we mainly focused on symptom nodes that bridge burnout and depression, as indicated by their BEI values (Table 2).The node "anhedonia" (D1, BEI = 0.161) had the highest BEI value, followed by "fatigue" (D4, BEI = 0.109) and "I doubt the significance of my studies" (CY4, BEI = 0.107).The results indicated that "anhedonia" (D1) and "fatigue" (D4) had the strongest contagion for the onset of burnout symptoms and that "I doubt the significance of my studies" (CY4) had the strongest contagion for the onset of depression symptoms.The CS coefficient of node BEI was 0.52, suggesting that the estimates of the node BEI values were sufficiently stable (Figure S9).The bootstrap difference test results for the BEI values are presented in Figure S10.
The network edges with the largest edge weights occurred within the burnout community.The edges between "I have learned many interesting things during the course of my studies" and "In class, I feel confident in my ability to get things done" (EF5-EF6, weight = 0.496); between "I have become less interested in my studies since enrolling at the university" and "I have become less enthusiastic about my studies" (CY1-CY2, weight = 0.459); and between "I believe that I contribute effectively in the classes that I attend" and "In my opinion, I am a good student" (EF2-EF3, weight = 0.427) had the largest weights.While, "I feel used up at the end of a day at university" and "fatigue" (EX2-D4, weight = 0.044) was the strongest edge connecting the burnoutdepression community.The bootstrap 95 % confidence intervals of the edge weights (Figure S11) indicated that the edge weights were relatively stable and accurate.Bootstrap difference tests showed that the edges with the strongest weights were significantly different from most of the edges (Figure S12).
Community detection
Three communities were found in the burnout network (Figure S13): an exhaustion community (from EX1 to EX4), a cynicism community (from CY1 to CY4, plus EX5), and a professional efficacy community (from EF1 to EF6).These corresponded to the three dimensions of burnout, except for EX5, which was included in the cynicism community.In the co-occurrence network, four communities were identified (Figure S14): a depression community (from D1 to D9), an exhaustion community (from EX1 to EX4), a cynicism community (from CY1 to CY4, plus EX5), and a professional efficacy community (from EF1 to EF6).The communities within which burnout symptoms occurred were consistent between the two networks, and no overlapping symptoms were observed between the burnout and depression communities.Bootstrap EGA validated the above results (Figure S15 and S16), showing excellent dimension (Table S1) and item stabilities (Figure S17 and S18).Specifically, the dimension stability was 1.000, 1.000, and 1.000 in the burnout network and 1.000, 1.000, 1.000 and 0.962 in the co-occurrence network, respectively.
Discussion
In this study, we examined a conceptual model of burnout and investigated its overlap with depression among college students.Unlike the previous research that primarily focused on exhaustion, we found that symptoms of cynicism and professional efficacy had the most central role in burnout, highlighting the importance of "interest in studies" and "sense of contribution".According to the burnout-depression cooccurrence network, depressive symptoms "anhedonia" and "fatigue" had the strongest contagion for the onset of burnout.Similarly, the symptom of burnout most likely to trigger depression was also from the cynicism dimension "I doubt the significance of my studies".In addition, community detection revealed three communities of burnout as established by factor analysis, with no overlap between burnout and depression.Taken together, the findings supported the multidimensional structure of burnout and suggested that it was a concept distinct from that of depression.Additionally, cynicism, rather than exhaustion, was the key component of burnout and the burnout-depression 2 for the symptoms assigned to the respective items. 2 for the symptoms assigned to the respective items.comorbidity.
In the burnout network, we explored the symptoms that were most closely associated with the pathogenicity and maintenance of burnout.The most central symptoms emerged at nodes CY1, EF5, and EF2.Consistent with a previous network analysis of depression and burnout in Chinese nurses (Wu et al., 2021), we found that the most associated symptom was cynicism (CY1) rather than exhaustion.These results consistently indicate the overlooked value of cynicism in causing and maintaining overall burnout symptoms.In other words, apart from emotional exhaustion, the lack of motivation, diminished empathy, and a passive approach to studying may be more crucial factors contributing to burnout in students.Additionally, symptoms related to professional efficacy (EF2 and EF5) were among the most central symptoms, emphasizing the significance of academic achievement in burnout among college students.These results indicated that students' interest in their studies or work (CY1 and EF5) and their sense of contribution (EF2) were essential.Hence, the symptoms that played the most central role were potential treatment targets.Based on the network model, stimulating interest and increasing one's sense of contribution may be effective interventions to alleviate burnout among college students.Moreover, "fatigue" (D4) was the most central symptom of depression, which underscored the importance of fatigue-related symptoms in depression, whereas exhaustion symptoms did not hold the same level of significance in the burnout network.
In the burnout-depression co-occurrence network, we examined the bridging symptoms that directly communicated between burnout and depression.We found that fatigue-related symptoms "anhedonia" and "fatigue" in the depression community had the strongest connection with burnout symptoms.This is in accordance with previous results showing that exhaustion-or fatigue-related symptoms are closely correlated with depression and burnout (Bianchi et al., 2015;Chen and Meier, 2021).Surprisingly, in the burnout community, symptoms that directly communicated with depression did not arise from exhaustion.Instead, cynicism symptom "I doubt the significance of my studies" (CY4) had the closest connection with depressive symptoms.This finding also contradicts the previous consensus that the conceptual overlap between burnout and depression lies primarily in the exhaustion dimension.Therefore, among students, the most significant causal relationship between burnout and depression was the loss of meaning in studies, rather than feeling exhausted.Additionally, the loss of meaning in studies aligned with one of the core depressive symptoms, the loss of meaning in life.Furthermore, item CY4 had the lowest EI value in the burnout network but one of the highest BEI values in the co-occurrence network.This indicated that the most central component of burnout symptoms and the bridging symptoms between burnout and depression did not overlap.
One of the crucial questions we attempted to resolve was the relationship between burnout and depression.By combining the aforementioned results and analyses, we concluded that burnout was a distinct construct rather than a largely overlapping concept with depression.In addition to the finding that there was no overlap in symptoms between burnout and depression communities, three reasons support this conclusion.First, most central and bridging burnout symptoms were not related to exhaustion or fatigue.Based on empirical evidence, redundancies or overlaps between burnout and depression mainly occur in the exhaustion dimension (Maslach and Leiter, 2016).Additionally, the most central and bridging symptoms of depression observed in this study were related to fatigue.However, our results emphasize the importance of cynicism symptoms in burnout and the burnout-depression comorbidity, which is not consistent with previous assumptions.Second, the symptom that most closely connected burnout to depression (CY4) in the co-occurrence network had the lowest centrality in the burnout network.It was assumed that burnout and depression overlapped excessively and that burnout symptoms influencing depression would at least have decent centrality in the burnout network.In contrast, the most connecting symptom in the co-occurrence network was the most marginalized symptom in the burnout network, which did not support this assumption.Finally, in the co-occurrence network, the strongest edges occurred within the burnout community, while the edge weights between the burnout and depression communities were considerably lower.If a large overlap existed between burnout and depression, the edges between burnout and depressive symptoms, which represent direct correlations, should have been notable.Instead, the strongest and most distinguishable edges emerged in the burnout community.In summary, the correlation between burnout and depression was not as strong as expected.
Finally, the third question this study aimed to answer was whether a single-or multidimensional structure for burnout should be applied.The results of the study suggest that a multidimensional burnout structure is appropriate.The main reason for this conclusion was that community detection revealed three separate communities under the burnout construct, which coincided with the three dimensions proposed by the 2 for the symptoms assigned to the respective items.factor analysis.Although EX5 was included in the cynicism community, which may be owing to the characteristics of college students, the remaining symptom nodes remained in the communities corresponding to their original dimensions.Additionally, in the co-occurrence network, symptoms from the three burnout dimensions had different bridging correlations with depression symptoms, suggesting that their different comorbidity mechanisms underlie these three dimensions.For example, the strongest bridging connection between exhaustion and depression symptoms was EX2-D4, whereas in the cynicism and professional efficacy communities, the strongest connections emerged in CY3-D7 and EF6-D7.The presence of different cross-community edges indicated that symptoms of different dimensions had different pathways for influencing depression, which also supported the conclusion that burnout had more than one dimension.
Several limitations need to be considered when interpreting the study results.First, most of the study participants were male.Sex-related perspectives toward studies may be a confounding factor in academic burnout symptoms and their relationship with depression.Second, the sampling method prevented us from obtaining a complete picture of college students' burnout.As we adopted a dimensional method to conceptualize burnout and depression, the results may lack clinical implications for those who meet the diagnostic criteria.Future studies that focus on categorical methods may yield novel findings.Additionally, this cross-sectional study lacked longitudinal observations.Future studies should include more diverse and comprehensive study participants and a follow-up design to advance our research.Finally, the network structure was limited to the selected scales.Given that the MBI is not the only or most reliable instrument for measuring burnout (Shoman et al., 2021), further validation of the burnout structure and its overlap with depression using other instruments is required.
Conclusion
The current study was designed to determine the structure of burnout and the burnout-depression overlap among college students at the network level.It specifically identified cynicism symptoms as crucial factors for burnout and burnout-depression comorbidity.Additionally, community detection analysis suggested a multidimensional structure for burnout, emphasizing that burnout should be considered distinct from the concept of depression.This study provides insights into the structure of burnout from the perspective of network analysis, and delineates the overlap between burnout and depression.
Fig. 1 .
Fig. 1.Burnout network and node EI values.The three different node colors represent three dimensions of burnout: exhaustion, cynicism, and lack of professional efficacy.Dark blue and red edges represent positive and negative correlations between nodes, respectively.The node EI raw values are listed on the right.Refer to Table2for the symptoms assigned to the respective items.
Fig. 2 .
Fig. 2. Depression network and node EI values.Dark green nodes and dark blue edges indicate depressive symptoms and positive correlations between nodes, respectively.The node EI raw values are presented on the right.Refer to Table2for the symptoms assigned to the respective items.
Fig. 3 .
Fig. 3. Co-occurrence network of burnout and depression and node BEI values.Dark green color indicates depressive symptoms, and the three dimensions of burnout are distinguished by three different colors, as seen in the burnout network.Dark blue and red color edges point to positive and negative correlations, respectively.Node BEI raw values are listed on the right in descending order.Refer to Table2for the symptoms assigned to the respective items.
Table 1
Descriptive information of demographic variables.
Table 2
Descriptive data of node psychometrics.
|
2024-03-08T14:47:18.866Z
|
2024-03-01T00:00:00.000
|
{
"year": 2024,
"sha1": "ec3856bbd3336a762d438e152090ca16ef1000aa",
"oa_license": "CCBYNC",
"oa_url": "https://doi.org/10.1016/j.psychres.2024.115828",
"oa_status": "HYBRID",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "8dd6f4cc58bbdf30cf83427002760978f46bccf3",
"s2fieldsofstudy": [
"Psychology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
259742953
|
pes2o/s2orc
|
v3-fos-license
|
Procumbent anterior premaxillary teeth in Stenorhynchosaurus munozi (Plesiosauria, Pliosauridae), evidence from new material
A recently prepared fragment of the anterior end of a snout comprising occluded upper and lower jaws of a juvenile individual of Stenorhynchosaurus munozi Páramo-Fonseca et al. , 2016 (Plesiosauria, Pliosauridae) is described herein. The specimen was found in the 1990s at Sutamarchán (Boyacá, Colombia), in Barremian beds of the Arcillolitas abigarradas Member of the Paja Formation. Its description provides hitherto unknown valuable morphological information on the species: dorsal anterior extension of the vomer, five premaxillary teeth (not four as was previously suggested), and anteriorly directed orientation of the first premaxillary alveolus (procum-bent). New observations made on previously studied material corroborated the existence of procumbent anterior premaxillary teeth in the species. The presence of this trait highlights the close relationship of S. munozi with other early-diverging brachauchenines from the Lower Cretaceous of Russia.
Introduction
The knowledge of Lower Cretaceous pliosaurids has recently increased with the description of several specimens from the Berriasian-Hauterivian of western Russia (Fischer et al., 2015;Fischer et al., 2017;Zverkov et al., 2018), the Barremian-Aptian of Colombia (Gómez-Pérez & Noè, 2017;Páramo-Fonseca et al., 2016, 2018, 2019), the Aptian-Albian of Australia (Holland, 2018) and the Albian of Europe (Bardet et al., 2016;Madzia, 2016;Madzia & Machalski, 2017).The Colombian pliosaurids are currently represented by four species, Acostasaurus pavachoquensis Gómez-Pérez and Noè (2017), known through a three-dimensionally preserved skull and some poscranial elements from the lower Barremian (Gómez-Pérez & Noè, 2017); Sachicasaurus vitae Páramo-Fonseca, Benavides-Cabra and Gutierrez (2018) known from a gigantic nearly complete skeleton from the upper Barremian (Páramo-Fonseca et al., 2018); Stenorhynchosaurus munozi Páramo-Fonseca, Gómez-Pérez, Noè and Etayo-Serna (2016) known from a nearly complete skeleton and a skull with some articulated cervical vertebrae from the upper Barremian (Páramo-Fonseca et al., 2016;2019); and "Monquirasaurus" boyacensis (Hampe, 1992) represented by a nearly complete skeleton from the upper Aptian.This last species, previously assigned to Kronosaurus by Hampe (1992), has recently been attributed to a new genus Monquirasaurus Noè and Gómez-Pérez (2022); however, we consider that most of the characters used by these authors to differentiate this species from S. vitae are based on traits that are poorly preserved in the only known specimen of "M." boyacensis and are not obvious enough to warrant a generic distinction.Therefore, we prefer to use the name "Monquirasaurus" (within quotes) until new characters, such as the morphology of the palate, as discussed by Páramo-Fonseca et al. (2018), are available to be compared.
The specimens of Stenorhynchosaurus munozi (the holotype and the specimen MP050310-1) have the anterodorsal end of the snout lost or damaged (Páramo-Fonseca et al., 2016;2019).A well-preserved fragment of the anterior end of a snout, housed in the Museo Paleontológico of the Universidad Nacional de Colombia at Villa de Leyva, was recently prepared.Based on the presence of diagnostic characters the specimen is referred to S. munozi.Herein, we describe this specimen and provide hitherto unknown features of the anterior end of the snout of S. munozi.
Materials and methods
The specimen described herein (MPVL-667) is an anterior end of a snout comprising occluded upper and lower jaws, which was collected by the staff of the Museo Paleontológico of the Universidad Nacional de Colombia at Villa de Leyva in the 1990s.It was collected as a trundled concretion found in the riverbed of La Yuca creek, in the municipality of Sutamarchán, Boyacá, Colombia, a municipality adjacent to Villa de Leyva.The discovery site is located approximately 8 km north-west of Villa the Leyva, (Fig. 1A).We visited the La Yuca creek and examined the concretions transported by the stream, both at the finding site and in the nearest 150 m upstream.Based upon a generic identification of ammonoids collected from the concretions, we propose the stratigraphic origin of the specimen MPVL-667.The ammonoid were identified at generic level by one of the authors (CDB-C).
The specimen was embedded in micritic matrix, so chemical treatment was the selected preparation method.Successive immersions of 5-6 hours in formic acid at 3% concentration followed by a minimum of 12 hours of water rinsing and subsequent drying were performed.Three water baths of 24 hours each were performed at the end of the preparation to ensure the elimination of acid residues.
Given that the new specimen exhibits many morphological similarities with Stenorhynchosaurus munozi, the anatomical comparisons were focused on this species and specifically on specimen MP050310-1, which although damaged, preserves the same region of the snout.To corroborate the affinity of the new specimen with S. munozi, new observations were made directly on specimen MP050310-1.The anatomical description of the dentition follows the terminology proposed by Zverkov et al. (2018).
A 3D photogrammetric model of the specimen is provided as support of the description (see Supplemental Data 1).The model was obtained from raw 18 megapixels images using high-resolution structure from motion techniques.
Each fragment was photographed with a Canon EOS rebel 4000D camera and processed individually using 275 pictures for the upper piece and 225 for the down one.Photos were taken surrounding each piece at three different angles to ensure the maximum possible resolution and coverage.Each model has been scaled using two scale bars located in each image.The correct threedimensional position of both 3D models of the snout fragments was achieved manually, having single pictures of the right location of both pieces.Agisoft Metashape Professional 1.5.1,Meshlab v2022.02,and Blender 3.3.1 programs were used to obtain the 3D models.
A phylogenetic analysis was performed to determine the position of S. munozi within the Plesiosauria, based on our enhanced anatomical understanding of the species.The analysis was based on the data set of Páramo-Fonseca et al. (2019) which is a modified version of the data set of Fischer et al. (2017) and that of Benson and Druckenmiller (2014).The dataset was compiled using the Mesquite software 3.51 (Maddison and Maddison, 2018).The resulting data matrix has 102 OTUs and 271 characters (Supplemental Data 2).The Cladistic analysis was carried out using TNT (Goloboff and Catalano, 2016).All characters were treated as unordered and were equally weighted.An initial exploration for the shortest-length tree islands was performed using the new technology search options (Ratchet + Drift + Tree Fusing) with 200 ratchet iterations and 99,999 trees limit.Five replicates, one hit on the shortest tree, and seven trees held per replicate were obtained.The resulting trees were then used as the starting point for a tree bisection reconnection (TBR) branch swapping algorithm.A strict consensus was obtained, and the unstable OTUs were identified by applying the iterPCR algorithm (Pol and Escapa, 2009).The unstable OTUs were pruned from the consensus cladogram.The consistency (CI) and retention (RI) indices (Farris, 1989) were calculated, and Bremer support values (Bremer, 1994) were computed for the nodes in the consensus tree using TNT tools.
Geologic and biostratigraphic settings
The specimen was found in the riverbed of La Yuca creek, in a trundled concretion in which no ammonoids, except for a minute unidentifiable shell, were preserved.We collected a matrix sample and analyzed it for calcareous nannofossils and palynomorphs, but the analysis showed no microfossil content.Although no outcrops were found in the La Yuca creek, the stream cuts the La Yuca hill where the almost complete stratigraphic sequence of the Arcillolitas abigarradas Member of the Paja Formation outcrops (Figs.1A-B).The topography and the geology of the finding region allowed us to establish with certainty that the specimen MPVL-667 comes from the Arcillolitas abigarradas Member of the Paja Formation.
Abundant trundled concretions bearing ammonoids were found in the riverbed of La Yuca creek.In some of the concretions we identified lower Barremian ammonoid genera, such as Nicklesia Hyaat, 1903 and Pulchellia Uhlig, 1883 (Patarroyo, 2000;2020) (Figs. 1C-D), and in other concretions we identified the genera Gerhardtia sensu Hyatt (1903) and Heinzia sensu Bürgl (1956) (Figs.1E-F) which are indicators of the upper Barremian in Colombia (Patarroyo, 2000;2020).We did not find any Aptian ammonoids on the creek, which suggests that the specimen came from Barremian rocks.Furthermore, in the nearest outcrops in the La Yuca hill we found the same ammonoid genera.In this context, we strongly assert that the specimen MPVL-667 originated within Barremian Beds of the Arcillolitas abigarradas Member that can be correlated with the segments A, B, and C of Etayo-Serna (1968) (see Fig. 1B).A more accurate stratigraphic origin of the specimen within the Barremian sequence is not possible.
Geographic and stratigraphic occurrence: Alto Ricaurte region (Villa de Leyva, Sáchica and Sutamarchán), Boyacá, Colombia; Barremian beds of the Arcillolitas abigarradas Member of the Paja Formation (Etayo-Serna, 1968;Páramo-Fonseca et al., 2016;2019) Emended Diagnosis: The diagnosis of Stenorhynchosaurus munozi follows Páramo-Fonseca et al. (2019) except for the following features, which are observed in the new specimen.Premaxilla with five (not four) functional alveoli, the anterior one procumbent and smaller than other premaxillary teeth.The autapomorphy concerning the vomer is specified as follows: vomer extending dorsally above the palatal processes of the premaxillae, but in ventral view ending posterior to the palatal premaxilla-maxilla suture.
Preservational features
The specimen is three-dimensionally preserved and exhibits accurate detail of bone textures.It consists of the anterior end of a snout with the occluded mandible, extending to the level of the first maxillary alveolus (Fig. 2).The upper jaw was preserved slightly turned to its left side (Fig. 2B).The external surfaces of the first alveolus in the left premaxilla as well as the anterior tip and the right lateral side of the mandible are eroded.Most of the functional teeth are preserved in place on the right side of the mandible, whereas the remaining alveoli in the upper and lower jaws are either empty or they lodge replacement teeth.
Anatomical features
Premaxilla and maxilla.The anteriormost portion of the premaxillae, forming the anterior end of the snout, and a small portion of the anterior end of the maxillae are present in the specimen (Figs.3A-D).The preserved portion of the snout is narrow with near parallel borders in dorsal view (Fig. 3B).Only a very slight narrowing of the snout at the level of the fifth alveolus and a faint ,C).Ventrally, the premaxilla-maxilla suture crosses the alveolar surface on a diastema of 19 mm in length that separates the fifth and sixth functional alveoli (Fig. 3C).
In ventral view, the palatal process of the premaxilla extends posterior to the premaxilla-maxilla suture (Fig. 3C).This condition indicates that the anterior end of the vomer exposed in palatal view is located posterior to the palatal premaxilla-maxilla suture.This trait is an autapomorphy of Stenorhynchosaurus munozi (Páramo-Fonseca et al., 2016).The palatal process of the premaxilla diverges dorsally forming a medial longitudinal lamella that contacts the vomer (Fig. 3D).Medial to the alveolar row, a narrow longitudinal canal runs, joining the concavities into which the replacement teeth erupt (Fig. 3C).This canal is interrupted in the diastema.At the level of the fourth alveolus an oblique posteromedial groove cross from the longitudinal canal to the midline where the two premaxillae meet (Fig. 3C).
Each premaxilla has five alveoli.The first alveolus is the smallest and the fourth alveolus is the largest (Fig. 3A, C).The first alveolus is practically horizontal and opens anteriorly on a vertical plane above the second alveolus, indicating a procumbent condition (Fig. 3A).The procumbent alveolus from both premaxillae is very close to each other; there is a thin bone wall of 0.26 mm separating them.The procumbent alveolus is vertically oval in outline, 19 mm in dorsoventral diameter and 13 mm in mesiodistal diameter (Fig. 3A).The subsequent premaxillary alveoli are circular in outline and anterolaterally oriented.Their mesiodistal diameter measure 22 mm (in the second alveolus), 23 mm (in the third), 25 mm (in the fourth), and 23 mm (in the fifth).The interalveolar spaces range from 6 to 8 mm.In the maxilla, only the right first maxillary alveolus is complete (Fig. 3C), it is circular in outline and its mesiodistal diameter is 22 mm.
Vomer.The vomers are fused forming an elongated bar that anteriorly extends dorsal to the palatal processes of the premaxillae (Fig. 3D).This dorsal extension of the vomers was not previously known in S. munozi; therefore, it is specified in the diagnosis of the taxon.The bar is trifurcated ventrally, forming three ventral lamellae that intercalate with the medial lamellae from the premaxillae, in such a way that, in cross-section, the joint between the vomers and the premaxillae is M-shaped (Fig. 3D).The dorsal surface of the fused vomers is concave and is textured by numerous longitudinal ridges.
Mandible.The preserved mandible includes the anterior portion of the dentaries and the anteriormost ends of the splenials, forming the anterior part of the mandibular symphysis (Figs.3E-H).Most of the right external surface of the preserved mandible is eroded, but the left lateral surface is well preserved and demarcates the alveolar margins.As the rostrum, the mandible does not exhibit noticeable lateral expansion nor a marked increase in the size of the functional alveoli.The symphysis is not tightly closed; both mandibular rami run nearly parallel to each other, slightly further apart in the back (5 mm) than in the front (2 mm) (Figs.3E, G-H).As in the upper jaw, a narrow longitudinal canal running medial to the alveolar row joins the concavities into which the replacement teeth erupt (Fig. 3F).Coinciding with the upper jaw, at the level of the fourth alveolus an oblique posteromedial groove cross from the longitudinal canal to the midline where both dentaries meet (Fig. 3F).
Each mandibular ramus preserves six complete and a seventh incomplete functional alveoli.The first functional alveolus is the smallest and the fifth is the largest (Fig. 3F).The first alveolus is anteriorly directed and nearly horizontal but opens in an anterodorsal plane.Differing from the first alveolus in the premaxilla, which opens directly above the second one, the anterior alveolus in the mandible does not open under the second alveolus, but anteroventral to it (Fig. 3E).The described condition suggests the first mandibular tooth was not procumbent.Except for the first alveolus, the mandibular functional alveoli are inclined slightly anterolaterally.The fifth and sixth mandibular teeth positions agree with the diastema of the upper jaw, without having an intercalated tooth.The measurements of the mesiodistal diameter of the mandibular alveoli are: 17 mm (anterior alveolus), 23 mm (second alveolus), 24 mm (third alveolus), 22 mm (fourth alveolus), 26 mm (fifth alveolus) and 22 mm (sixth alveolus).The space between successive alveoli in the mandible is slightly shorter than in the premaxilla; it ranges from 5 to 7 mm.
Teeth.Only replacement teeth are present in the premaxillae and maxillae (Fig. 3C).On the mandible, in addition to some replacement teeth, there are five functional teeth, four in the right mandibular ramus and one in the left (Fig. 3F).Replacement teeth emerge in small cavities placed within a canal that extends medially to the alveolar row.These cavities, named the replacement alveoli by Edmund (1960), are positioned distal-lingual to the marginal (functional) alveoli, and become elongated when the replacement tooth migrates to the marginal alveolus (Edmund, 1960;Sassoon et al., 2015).The replacement tooth of the first premaxillary position is located ventrally to the functional alveolus and the first replacement tooth in the mandible is placed dorsally to the marginal alveolus (Figs.3C, F).In both premaxillae, the third replacement tooth is the largest; their crown is practically emplaced into the alveolus, but their root is still short.The root in the second and third functional teeth of the right mandibular ramus is visible due to erosion (Fig. 2).They are 59 mm and 55 mm in heigh respectively and protrude 15 to 17 mm from the alveolar rim.The root height is about twice that of the crown.The root is slightly bulbous, circular in crosssection, and its surface is smooth.
The crowns of all functional teeth is poorly preserved and labially eroded (Fig. 2).However, in the right sixth functional tooth the lingual surface is visible and in the largest replacement teeth the crowns are well preserved.The crown of the largest replacement teeth is 31 mm in height.The crown is nearly conical, with a flat labio-mesial surface delimitated by two sharps smooth carinae, one mesial and one distal-labial, running the entire height of the crown (Figs.4A-C).This morphology gives a sub-trihedral cross-section of the crown (Fig. 4H), as defined by Benson et al. (2013) for P. kevani.An additional distal apicobasal ridge appears toward the apex of the crown forming a trihedral tip of the tooth (Fig. 4D), as in the holotype of Stenorhynchosaurus munozi (Páramo-Fonseca et al., 2016).The flat surface of the crown only bears a few fine ridges at its base, whereas on the remaining enameled surface there are coarse, abundant, and equidistant longitudinal ridges reaching the lower half of the crown (Fig. 4E-G).For every one or two ridges, a shorter inserted ridge is intercalating.These crown features have been described for the specimen MP050310-1 (Páramo-Fonseca et al., 2019).
Phylogenetic analysis
Based on the anatomical input provided by the described specimen we modify the scoring of character 131 of Stenorhynchosaurus munozi from state 0 (four premaxillary teeth) to 1 (five premaxillary teeth).Furthermore, we added one character to the matrix: 271-Direction towards which the anteriormost premaxillary alveolus faces.0. The alveolus faces ventrally or weakly anteroventrally; 1.The alveolus faces directly anteriorly (procumbent).The scoring of the new character [271] can be found in Supplemental Data 3.
The expected early diverging branch grouping Stenorhynchosaurus munozi, Makhaira rossica and Luskhan itilensis, does not appear in the cladogram, probably due to the unstable position of M. rossica.The node of S. munozi and L. itilensis does not appear as early-diverging as expected by their shared characteristics with the Jurassic thalassophoneans, instead it appears in a distant position to the ancestral node Brachaucheninae.This might be due to the unordered treatment of the characters related to the tooth morphology.
Discussion
One of the two autapomorphies defining Stenorhynchosaurus munozi concerns the anterior region of the snout and is present in the new specimen: The anterior external exposure of the vomer is located posterior to the palatal premaxilla-maxilla suture in palatal view (Páramo-Fonseca et al., 2016;2019).Moreover, the morphological features of the new specimen, including the size and shape of the snout, the size and orientation of the alveoli, and the shape and ornamentation of the teeth are consistent with the morphology previously described for Stenorhynchosaurus munozi (Páramo-Fonseca et al., 2016;2019).The dimensions of the studied specimen agree with the size of the holotype of S. munozi, which was considered a juvenile individual (Páramo-Fonseca et al., 2016).In both specimens, the symphysis is not fused indicating that they represent juvenile individuals.The new specimen is thus reliably referred to as a juvenile individual of S. munozi.
The ventral exposure of the anterior end of the vomer is not observable in the only adult specimen of S. munozi known so far (Páramo-Fonseca et al., 2019).The anterior dorsal extension of the vomer observed in the new specimen raises the possibility that the shorter ventral exposition of the vomer in young specimens is the result of ontogenetic allometry.However, this doubt cannot be clarified until the ventral exposure of the vomer in the palate of an adult individual is known.
Stenorhynchosaurus munozi was erected for a nearly complete skeleton (VL17052004-1, the holotype) whose skull lacks the anterior tip of the snout (Páramo-Fonseca et al., 2016).A second specimen (MP050310-1), a skull with some articulated cervical vertebrae, was later described by Páramo-Fonseca et al. (2019).The dorsal region of the anterior end of the snout in the second specimen is damaged (Páramo-Fonseca et al., 2019: Figure 3C; Figs 5A-B).Therefore, the anterior dorsal end of the snout was unknown in S. munozi.In the published descriptions of these specimens, the authors proposed the number of premaxillary teeth of S. munozi as four based on the preserved material (Páramo-Fonseca et al, 2016;Páramo-Fonseca et al., 2019).The specimen herein described has five premaxillary teeth, the first being a dorsal procumbent tooth smaller than the other premaxillary teeth.This allows us to infer that S. munozi had five premaxillary teeth and not four as previously proposed.
New and detailed observations made on the dorsal end of the snout of the specimen MP050310-1, allowed us to identify two circular concentric structures on the broken surface that this specimen has above the alveolar margin (Fig. 5B-C).The morphology of these structures agrees with traces of two broken anterior procumbent teeth.Nevertheless, these structures seem to be in a somewhat higher position than that occupied by the procumbent alveoli in the new specimen.This discrepancy does not imply a taxonomic difference, since: 1-the broken surface where the concentric traces are registered in MP050310-1 is placed at a considerable distance from the frontal alveolar margin, so the traces could be showing the posterior portion of alveoli of the two slightly inclined procumbent teeth; and 2-the MP050310-1 represents an adult individual and the new specimen comes from a juvenile individual, so it could be expected an ontogenetic difference between the two specimens, with a more robust snout and procumbent teeth more separated from the alveolar margin in the adult.
Based on the above discussion, we can affirm that Stenorhynchosaurus munozi had five premaxillary teeth, the anterior one procumbent, nearly horizontal, and smaller than the other premaxillary teeth.The presence of procumbent premaxillary teeth is rare in thalassophonean pliosaurids; it has (Fischer et al., 2015;Fischer et al., 2017).In addition, L. itilensis, M. rossica, and S. munozi share the elongated rostrum without lateral expansions, the isodont dentition, and the subtrihedral crown with trihedral apex (Fischer et al., 2015;Fischer et al., 2017;Páramo-Fonseca et al., 2019), a feature that differentiates these taxa from the other brachauchenine pliosaurids, which have a subcircular tooth cross-section (Zverkov et al., 2018).
The differences between S. munozi and the Russian Hauterivian taxa have already been discussed by Páramo-Fonseca et al. ( 2019).Here we give only a few remarks on the number of premaxillary teeth (five in the S. munozi), which differs from that of the Russian taxa (six in M. rossica and seven or probably six in L. itilensis) (Fischer et al., 2015;Fischer et al., 2017;Zverkov, personal communication) and from that of all other brachauchenines (four functional teeth) (Schumacher et al., 2013;Albright et al., 2007;White, 1935;Gómez-Pérez and Noè, 2017;Páramo-Fonseca et al., 2018) Nevertheless, all these taxa differ from S. munozi by having a constricted rostrum with large teeth near the premaxilla-maxilla suture and short symphysis (Andrews, 1909;1913;Halstead, 1971;Benson et al., 2013;Fischer et al, 2015, appendix data set;Ketchum & Benson, 2022).The combination of features in the premaxillary dentition of S. munozi is therefore unique among thalassophonean pliosaurids.
The new traits are therefore added to the diagnosis of the species.The presence of anterior procumbent teeth in the premaxilla of the Barremian taxon S. munozi from Colombia has significant implications in the temporal distribution of this trait in the history of long-snouted brachauchenines.It is noteworthy that within the thalassophonean pliosaurids the procumbent teeth in front of the upper jaw only occur in the long and narrow snouted taxa of the Early Cretaceous.These taxa have morphological affinities with the Cretaceous brachauchenine pliosaurids but retain traits of the dental crown morphology found in the Jurassic thalassophonean pliosaurids.These long and narrow snouted taxa with procumbent teeth formed, early in the history of the brachauchenines, a derived evolutionary lineage with a particular ecological adaptation that successfully expanded across the Tethys Sea from Eurasia to northern Gondwana during the Early Cretaceous.
The apparent temporal discontinuity of the ecological adaptation of a narrow snout with procumbent front teeth on premaxilla among brachauchenines could be due to a collection bias, but also to ecological changes.It would not be unreasonable to suggest that the polycotylid plesiosaurs, having elongated snouts, relatively isodont dentition, and frequent teeth with carinae (Fischer et al., 2018), which begin their history in the Aptian-Albian (Druckenmiller, 2002;Kear, 2003) and become diverse during the late Cretaceous (Druckenmiller and Russel, 2009;Fischer et al., 2018;Sachs et al., 2020), could have occupied the ecological niche of the Early Cretaceous long-snouted brachauchenines.
Conclusions
We identify the specimen MPVL-667 as a juvenile individual of Stenorhynchosaurus munozi by presenting juvenile traits in a morphology consistent with that described for this species and showing the anteriorly exposed end of the vomer in palatal view located posterior to the palatal premaxilla-maxilla suture, which is one of the autapomorphies that define S. munozi.The study of the new specimen reveals information about traits that were previously unknown or were interpreted differently for the species, such as the dorsal anterior extension of the vomer, the presence of five premaxillary teeth (previously interpreted as four) and the existence of a procumbent anteriormost premaxillary tooth.This last feature is corroborated by new observations in the specimen MP050310-1.The procumbent anterior premaxillary teeth and the subtrihedral crown with trihedral apex is shared by S. munozi, Luskan itilensis, and Makhaira rossica and differentiate these taxa from the other brachauchenines.These shared traits show a close relationship between the Hauterivian brachauchenines from Russia and the Barremian brachauchenines from Colombia.Moreover, our cladistic analysis supports the phylogenetic proximity of S. munozi and L. itilensis by 11 synapomorphies, including the new character of the procumbent anteriormost premaxillary teeth.We suggest that the niche occupied by the Early Cretaceous long-snouted brachauchenines (S. munozi, L. itilensis, and M. rossica) could have been later occupied by the polycotylid plesiosaurs, which also have elongated snouts, relatively isodont dentitions and frequent teeth with carinae.
Figure 5 .
Figure 5. Stenorhynchosaurus munozi, specimen MP050310-1.A-B, Anterior tip of the snout in A, anterolateral and B, anterior views.C, detail of the anterior end of the snout showing the two concentric structures (broken teeth) found in the broken anterior surface.Scale bar = 30 mm.
Figure 6 .
Figure 6.A, B, consensus trees of 99,999 trees of 1,620 steps based on data set of Plesiosauria with 102 OTUs and 271 characters (see Materials and Methods).A, strict consensus tree, with Bremer support indicated in some nodes.B, reduced consensus found after pruning the unstable OTUs.
|
2023-07-12T08:07:52.925Z
|
2023-05-23T00:00:00.000
|
{
"year": 2023,
"sha1": "4db79802506a909fb3102c0721d6d4790a236384",
"oa_license": "CCBY",
"oa_url": "https://revistas.unal.edu.co/index.php/esrj/article/download/105689/88471",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "49202a61122c9d693b40029c784c168b42018879",
"s2fieldsofstudy": [
"Biology",
"Geology"
],
"extfieldsofstudy": []
}
|
121405609
|
pes2o/s2orc
|
v3-fos-license
|
Self-organized model of cascade spreading
We study simultaneous price drops of real stocks and show that for high drop thresholds they follow a power-law distribution. To reproduce these collective downturns, we propose a minimal self-organized model of cascade spreading based on a probabilistic response of the system elements to stress conditions. This model is solvable using the theory of branching processes and the mean-field approximation. For a wide range of parameters, the system is in a critical state and displays a power-law cascade-size distribution similar to the empirically observed one. We further generalize the model to reproduce volatility clustering and other observed properties of real stocks.
Introduction
Cascade spreading is an important emergent property of various complex systems. Real life examples of cascades are numerous and range from infrastructure failures and epidemics to traffic jams and cultural fads [1,2]. Theoretical models of cascades usually assume that agents can be in one of two states (healthy or failed) and an agent's failure puts some stress on its neighbors which may consequently fail too. See [3] for a recent survey of this field offering a novel unifying view.
In this paper we focus on cascades in economic systems which can be identified with stock prices suddenly dropping in a major market crash [4] or with companies going bankrupt simultaneously and leading to global recession [5]. Theoretical models of such cascades are based on shortage and bankruptcy propagation in production networks [6], default propagation in credit networks [7,8], interaction of firms through one monopolistic bank [9] or in a complex credit network economy [10], and herding behavior of traders [11,12]. While these models help us to understand cascade processes in economic systems, they are mostly too involved to allow for analytical solutionstheir study hence relies on numeric simulations and agentbased modeling [13].
A simpler point of view on cascade phenomena is offered by the concept of self-organized criticality (SOC) which has had a deep impact on the science of complexity. First introduced more than twenty years ago to explain the ubiquitous 1/f noise [14], it caused a blossoming of toy models, computer simulations, and real life experiments [15]. The analytical techniques employed include scaling arguments [16], mean-field theories [17], branching processes [18], renormalization methods [19,20], and rigorous algebraical techniques [21]. SOC is a mechanism which explains the emergence of complex behavior in many diverse real world systems [22,23]. The generic behavior of SOC models is: (a) they evolve so that they always stay close to the critical point, (b) long periods of robustness and moderate activity are interrupted by sudden breakdowns. This qualitatively resembles "stock markets which expand and grow on relatively long time scales but contract in stock-market crashes on relatively short time scales" [15] and "stock crashes caused by the slow buildup of long-range correlation leading to a global cooperative behavior of the market eventually ending into a collapse in a short time interval" [4]. This similarity provides the main motivation for the present study.
We begin our work with an empirical investigation of simultaneous price drops of real stocks and show that the size distribution of observed events is broad (for high drop thresholds it follows a power-law distribution). This observation suggests that simultaneous stock downturns are a collective phenomenon. We propose a simple dynamical model which for a wide range of parameters selforganizes into a critical state. Unlike most SOC models, our model assumes a probabilistic response mechanism where a node has only a certain probability of reacting to the current stress conditions. The basic idea behind modeling simultaneous stock downturns with cascades is that decline of a single stock may provoke investors' reactions which consequently may cause other stocks to decline and a "cascade" to spread. The key premise is that while failed nodes become significantly more resistant in the next time step, healthy nodes become slightly less resistant. This close parallel with the slow growth/fast decay picture described above is further supported by our analysis of empirical data which shows that majority of stocks behave in this way. While there are certainly many other effects contributing to the dynamics of market crashes (external shocks, for example), we show that failure propagation alone can reproduce some of the observed patterns.
The minimal model proposed here has the advantage of being simple, not relying on fine-tuning of parameters, analytically solvable in some cases, and easily generalizable to more complicated settings. We analyze it using the formalism of branching processes, the mean-field approximation and, for complex topologies of nodes' interactions, using numerical simulations. Obtained cascade-size distributions exhibit a close similarity to our empirical observations. Introduction of memory within the model allows us to reproduce other empirically observed features, such as volatility clustering, though at the cost of analytical tractability. We conclude our study with a discussion of further model's generalizations and possible areas of application.
Empirical data
Here we investigate co-occurring price movements of real stocks. Adopting the vocabulary of cascade models, we say that a stock fails when the relative loss of its price over a given time interval ∆t exceeds a certain threshold H. Denoting the price of stock i at time t as p i (t), its failure occurs when [p i (t) − p i (t + ∆t)]/p i (t) > H. The number of stocks failing at time t, n F (t), is a direct analog of the cascade size in a model of cascade spreading. As the input data we use daily closing prices (hence ∆t = 1 day) of 500 stocks from the standard U.S. index S&P 500 (this data is freely available at, for example, finance.yahoo.com). To achieve a fixed system size, we consider only those 332 companies which are in the stock market since the beginning of 1992 and use their prices during the 18-years long period ending in May 2010 for our analysis.
The empirical distribution of failure sizes is shown in Fig. 1 for H = 0% and H = 10%. We see that for the large value of H (which is in line with the notion of stock failures), the observed size distribution has a power-law shape. Using the methodology described in [24], we obtained the power-law exponent 2.19 ± 0.05 with the lower bound for the power-law behavior n min = 3. The corresponding p-value (obtained using the standard Kolmogorov-Smirnov statistic) is 0.92 which confirms that the data is consistent with the hypothesis of a power-law distribution. Similar results are obtained also for other threshold values so long as H 8%. When H 8%, the resulting size distributions are broad but probably not power-law. Finally, when H = 0% (i.e., any price drop is interpreted as a failure), the size distribution is roughly symmetric around the value corresponding to one half of the system size (see Fig. 1). In the following analysis of empirical data we use the threshold H = 10%.
The power-law shape itself suggests that the observed simultaneous stock downturns are rather a collective phenomenon than independent events. This hypothesis is further supported by the average correlation of simultaneously failing stocks, 0.35 (again including only events with at least three simultaneously failing stocks), which is significantly higher than the overall average stock correla- tion, 0.25. Another sign of a strong connection among simultaneously failing stocks comes from their division to ten different industrial sectors according to the GICS classification. The effective number of sectors participating in a cascade is defined as where r i is the relative share of sector i in the cascade and 10 i=1 r i = 1. By averaging this quantity over all cascades of a given size S, we obtain e(S). This number can be compared with the effective number of sectors corresponding to selecting failed stocks at random, e ′ (S). The analysis of stock prices shows that for any S > 3, e(S) is significantly smaller than e ′ (S) which implies that simultaneous stock failures preferentially affect strongly connected stocks in one sector or in a small number of sectors. Now we turn our attention to time correlations of failures. The autocorrelation of the number of failing stocks with the time lag one day, C(n F (t), n F (t + 1)) ≈ 0.15, is comparable with the autocorrelation of absolute returns, C(|r(t)|, |r(t + 1)|) ≈ 0.25 (the latter result agrees with previous studies [25,26]). The positive autocorrelation values are signs of volatility clustering which is commonly observed in financial data [27]. (Loosely speaking, volatility clustering means that large changes tend to be followed by large changes and small changes tend to be followed by small changes, as first noted by Mandelbrot [28]. ) We further estimate conditional failure probabilities for individual stocks. For example, P (F |N ) denotes failure probability of a stock given that this stock didn't fail in the previous time step (other three quantities, P (N |F ), P (F |F ), and P (N |N ), follow the same logic). When the results are averaged over all stocks, we obtain P (F |F ) = 0.039 which is much higher than the overall failure probability P (F ) = 0.003-this is another sign of volatility clustering in our data. On the level of individual stocks, however, 62% of all stocks with at least three failures strongly satisfy the inequality P (N |F ) > P (N ) which is equivalent to P (F |F ) < P (F ) (because P (F |F ) + P (N |F ) = 1). (By strong satisfying we mean that the difference of the two probabilities is greater than the sum of their uncertainties.) We see that despite volatility clustering in the data, most stocks are more "resistant" to failures after they have just undergone one. For the remaining stocks, probabilities P (N |F ) and P (N ) either differ less than the sum of values' uncertainties (for 14% of stocks) or even strongly satisfy the opposite inequality P (N |F ) < P (N ), with corresponding values of P (F |F ) often as high as 0.30 (24% of stocks).
To summarize, after a failure (a major price drop), most stocks become more resistant to another failurethis observation will serve as a basis for the mathematical model presented in the following section. At the same time, there is a fraction of stocks which are prone to consecutive failures-this particular feature will be discussed in detail in Section 4.
Basic model and its mean-field solution
In this section we present a basic model which is amenable to analytical treatment and qualitatively reproduces some of the features observed in empirical data. In its original formulation, this model is particularly suitable for stocks that, as discussed in the previous section, after a failure become more robust. A generalization of the model aiming at reproducing other observed features (volatility clustering, for example) is presented in Section 4.
Consider a system of N nodes where node i (i = 1, . . . , N ) has only two possible states: failed (i ∈ F ) and healthy (i ∈ F ). With each node i we further associate fragility f i ∈ [0, 1] which measures how this node reacts to failures of its neighbors (the higher the fragility, the more likely is the node to follow a neighbor's failure). The dynamics of the model is governed by the following simple rules. (i) In each time step, the first failed node ("trigger") is chosen at random and may induce failures of other nodes. (ii) If a neighbor of node i fails, node i follows it with probability f i and resists with probability 1 − f i . (If several neighbors of node i fail simultaneously, in order to stay healthy, node i has to resist each individual failure.) The cascade of failures propagates until all remaining nodes resist the damage. (iii) At the end of the time step, fragilities of all nodes are updated according to where 0 < β ≪ 1 and λ ∈ (0, 1) are parameters of the model (in effect, failed nodes become less fragile and healthy nodes become slightly more fragile in the next time step). All values f i (t + 1) > 1 are truncated to 1 (this may occur when β is large). After this update is finished, all nodes are again marked as healthy, the current time step ends and a new one begins with point (i). Note that unlike some other models of cascade spreading, failed nodes are not removed from the system in our case. If a long enough equilibration period is applied before measuring the system behavior, the initial fragility values f i (0) are of little importance (see Section 3.5 for a detailed discussion). Unless stated otherwise, we set them randomly in the range (0, 1) in our simulations. According to the rules above, when n neighbors of node i fail, node i resists with the probability (1 − f i ) n and fails with the complementary probability This response to failures is "path-independent" in some sense: the probability that a node resists n failures of its neighbors, (1 − f i ) n , is the same as the probability of resisting two consequent waves of failures of x and n − x neighboring nodes, We simplify the system by assuming that interactions of all nodes are equally strong (the general case will be studied in Section 3.4). This renders the notion of "node's neighbors" superfluous because every failure affects all remaining healthy nodes in the system. Now assume that after the initial failed node is chosen, n 1 nodes respond to this failure and fail too. Each of the remaining N −n 0 −n 1 nodes (here n 0 = 1 is the initial number of failed nodes) then has some n 1 -dependent failure probability which results in n 2 new failures, and so on, until in iteration m, n m = 0 is achieved. The cascade size is then defined as the total number of failures, S = n 0 + · · · + n m , and node fragilities are consequently updated according to Eq. (2). Since in one turn nodes can only fail once, cascade sizes are limited by the system size and S ≤ N .
The dynamics of the system, based on failure propagation and fragility updating, is fully contained in the three above-described rules. In the following paragraphs we shall study when these rules drive the system to a critical state and what is the distribution of cascade sizes P (S).
Failure probability
Let P F be the average failure probability of a given node in one time step (or, equivalently, the average fraction of failed nodes in one time step). Assuming that t eq is some sufficiently long equilibration time (we use t eq = 10 4 for all our simulations), later fragility values averaged over realizations, f i , do not evolve anymore. All nodes interact equally strongly, hence f i is independent of i and it can be replaced with f . Since in a large number of time steps T each node undergoes P F T failures and Using the equilibrium condition f (t eq + T ) = f (t eq ) , we can solve this equation with respect to P F to get When β ≪ 1, this can be approximated with P F (β, λ) ≈ −β/ ln λ (Fig. 2 compares these results with numerical simulations). A node may fail because it is selected as the first failed node (with probability 1/N ) or due to failure propagation (with probability P P ); P F thus can be written as P F = 1/N +P P . Since the value of P F depends solely on β and λ, P P = P F − 1/N may be negative for a small system which is, of course, impossible in practice. This situation occurs when for given λ, N , the value of β is smaller than a certain threshold β 0 and hence it does not suffice to compensate for the fragility decay due to λ. Eq. (4) then has only the trivial solution f = 0 and hence P F (β, λ) = 1/N (failures do not spread). When β is small, the approximate form of P F can be used to solve this equation with respect to β and we get which agrees with numerical simulations (see the vertical line in Fig. 2). Note that if the number of initial failed nodes is assumed to grow with the system size as wN (w ≪ 1), we get β 0 ≈ −w ln λ which is independent of N . When model parameters are set to extreme values (for example, N = 10 3 , β = 10 3 , λ = 10 −3 ), the system exhibits unusual modes of behavior where active turns (with nearly all nodes failed) alternate with calm turns (with nearly all nodes healthy). While Eq. (5) holds also in such conditions, our further analysis focuses on β ≪ 1 which renders more realistic behavior.
Average fragility
When nf i ≪ 1, P F given by Eq. (3) can be approximated as P F (f i , n) ≈ nf i which can be interpreted as independence of stress inflicted by n individual failed nodes. This further means that each failed node has its failing descendants independently of other failed nodes and hence one can use the theory of branching processes [29] to describe the cascade spreading. Note that by use of this theory we implicitly assume that the system size is infinite. For a discussion of the finite-size effects on the size of an epidemic outbreak see [30].
As already mentioned, when interactions of all nodes are equal, f i is independent of i. If we further neglect fluctuations of f i , then all nodes have identical fragility f . This is a mean-field-like approximation which replaces the exact cascade spreading with cascade spreading in a homogeneous averaged medium. Since the number of direct descendants now follows a simple binomial distribution with mean N f , we can use elementary results of branching process theory to express the average cascade size (the total progeny) as S = 1/(1 − N f ). Further, using S = N P F (β, λ) we obtain the average fragility Since β > 0 and λ < 1, f is always less than 1/N . Comparison with numerical simulations (not shown) confirms that Eq. (7) is valid only for β ≪ 1.
Cascade size distribution
The theory of branching processes is well studied [31] and can be easily applied to our model. According to a theorem from [32], if the generating function for the number of direct descendants d is π(x), the total progeny of the resulting branching process Y has the distribution a is defined using and n 0 is the number of ancestors (in our case, the number of initial failed nodes). Since d obeys a binomial distribution, its generating function is π(x) = (1 − f + f x) N and we get where we used n 0 = 1 and f is given by Eq. (7). Note that the resulting probability is positive for S > N which contradicts the model assumptions (each node fails at most once in a given turn). This is a direct consequence of using the theory of branching processes which assumes that the system size is infinite. This problem is of little importance for small values of β when the obtained values of P (S) are negligible for S > N . When 1 ≪ S ≪ N , Eq. (10) can be approximated with According to Eq. (7), lim N →∞ N f = 1 for any given β, λ and hence in the limit of large system size is P (S|β, λ) ∼ S −3/2 which corresponds to the classical critical branching process. For a finite system, the smaller the value of β, the larger the value of 1 − N f . Consequently, the power-law scaling holds only for S ≪ βN (this agrees with Fig. 3 where for β = 10 −3 , the power-law behavior disappears at S ≈ 10). On the other hand, the range of β and λ for which the system self-organizes to a critical state is wide and we can say that this is an SOC system. A comparison of the obtained analytical results with numerical simulations is shown in Fig. 3. The agreement is good for small values of β (β 0.01) and the initial slope of the distributions (before the finite-size effects become apparent) is close to −3/2. Results obtained with β = 0.001 confirm that when β is small enough, P (S) decays faster than as a power law. When β is large, true P (S) deviates from the analytical prediction and exhibits a secondary maximum at a large size value-this effect is well visible in Fig. 3 for β = 0.1. This maximum, formally simply a super-critical phase of the model, resembles socalled meaningful outliers discussed in [33]. To estimate the value of β at which the secondary maximum appears and Eq. (10) ceases to hold, we take the average number of failures computed both from Eq. (10) and from Eq. (5). By comparing the two results we obtain When β is small, both sides of this equation depend on β and the equality can hold. However, Eq. (11) shows that when β is sufficiently large, the size distribution is approximately power-law and it is independent of β. As we increase β further, the power-law distribution does not suffice to provide enough failures and for Eq. (12) to hold, an additional contribution must appear on the rights side. The value β 1 when this happens can be found by substituting P (S) ∼ S −3/2 on the right side and approximating the summation with integration. When N is large, we obtain which complements the previously found threshold β 0 . For N = 10 4 and λ = 0.1, we obtain β 1 ≈ 0.02 which agrees with our empirical observation (β 0.01 for Eq. (10) to hold) above.
Finally, by comparing the empirical observations presented in Fig. 1 with the obtained analytical results, we can conclude that the presented model exhibits qualitative agreement with the studied real system.
Generalizations
To test how robust are the obtained results, we consider simple generalizations of the proposed model. First of all, when the multiplicative fragility update rule Eq. (2) is replaced by an additive one, the behavior of the system does not change considerably. The second generalization relates to the assumed even influence of a node's failure on all the remaining nodes. Denoting the strength of failure propagation from node i to node j as C i,j , the probability that node j fails as a result of i's failure can be generalized to C i,j f j . The probability that node j fails as a result of a group F of failed nodes (given by Eq. (3) before) generalizes to the form Matrix C encodes the structure of the network of node interactions.
When the elements C i,j are drawn independently from a given distribution and the system size is large, the meanfield approximation is again appropriate to describe the system behavior and the power-law size distribution with exponent 3/2 results. Similarly when C contains a block structure with inter-block elements drawn from a different distribution than intra-block elements (this mimics the sector structure of the stock correlation matrix [26,34]), the original power-law size distribution remains largely unchanged (unless either the block division of C or one of the two probabilistic distributions are such that they do not allow to use the mean-field approximation). Analogous behavior results from the "random neighbor approximation" in which node's neighbors are chosen anew repeatedly (see [35] for this kind of analysis of a different model).
When all elements C i,j are either zero or one, matrix C can be represented by a network and a complex topology of node interactions can be introduced by network models [36]. We studied two different types of networks: the Erdős-Rényi network where C i,j = 1 with probability p and C i,j = 0 otherwise and the growing Barabási-Albert network where each new node is attached to I old nodes. (These two kinds of networks are structurally very distinct as the former consists of nodes of approximately identical degree and the latter exhibits a power-law degree distribution.) Numerical results for both cases are shown in Fig. 4. As expected, for the Erdős-Rényi network with p > 1/N , the size distribution exponent remains unchanged. When p < 1/N , the network consists of small isolated components and hence big cascades cannot occur. The irregular size distribution P (S) observed for β = 5 · 10 −5 is due to topological properties of the particular network realization where the model was simulated (i.e., positions of respective ups and downs of the size distribution depend on the network realization). These results agree with a previous study of the sandpile dynamics [37] (see [38] for an extensive recent review of critical phenomena in complex networks). By contrast, Barabási-Albert networks yield cascade size distributions with significantly higher exponents (approximately 1.65) which is probably due to strong inhomogeneity of the network. When I = 1, P (S) deviates from a power law, probably as a consequence of the scalefree network topology (the same shape of the distribution is observed for different realizations of the network).
Role of the initial fragility values
While it sounds plausible that due to model's stochasticity, the initial fragility values have no influence on the equilibrium fragility distribution, the situation is in fact more complicated. For example, a simple numerical simulation with f i (0) = 1/N for all i shows a case where: (i) no stationary fragility distribution arises, (ii) at any time step, only a small number of distinct fragility values is observed (see Fig. 5). What causes the discreteness of fragility values? Denoting the number of failing and healthy time steps of node i as F i and H i , respectively, it must hold that F i + H i = t where t is the current time step. This node's fragility now can be written as When all f i (0) are identical, the possible values of f i (t) are discrete at any time step t and the ratio of neighboring possible values is (1 + β)/λ. If λ is small (as it is in our simulations), this ratio is large and hence the number of actually observed fragility values is small (because values much smaller or greater than the average fragility are unlikely). Eq. (15) implies that possible fragility values depend on t and hence there can be no stationary fragility distribution-this is confirmed by Fig. 5 where fragility peaks constantly shift to higher values and change their relative heights. Interestingly, even this peculiar setting of f i (0) does not alter the long-term model's behavior substantially and the aggregate quantities (such as the average failure probability or the cascade size distribution) are similar to those found for randomized initial fragility values before. Differences between neighboring peaks are λ/(1 + β), hence the time after which the fragility distribution pattern repeats can be estimated as ln λ/(1 + β) / ln(1 + β). Since this is a typical time of fragility evolution, one can use it also as an estimate of the initial equilibration time T eq . For the smallest value of β in our simulations (β = 0.005) we obtain T eq ≈ 4 600 which ex post confirms our setting of the equilibration time to 10 4 . Finally, note that while the random setting of f i (0) prevents discrete fragility values from appearing, some remnants of the initial fragility values can be preserved by Eq. (15). To ob-tain a fragility distribution truly independent of the initial values, one has to assume annealed dynamics, i.e. fragility updating by randomized values of β and λ.
Generalized model with partial memory
Fragility updating rules defined by Eq. (2) imply that nodes become more robust after a failure and hence autocorrelation of their failures as well as autocorrelation of the total number of failures are negative (their magnitudes depend on β and λ). As discussed in Section 2, this is true for majority of stocks but certainly not for all of them. To allow for repeatedly failing stocks, we introduce the probability α with which a failed node stays failed also in the next time step (and consequently acts as an additional initial failed node). This probability has the role of partial memory in the system and, as we shall see, gives rise to volatility clustering and other effects observed in real financial data. Note that memory or delayed stress propagation are quite often part of cascade spreading models as in, for example, [39]. We assume that fragilities of nodes which stay failed due to α are not updated in the given time step (when α is small, this assumption has little influence on the results).
Since P (F ) ≪ 1, the probability of a node's repeated failure is now P (F |F ) ≈ α which, in combination with the empirical results presented in Section 2, motivates us to set α = 0.04. We further choose β = 0.01 and λ = 0.1 which best correspond to the critical regime in Fig. 3. Using this setting we numerically obtain conditional probabilities consistent with those observed in the empirical data: P (F |F ) = 0.041 (empirical value is 0.039), P (F ) = 0.004 (empirical value is 0.003) and P (F |N ) = 0.004 (empirical value is 0.003). As long as we stay in the critical regime, these values depend on β and λ weakly. Presence of volatility clustering is confirmed by significantly positive autocorrelation of the number of failures C(n F (t), n F (t + 1)) ≈ 0.3 (empirical value is 0.15). The precise value depends on α and λ (and much less on β) but positive autocorrelation naturally arises for α which is large enough. By contrast, α = 0 yields P (F |F ) ≈ P (F ) and C(n F (t), n F (t + 1)) ≈ −0.04. Finally, Fig. 6 shows P (S) for different values of α. We see that for small values of α, the size distribution remains power-law with exponent gradually decreasing as α grows. Due to the additional complexity introduced by partial memory, an analytical cascade size distribution for this generalized model has not been obtained yet.
Discussion
We studied empirical stock prices and found that large simultaneous downturns follow a broad distribution consistent with a power law with exponent 2.19 ± 0.05. To reproduce this behavior, we proposed a minimal stochastic model of failure propagation. Using a mean field approximation and branching process theory we derived the general cascade size distribution and determined the range of parameters which give rises to the critical regime. To reproduce other features observed in financial data, such as volatility clustering, partial memory was introduced within the basic model. While our model implicitly assumes arrival of news to the market (they cause the initial nodes to fail and allow cascades to be created), we minimize the influence of news on the system's behavior by assuming their equal impact (in each time step, exactly one initial node is chosen to fail). This approach is motivated by the extensive study of excess volatility which shows that it is difficult to link the observed trading volumes and volatility to the arriving information [40] and even the large crash of 1987 does not seem to be triggered by particular news [41]. In reality, of course, the impact of news on the market differs from one day to another. It could be therefore interesting to test how different ways of choosing the initial failed nodes influence the model's behavior (for example, the number of the initial nodes can be random or network hubs may be preferentially chosen to trigger a cascade).
There is a number of other challenging questions which deserve further investigation. Firstly, since the cascade sizes corresponding to the secondary maximum in Fig. 3 are comparable with system size, this behavior cannot be described within the formalism of branching processes where an infinite system size is assumed. While we found an approximate condition for the appearance of the secondary maximum, how to proceed further towards an exhaustive description of the resulting size distribution is still an open question. Secondly, it would be interesting to find an analytical expression for the size distribution exponent in scale-free networks where it appears to differ from the mean-field value 3/2. Thirdly, generalized model with "partial memory", studied only numerically here, calls for analytical approaches. Fourthly, it would be interesting to know whether the model can be modified to produce power-law size distributions with exponents considerably higher than those reported here. One opportunity for such a generalization is to assume a dynamic network structure whose evolution depends on nodes' failures, similarly to the approach used in [42,43] for different models. Alternatively, as a generalization of the current binary model where nodes are either healthy or failed, one could define a multi or continuous-state model in which the probability of following a neighbor's failure depends on the failure's magnitude.
We stress that the probabilistic spreading mechanism proposed here is a general one and its use is not limited to market crashes or firm bankruptcies. For example, economic exchanges between countries are so intense that decline in one country may propagate to a neighboring one (take, for example, how growth in many European countries depends on spending of German consumers). On a two or three dimensional lattice, a similar mechanism might be employed to model earthquakes because, similarly to the proposed model, a failure at one place of the Earth's crust exerts some stress on its neighborhood (the number of failed nodes would then represent the earthquake size). In summary, the proposed model, together with its generalizations, has proven to be simple yet rich in behavior. It poses a variety of new research questions and we are looking forward to its future development and applications.
|
2010-11-10T06:32:56.000Z
|
2010-03-16T00:00:00.000
|
{
"year": 2010,
"sha1": "c78603412e53a1eaaa0d915a0a00f407e6568b6e",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/1003.3114",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "1baa9eaf866b8b4014ac0ae450a8446efc41a47a",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Mathematics",
"Physics",
"Economics"
]
}
|
55429229
|
pes2o/s2orc
|
v3-fos-license
|
First-principles derived complexion diagrams for phase boundaries in doped cemented carbides
This article reviews a method for calculating an equilibrium interfacial phase diagram depicting regions of stability for different interface structures as function of temperature and chemical potentials. Density functional theory (DFT) is used for interfacial energies, Monte Carlo simulations together with cluster expansions based on DFT results for obtaining configurational free energies, and CALPHAD-type modeling for describing the thermodynamic properties of the adjoining bulk phases. An explicit case, vanadium doped cemented carbides, is chosen to illustrate the progress in the research field and the interfacial diagram, a complexion diagram, for the phase boundary WC(0001)/Co is constructed as function of temperature and chemical potentials.
Introduction
Computational techniques have been developed to model properties of complex multicomponent materials. Methods such as CALPHAD are used to describe the thermodynamic phase behavior [1]. Experimental data and in some cases results from first-principles methods are then utilized in an efficient way to compute the corresponding phase diagrams.
For a polycrystalline material, not only the phase behavior is important, but also the structure, composition and energetics of its internal interfaces are decisive for the properties of the material. The study of interfaces is hence of great importance and is a growing and rapidly advancing field within materials science. It has been well established that the appearance of certain interfacial "phases" different from those of the adjoining bulk phases (e. g. segregation structures, amorphous wetting films, etc.) can be understood from thermodynamic considerations [2,3,4,5,6,7,8,9,10,11,12,13,14,15]; the disparate interfacial phases are equilibrium states stabilized by the interface. The term complexion has been suggested to separate these interfacial phases from ordinary bulk phases [16,17,18].
Development of computational methods to derive the corresponding interfacial phase diagrams has therefore become important. In contrast to the bulk behavior experimental data for the interfacial properties are scarce and hence one may expect results from first-principles methods to become more critical and decisive in the development of interfacial phase diagrams compared with ordinary bulk phase diagrams.
Here we review how interfacial phase diagrams can be constructed by combining results from first-principles density functional theory (DFT) calculations together with thermodynamic
Cemented carbides
Cemented carbides is a class of materials which consists of fine particles of a carbide cemented into a composite by a binder metal. The most common cemented carbide is produced by combining tungsten carbide (WC) with the binder metal cobalt (Co). Cemented carbides combine excellent hardness with high toughness and wear resistance [19]. They are used in cutting and wear resistant tools and are of considerable industrial importance [20].
The material is produced in a sintering process, in which carbide and metal powders densify into a microstructure of faceted WC grains embedded in the Co binder. Considerable research work has been focused on producing cemented carbides with WC grain sizes in the nanometer scale, since such a fine microstructure has the potential to dramatically improve on the mechanical properties of the material [21]. To mitigate WC grain growth and retain the fine structure of the starting powder, additions of VC, Cr 3 C 2 , NbC, TiC, TaC, etc. are made, of which especially VC works effectively as grain growth inhibitor [22]. The effects on microstructure of the additions have to large extent been empirically investigated, but still the exact mechanism behind the grain growth inhibition is unknown.
Several high-resolution electron microscopy studies have shown WC/Co interfaces that contain films with a thickness of a few atomic layers having an atomic arrangement incompatible with the underlying hexagonal WC lattice. The films have been seen especially in VC-doped WC-Co [23,24,25,26,27,28,29,30,31,32], where they have been associated with the cubic VC phase [25]. Thin films in WC/Co interfaces have also been imaged in Cr 3 C 2 - [33] and TiC-doped [34] cemented carbides.
It has been suggested, that thin films in WC/Co interfaces hinder the WC grain growth by forming a diffusion barrier for dissolving and/or reprecipitating W atoms [26]. However, the observed thin films at the WC/Co interface in the final material does not necessarily correspond to the equilibrium state during sintering [35]. The sintering process occurs at high temperature -including temperatures above the melting of the Co binder phase -where solubilities of W, C and dopant atoms in Co are high and where the interface region can be expected to be in a state of local chemical equilibrium. During cooling in the end of the sintering cycle, most of the dissolved atoms leave the Co binder phase [36] and reprecipitate onto existing carbide grains. It is thus conceivable that experimentally observed atomic structures and dopant segregation correspond to a lowtemperature equilibrium or some frozen-in state from the end of the cooling [37,38]. However, experimental studies using rapid temperature quench of the samples give strong evidence for the existence of thin films in WC/Co interfaces at the holding temperature during liquid phase sintering [30].
As most of the microstructural development occurs at high temperatures, it is essential to know the equilibrium state of the WC/Co interface in a doped WC-Co system as function of temperature and dopant addition i.e. one has to know the corresponding interfacial phase diagram. Only with this knowledge can a connection between WC growth kinetics and WC/Co interfacial structure and segregation eventually be established.
Simplified modeling
WC has a hexagonal lattice structure in cemented carbides. In the phase diagram, WC exists in a very small range of homogeneity with a carbon content corresponding to nearly perfect stoichiometry, meaning that in practice W and C vacancies can be assumed not to occur [39]. Furthermore, the solubility of transition metal atoms in WC is small. For V it is found to be only about 10 −3 atomic fractions at 1410 • C [40].
The hexagonal WC structure is the stable structure over the whole temperature interval relevant for sintering (below 2000 • C). However, a cubic WC phase can be obtained at sufficiently high temperatures (above 2500 • C) [41]. In contrast, the stable structures for other carbides, such as VC, TiC, and TaC, are cubic and the stoichiometry can deviate substantially from perfect.
During sintering, the WC grains develop a faceted triangular shape, bounded by basal {0001} and prismatic {1010} planes. Theoretical data for the WC/Co interface energies for the basal and prismatic planes are in the range 0.9-1.5 J/m 2 [42]. For the cubic WC structure the WC/Co interface energies are lower, about 0.5 J/m 2 for the WC(100)/Co interface [43].
In cemented carbides the stable lattice structure for WC is the hexagonal one. However, due to the lower WC/Co interface energy for cubic WC compared with hexagonal WC, one could speculate that the atomic stacking of the carbide near a WC/Co interface could transform locally into cubic to lower the interface energy. Indeed, experimentally a few layers of cubic stacking has been identified at some WC/Co interfaces in undoped WC-Co [44].
An intriguing scenario is then that in doped cemented carbides at low doping conditions the dopants may preferentially segregate to WC/Co interfaces and locally form intermixed cubic structures at the interface, interface complexions, stabilized by interfacial effects. At higher doping conditions the corresponding cubic carbide phase would precipitate, which may give undesirable properties to the cemented carbide material. With carefully controlled doping it would then be possible to preferentially influence the properties of the interfaces in the cemented carbide, while limiting the precipitation of the cubic carbide phase.
Simplified modeling
To model the possibility to create intermixed cubic structures, interface complexions, at WC/Co interfaces in doped cemented carbides we start with a very simple model. Consider a sharp interface between WC and the binder phase. The latter consists of mainly Co but also some dissolved metal atoms M, such as V, Ti or Ta, the dopants (see Fig. 1(a)). We then assume that the interface is replaced by a thin film of a cubic MC phase (see Fig. 1(b)). For a sufficiently thick film the interface can be regarded as split into a WC/MC interface and a MC/Co interface separated by a slab of MC. The energy for such a filmcovered interface can then be decomposed as where γ WC/MC (γ MC/Co ) is the interface energy between WC and MC (MC and Co). The last term of Eq. (1) represents the energetic cost per layer of building the MC phase and, thus, depends linearly on the number of layers N of the film. The energy ∆g MC is the negative of the driving force of nucleation of the bulk MC phase. It is positive if the MC bulk phase is unstable, if we are below the solubility limit of the metal atoms M in the binder phase. We consider coherently strained thin films and the last quantity of Eq. (1), e MC , is the strain energy per layer needed to bring the film into epitaxy with the underlying hexagonal WC phase. The energy γ film should be compared with the energy γ WC/Co of a WC/Co interface not covered by any MC film. The film coveraged interface is stable if γ film < γ WC/Co and, hence, a necessary condition for film formation in this simplified model is that i.e. one has to gain interfacial energy when the two interfaces WC/MC and MC/Co are created from the original WC/Co interface. In Tab. 1 we give the result for ∆γ MC for V, Ti, and Ta doped materials. The result is obtained from density functional theory (DFT) calculations and some computational details are given in Appendix A. In the calculations the bulk phases that are joined together are strained into coherency. The corresponding strain energy is cancelled, so that γ WC/Co , γ WC/MC , and γ MC/Co only represent the chemical contributions to the interface energy. The interface energy can then become negative and to model the true interfaces (see below) the elastic contribution has to be added. For all three dopants; V, Ti, and Ta, ∆γ MC is positive and there is a substantial energy gain when the films are formed. The positive values of ∆γ MC are mainly due to the large interface energy γ WC/Co for the WC/Co interface compared with the interface energies γ MC/Co containing the cubic carbide phases (cf. Tab.1). An interface energy can be expressed in terms of surface energies σ and the work of adhesion W according to the Dupré equation In Tab.2 the data both for the WC/Co interface with a basal (0001) hexagonal WC plane and for MC/Co interfaces (M=V, Ti, or Ta) with a cubic (111) plane are presented. Some computational details are given in Appendix A. The work of adhesion is a direct measure of the interface bond strength. Both for the hexagonal carbide WC and the cubic carbides MC (M=V, Ti, or Ta) the work of adhesion is similar, the interface bond strengths are similar. However, the surface energy for hexagonal WC is quite large and following the Dupré equation the large WC surface energy σ WC implies a large WC/Co interface energy γ WC/Co (and vice versa).
The computed interface energies γ WC/Co , γ WC/MC and γ MC/Co in Tab.1 only contain the chemical part. To make the results more realistic the elastic contribution has to be added. In Tab.1 we give the computed values for the elastic energy per layer e MC needed to bring the cubic films into epitaxy with the hexagonal WC phase. The mismatch varies considerably. It is 0.7%, 5.0% and 8.2% for VC, TiC and TaC, respectively, and hence the elastic energy also varies substantially. In Tab.1 we have explicitly added the effect of two cubic layers in the last column. The interface energy gain (∆γ MC − 2e MC ) is now only positive for VC. The prediction is therefore that for V doped material epitaxial thin film formation is possible, for Ti doped material maybe, but most likely not for Ta doped material [8].
Layer-by-layer approach
The next step in the modeling is to take the atomic scale layered structure into account. The decomposition of γ film in Eq. (1) is valid only if the film is sufficiently thick, so that the interior of the film is bulk-like and the interface energies are well-defined. For thin films, γ film will deviate from the linear dependence on N. An atomic-scale layer-by-layer calculation has therefore been performed in Ref. [5]. In Fig. 2 the result for γ film is shown for a stoichiometric cubic VC film on the WC (0001) basal plane. The VC film is stacked in the 111 VC direction with alternating layers of metal and carbon atoms. All different stacking sequences and the two different terminations of the WC basal plane have been investigated and the results for the lowest energy stacking sequence are presented in Fig. 2 as function of the number N of V layers. The theoretically predicted lowest energy stacking sequence for N > 1 is identical to the stacking sequence suggested from a high resolution transmission electron microscopy (HRTEM) study [28].
The filled circles in Fig. 2 show the result with no energetic cost per layer added to build up the thin film (i.e. ∆g VC + e VC = 0). The result should converge to (γ WC/VC + γ VC/Co ) for sufficiently thick films (the dashed line). We notice that the layerby-layer result is close to the asymptotic value already after two V layers. It implies that interface energy gain ∆γ VC is essentially obtained already after the addition of two layers. The relevant interactions are short ranged. Additional layers will then only add the energy cost (∆g VC + e VC ) per layer and we expect to obtain a minimum in energy for a thin film with two V layers.
The energetic cost ∆g VC depends on the temperature and on the chemical composition in the binder phase. At liquid phase sintering temperatures around 1400 • C and for relevant doping concentrations, ∆g VC varies between 0 and about 0.2 J/m 2 (0 to 0.09 eV per formula unit) [5]. This has been determined using the Thermo-Calc software [46] with a database for V additions in the W-C-Co system [47]. We have added (∆g VC + e VC ) = 0.12 J/m 2 per layer (0.055 eV per formula unit) to the data in It is clearly seen that the thin film structure has a minimum for 2 layers of V. Furthermore, this minimum will be present over a quite large temperature range. When the temperature is increased/decreased the slope of the dotted line will increase/decrease, but 2 layers of V will be the stable interfacial structure over a large temperature interval. When cooling and ∆g VC approaches zero the VC phase becomes stable and it can precipitate and start to grow, most likely on the thin films at WC/Co interfaces.
The two atomic layer thick VC film at the WC/Co interface is stabilized by interfacial effects and it is thermodynamically stable over a quite large temperature range. For this structure we may use the notion, an interface complexion.
Interfacial phase diagram
We now would like to construct the actual interfacial phase diagram. It should describe the equilibrium structure and composition of the WC(0001)/Co interface in vanadium doped cemented carbides as function of the intensive thermodynamic variables temperature T , pressure P and chemical potentials µ i . For more details, we refer to Ref. [10].
The propensity of the dopant to segregate to the interface is governed by its effect on the interface energy γ. In general, γ can be written as [48] Here, G is the Gibbs energy for the considered interface system, which contains a planar interface of area A. N i is the number of atoms in the system of component i with a corresponding chemical potential µ i determined by the reservoirs with which the interface is in equilibrium. The equilibrium interfacial state is given by the minimum of γ under fixed µ i , T and P.
Interface structural models
The first step is to define an interface structural model for which an expression of G for the WC(0001)/Co interface can be formulated. We are then guided by both results from highresolution transmission electron microscope (HRTEM) imaging [25,26] and from theoretical investigations [5]. The interface is assumed to have a layered geometry where each layer has well-defined atomic lattice sites. The structural model contains a set of WC layers with hexagonal stacking, mixed layers, and Co layers describing the binder phase. The mixed layers are in epitaxy with the hexagonal WC layers and consist of two metal layers and a carbon layer in between. The metal layers can have a mixed composition of V and W atoms and the corresponding site fractions are denoted y V and y W . We assume that y V + y W = 1, and hence, we neglect any metal vacancies and assume no Co occupancy in the mixed layers. We also neglect any carbon vacancies in the mixed region. The carbide phase is metal terminated against the binder phase and the Co binder phase may contain dissolved W, V and C atoms. However, in the present modelling we do not explicitly include these dissolved atoms in the binder phase. As usual, the hexagonal WC bulk phase is assumed to be perfectly stoichiometric.
Several different interface structural models are constructed (cf Fig. 3). They are denoted as k = 0, 1, . . . , 4. The model k = 0 corresponds to hexagonal stacking in the mixed region and with y W = 1 (y V = 0) we recover the unreconstructed WC(0001)/Co interface without any segregation of V. The model k = 1 corresponds to cubic stacking in the mixed region and with y V = 1 (y W = 0) the model derived in the layerby-layer approach [5] (see subsection 3.2) is obtained. For each structural model k there is a corresponding equilibrium interfacial state with energy γ k . The interface energy depends on the composition in the mixed region γ k (y V , y W ) and the equilibrium energy γ k is obtained by minimizing with respect to that composition. The final equilibrium interface energy γ eq and corresponding equilibrium structure are then given by the minimization with respect to k The computation of the interface energy γ k can be separated into three parts, a temperature independent part γ k 0 , a temperature dependent part γ k T and an environment dependent part γ env , This is schematically illustrated in Fig. 4, where G(T ) is the Gibbs energy for the interface system and G ref (T ) represents the Gibbs energy for a reference system used in the computations.
Reference states
The Gibbs energy for the reference system can be expressed as The use of well-defined and suitable reference states with chemical potentials µ ref i is an important part of the modeling strategy.
In the W-C-Co system, the WC+Co two-phase field borders to the three-phase fields WC+Co+graphite and WC+Co+η for high and low C content, respectively [39] (where η is a (Co,W) 6 C carbide phase). We use graphite as our reference state for carbon, i.e.
where g gr is the Gibbs energy per atom of graphite. The interface system is in equilibrium with a stoichiometric WC bulk phase, hence where g shp WC is the Gibbs energy per formula unit of hexagonal WC. The amount of V added to the WC-Co system will affect µ V , and thereby the interface system. A maximum of µ V is attained when the addition of V is so large, that a secondary carbide bulk phase forms. For V, this phase corresponds to (at typical sintering temperatures and C contents) a cubic (V,W)C x substoichiometric (x < 1) carbide of NaCl structure. To determine the vanadium reference state we use the corresponding stoichiometric carbide phase VC and define where g NaCl VC is the Gibbs energy per formula unit of fully stoichiometric cubic VC. Finally, for cobalt we use solid fcc Co as reference state, where g fcc Co is the Gibbs energy per formula unit of fcc Co.
Temperature independent part
The temperature independent part γ k 0 in Eq. (6) corresponds to the thermodynamic ground state of the interface at T = 0 K, i.e. the lowest energy configuration for given n W , n V , etc. For this part straight-forward DFT calculations are used to compute the energy difference within the methodology presented in earlier papers [49,50,5,8].
Temperature dependent part
The temperature dependent part γ k T in Eq. (6) can be written as and it is equal to the difference of temperature dependence of the Gibbs energy between the interface system ∆G(T ) = [G(T ) − G(0)] and the chosen reference system ∆G ref ( The temperature dependence of Gibbs energy can be caused by configurational changes, vibrational motion, magnetic and electronic effects as well as melting of the binder phase. The configurational changes in the solid phases are only present in the interface system, not in the reference system, and they are due to varying degree of W-V ordering in the mixed lattice sites. This effect is important and has been thoroughly taken into account. A cluster expansion (CE) has been developed for the mixed layers based on DFT calculated energies for a large set of different V-W configurations. The ATAT package [51,52,53,54] is used to derive the parameters in the CE by fitting to the DFT data. The CE coefficients are allowed to be different in the two different mixed layers. Monte Carlo (MC) simulations are then performed within the canonical ensemble and both the contribution to the temperature dependence of the energy (enthalpy) and entropy are determined [10].
Another contribution to the temperature dependence of the Gibbs energy is the vibrational motion. The contribution to γ k T will depend on the difference in the vibrational motion for 5 the atoms in the interface system compared with the atoms in the reference system. Due to the similarities between the interface system and the corresponding reference system we expect a large degree of cancellation to be present. In Ref. [5] the vibrational contribution to γ k T was estimated to be at maximum ±0.3 J/m 2 . It can be obtained from phonon spectrum calculations but the computational effort is substantial. In the present study we have therefore chosen to neglect the vibrational contribution to γ k T . More complex is the deviation that arises from the temperature dependent magnetic structure of Co. From theoretical calculations, it is known that the magnetic state of Co influences Co/carbide interface energies [55]. The absolute value of a Co/carbide interface energy can change substantially (∼ 0.4 J/m 2 ) when treating Co as either ferro-or nonmagnetic [55,8], although the relative differences between various Co/carbide interfaces do not [8]. Another complication is posed by the melting of Co that appear in the temperature interval relevant for sintering.
Accurate absolute values of WC/Co interface energies are therefore difficult to obtain, in particular at high temperatures. However, both the effect of Co magnetism and the liquid phase of Co should to large degree cancel when considering a difference γ k 2 (y V 2 , y W 2 )−γ k 1 (y V 1 , y W 1 ) between interfaces of different atomic setups and site fractions calculated at the same thermodynamic conditions T and µ i . Therefore, in the following, we will concentrate on the relative interface energies i.e. the difference in energy between an interface k containing V at site fraction y V and the unreconstructed (k = 0) interface containing no segregation of V. Strictly speaking, this means that we are calculating free energy differences between systems containing an interface rather than the actual interface energy. Furthermore, we do not consider any segregation of Co atoms into the mixed region. Neither do we consider W or V segregation on the binder phase side of the interface. Consequently, the number of Co atoms is the same in the two systems γ k (y V , y W ) and γ k=0 (0, 1), and hence the value for ∆γ k (y V , y W ) in Eq. (14) will be independent on the chemical potential for Co. It implies that the particular choice of µ Co (T ) for the current results for ∆γ is of no importance.
Environment dependent part
The environment dependent part γ env of Eq. (6) introduces the connection between the DFT computations and CALPHAD-type modeling. This is independent on the chosen structural model k. It is given by the difference between the chemical potentials for the chosen reference states µ ref i in the DFT modeling and the actual experimental chemical potentials µ i , Calculating chemical potentials with respect to chosen reference states is a standard procedure in thermodynamic software.
For well-defined experimental conditions (temperature, pressure and composition), the difference (µ ref i (T ) − µ i (T )) can thus be obtained given databases for the system at hand. An example is provided in subsection 4.6.
Interfacial phase diagram
The final step is the actual construction of the interfacial phase diagram. We will make predictions for a real V-doped WC-Co material, which was produced and investigated by Weidow et al. [56,57]. The composition of their material is given in Tab. 3 and details regarding its manufacture can be found in Ref. [56]. The V addition used is rather typical for WC grain growth inhibition and a signicantly smaller WC grain size was reported for this material compared with a reference straight WC-Co [56]. [46] together with data from Refs [39,47]. The stability line of the (V,W)C x phase is shown as a black curve (see text). To the right of the (V,W)C x stability line V is predicted to be dissolved in the binder phase. The green curve shows the calculated ∆µ VC − T path (cf Eq. 16) that corresponds to the experimental conditions in Refs [56,57]. Following the green curve, above 1550 K the (V,W)C x bulk phase becomes dissolved in the binder phase. The green cross marks the holding temperature during liquid phase sintering and the dashed green line indicates the behavior if one would had continued to increase the temperature in the experimental study. Data from Ref. [10].
We first determine the thermodynamic bulk behavior. For a small addition of V atoms µ V is low and the V atoms will be dissolved in the binder phase. By increasing the amount of V µ V will increase and a maximum will be attained when the secondary cubic (V,W)C x bulk phase is formed. This defines the (V,W)C x stability line. Using the Thermo-Calc software [46] with appropriate databases [39,47], we have determined the location of the stability line. It is shown in Fig. 5 (black curve) as function of the difference where g NaCl VC is the Gibbs energy per formula unit of the (hypothetical) stoichiometric cubic VC phase and µ V and µ C are the chemical potentials for V and C, respectively. The exact location of the (V,W)C x stability line in the ∆µ VC -T diagram depends also on the chemical potential for C. In Fig. 5 we have used the same value for µ C as derived from the experimental set up in Refs [56,57]. With increasing temperature ∆µ VC for the stability line increases more and more indicating that the real nonstoichiometric (V,W)C x phase becomes more and more stable compared with the (hypothetical) stoichiometric cubic VC phase. To the right of the (V,W)C x stability line V is predicted to be dissolved in the binder phase. If approaching the stability line from right the (V,W)C x bulk phase would start to nucleate and the size and thickness of such precipitates would not directly depend on thermodynamic parameters.
We can now plot in the same figure the ∆µ VC -T path that the system in the experimental study [56,57] follows. This path is also determined from Thermo-Calc modeling and the result is shown as the green curve. Following the green curve from low temperatures the (V,W)C x bulk phase is stable up to 1550 K. Above this temperature the (V,W)C x bulk phase is dissolved in the binder. At 1582 K the Co binder phase start to melt (the solidus temperature for the binder) and it is completely melted at 1620 K (the liquidus temperature for the binder). This is seen as two kinks along the ∆µ VC -T curve. Within the solidusliquidus region, ∆µ VC increases sharply due to the larger V solubility in liquid Co than in solid Co. The holding temperature during liquid phase sintering is 1683 K, marked by a green cross in Fig. 5. If one would had continued to increase the temperature in the experimental study the system would had followed the ∆µ VC -T curve shown as a dashed green line. Fig. 5. The red area shows the region with high-segregation, with a two monolayer thick (V,W)C film with cubic stacking (k = 1). The blue area corresponds to an interface with a hexagonal WC stacking (k = 0) and with a very minor segregation of V to the topmost metal layer. The color within the regions k = 0, k = 1, and k = 2, respectively, shows the equilibrium site fraction y V . The black lines mark the interfacial transitions. Data from Ref. [10].
We are interested in the region to the right of the (V,W)C x stability line in Fig. 5 and the possibility of substantial segregation of V to interfaces and the formation of intermixed cubic structures, thin films, in that region. In Fig. 6 we show our re-sults for the WC(0001)/Co interface, superimposed on the bulk result. Three different regions exist in the relevant part of the ∆µ VC -T diagram. For low ∆µ VC , the structural model k = 1, a thin film with cubic stacking, is associated with the lowest free energy. The two metal layers have mixed composition but are V-rich with y V ranging from 0.96 to 0.74 in the depicted region. Appearing only as a thin wedge in Fig. 6, a region with the structural model k = 2 expands at higher T and is associated with a lower V content than k = 1 with y V about 0.48. At around ∆µ VC = 0.3 eV, the structural model k = 0 region appears which has a low V content of, at its limits, y V = 0.01 and 0.05 at T = 1000 K and T = 2000 K, respectively. This corresponds to an interface with hexagonal WC stacking with a small segregation tendency of V to the topmost layer. Thus, the three different stackings are well separated in terms of total V content; 1.5 to 1.9 (k = 1), 1.0 (k = 2) or less than 0.1 (k = 0) equivalent V monolayers, and the use of the concept complexions and thermodynamic complexion transitions are well justified in the present case. Figure 7: Layer-dependent site fractions y V,1 and y V,2 and total V content y V,1 + y V,2 in terms of equivalent V monolayers (eq. V m.l.) (left axis) and relative interface energy ∆γ (right axis) for a WC(0001)/Co interface at the experimental set up used in Refs [56,57]. y V,1 (y V,2 ) is the site fraction of V in the metal atom layer closest to Co (second metal atom layer). Data from Ref. [10].
The V concentration along the experimental ∆µ VC -T path is shown in Fig. 7 (blue curve). At 1550 K the concentration corresponds to 1.93 equivalent V monolayers (eq. V m.l.). It decreases to 1.88 eq. V m.l. when the Co binder phase has melted and it decreases further to 1.86 at 1683 K, the holding temperature at the liquid-phase sintering stage. The layer-dependent site fractions are also shown as function of temperature. At 1683 K, the holding temperature, the reduction of the V concentration is located exclusively in the topmost layer (y V,1 ). If one would have continued to increase the temperature in the experimental study the theoretical prediction is that the film would have dissolved at 1780 K and above that temperature the top-most layer (y V,1 ) would have contained only a very minor concentration of V atoms.
In the same figure we also show the temperature dependence of ∆γ, the difference between the interface energy and the interface energy for an unreconstructed WC(0001)/Co interface (k=0) without any V segregation (cf. Eq. 14). At 1550 K the reduction is quite large, 0.84 J/m 2 . The reduction decreases with increasing temperature and at 1683 K, the holding temperature, it is only 0.14 J/m 2 .
The important theoretical prediction from this study is that the WC(0001)/Co interfaces are in the high-segregation state (k=1), they are covered by a two monolayer thick cubic (V,W)C film, throughout the liquid-phase sintering process in the experimental study by Weidow et al. [56,57].
Future challenges
The presented interfacial phase diagram is a first step towards a complete complexion diagram for the doped WC-Co system, a technologically relevant material. A straightforward improvement of the calculation methodology would be to add an explicit calculation of the vibrational motion of the solid phases to obtain a more accurate description of the temperature dependence of the free energies. Addition of carbon vacancies in the cluster expansion would also improve the modelling and will be important when considering the effect of varying the carbon content in the WC-Co system.
More challenging is a proper description of the magnetic structure of cobalt and its liquid phase. Progress is being made in modelling paramagnetic systems [58,59] and for the modelling of the liquid state it may be necessary to make use of DFT-based inter-atomic potentials [60,61] and the moleculardynamics technique to obtain sufficient sampling of relevant configurations. However, here we have focused on the calculation of free energy differences between systems containing an interface (cf Eq. 14) and rely on substantial cancelation effects. Accurate computations of the absolute value of interface energies at high temperatures for complex structures constitutes an outstanding challenge.
Our method is based on the use of defined interfacial structural models for which a quantitative expression for Gibbs energy can be formulated. Input from experiments with atomic scale structural resolution has been important to guide the construction of the interfacial structural models [28]. The research area would benefit from further experimental studies, in particular on the direct identification of both the atomic structure and its chemical nature of boundaries in various doped cemented carbides. Segregation in the binder phase of metal atoms and carbon to the phase boundaries should also be taken into account in future developments of the methodology. One may also envisage the use of DFT-based machine learning techniques to improve on the interfacial structural modelling.
Another key ingredient in our methodology is the use of CALPHAD-type of modelling of the adjoining bulk phases together with well-defined and appropriate reference states in the DFT calculations. The connection to real systems can then be made using available thermodynamic software [46]. It relies on developed databases and application to other dopants and materials may require development and refinement of appropriate databases. In general, a challenge is to apply the presented methodology to other dopants and to other phase boundaries in cemented carbides. By comparing with experimental information it would then be possible to establish the validity of our approach.
Doping of cemented carbides is used to mitigate the WC grain growth in the sintering process [22]. The phenomenon is experimentally well-established but the exact mechanism behind the grain growth inhibition is still unclear. The interfacial phase diagram can be used to tune sintering temperature and dopant addition to perform interfacial engineering. The diagrams are important as they provide information at the sintering temperature, where most of the grain growth occurs and where experimental tools with atomic scale resolution are rare [30]. Outstanding challenges in the future are to establish the connection between the segregated interfacial structures, the interface complexions, and the growth mechanism as well as their influence on the mechanical properties of cemented carbides.
Summary
We have here reviewed a method for calculating an equilibrium interfacial phase diagram, a complexion diagram, depicting regions of stability for different interface structures as function of temperature and chemical potentials. We use DFT for interfacial energetics, Monte Carlo simulations together with cluster expansions based on DFT results for obtaining configurational free energies, and CALPHAD-type modeling for describing the thermodynamic properties of the adjoining bulk phases.
The concept interface complexion is also demonstrated using a simplified model. We then show that atomically thin VC films can be stabilized at WC/Co boundaries by interfacial effects and that two VC layers become the thermodynamically stable configuration over a wide temperature range.
Appendix A. Computational details
All DFT calculations have been performed using the Vienna ab-initio simulation package (VASP) [62]. For the exchangecorrelation effects, we employ the generalized gradient approximation (GGA). The exchange-correlation functional is approximated in the Perdew-Burke-Ernzerhof (PBE) scheme [63] and the calculations are performed non-spinpolarized. A planewave implementation with projector augmented wave (PAW) potentials is used [64,65] and the plane-wave energy cutoff is set to 400 eV in all calculations. Partial occupancies are set with the method of Methfessel-Paxton [66] of first order with a smearing parameter of 0.2 eV. Atomic relaxations are performed until all atomic forces are smaller than 0.02 eV/Å.
For the interface energies presented in Tabs 1 and 2, the modeled interface orientation is the same as in Ref. [10]. For the carbide/Co calculations, we use metal-terminated carbide slabs containing 5 M + 4 C atomic (111) (for cubic MC) or (0001) (for WC) layers each with 3 atoms. The Co slab contains 7 (111) layers each with 4 atoms. For the carbide/Co calculations, a slab+slab setup without any vacuum region is used. For the WC/MC calculations, we use a setup with a total of 4 M, 4 W, and 7 C atomic layers. For consistency, we keep the supercell dimensions in the interface plane the same as in the MC/Co calculations, but due to incompatible stacking sequences between WC and MC, a slab+slab+vacuum (∼ 10 Å vacuum) setup is used in the direction perpendicular to the interface plane. The values presented in Tab. 1 corresponds to the stacking sequence of lowest energy and the C chemical potential is assumed to be in the graphite limit. For the corresponding free surfaces, separate carbide and Co slabs with the same number of atoms as in the interface calculations are used. The k-point sampling is done with a Γ-centered grid with a division of 9 × 9 × 1 to converge all interface energies to an estimated computational error within 0.03 J/m 2 .
|
2016-01-07T16:31:24.000Z
|
2016-01-07T00:00:00.000
|
{
"year": 2016,
"sha1": "5f91e3b89553bb97912ca9a56bf8ab6169d70b45",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/1601.01588",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "5f91e3b89553bb97912ca9a56bf8ab6169d70b45",
"s2fieldsofstudy": [
"Materials Science"
],
"extfieldsofstudy": [
"Physics",
"Materials Science"
]
}
|
258369386
|
pes2o/s2orc
|
v3-fos-license
|
Tubular basement membrane amyloid deposition: is it an indicator of renal progression in light chain amyloidosis?
Abstract In light chain amyloidosis (LA), the massive glomerular and vascular amyloid deposition leading to interstitial fibrosis/tubular atrophy (IFTA) is thought to be responsible for renal failure. The amyloid deposition in the interstitium and the tubular basement membrane (TBM) has received less attention in the study of LA. We, therefore, collected clinical and laboratory data on patients diagnosed with LA in our Nephrology Department and studied amyloid deposition in the TBM. Twelve LA patients were diagnosed by renal biopsy during a seven-year period. In 4 of the 12, amyloid deposition could also be detected in the TBM. In our first case of a patient with diabetes mellitus, non-amyloid fibrils resembling ‘diabetic fibrillosis’ were also seen by electron microscopy. Despite the double damage, IFTA was negligible, blood vessels were unaffected, and the glomerular deposition was segmental. In the other three cases, significant (>50%) IFTA and a severely reduced estimated glomerular filtration rate were already detected at the time of diagnosis and amyloid deposition was also observed in the blood vessels. These findings indicate the importance of TBM amyloid deposition in the progression of renal disease. This may represent a late-stage presentation of the disease with a heavy LC burden.
Introduction
in light chain amyloidosis (LA), abnormal insoluble proteins are formed mainly in the serum and then deposited in predilected organs. in the kidney, amyloidosis usually begins in the mesangium and later diffusely affects the other renal structures. Dominant glomerular deposition is the most common, resulting in significant nephrotic proteinuria [1]. When amyloid is loaded predominantly within arteries and arterioles, proteinuria is lower, and patients usually present with renal insufficiency [2]. Massive glomerular and vascular amyloid deposition leading to interstitial fibrosis/tubular atrophy is thought to be responsible for renal failure.
However, tubulointerstitial changes, alterations in the tubular epithelium, and interstitial fibrosis may also be consequences of interstitial amyloid deposition. Amyloid deposition surrounding the tubules, peritubular capillaries, and along the tubular basement membrane (tBM) has received less attention in LA research. However, the recently proposed standardized histopathological scoring system does take interstitial amyloid deposition into account [3]. in fact the amyloid score correlated well with renal outcome in an independent cohort [4]. recently, a large prospective study confirmed that an elevated urinary retinol-binding protein level, indicating proximal tubule cell dysfunction, is a good predictor of renal outcome in LA, independent of estimated glomerular filtration rate (eGfr), urinary protein-to-creatinine ratio (upCr), and clonal response [5].
Light chains can cause more than one type of kidney disease. for example, patients with LA may also have light chain deposition disease (LCDD) with or without light-chain cast nephropathy [6]. Non-organized, granular deposition along the tBM is well known in LCDD; however, little is known about the frequency, clinical and prognostic importance of amyloid fibril deposition along the tBM in systemic LA. the possible role of tBM amyloid in disease progression has not been investigated. therefore, we collected clinical and laboratory data of patients diagnosed with LA in our Nephrology department between 2015 and 2022 and studied amyloid deposition in the tBM.
Case reports
twelve LA patients were diagnosed by renal biopsy in the 7-year period. table 1 shows the essential clinical and histological data of patients. in 4 of the 12 cases, amyloid deposition in the tBM was also present (cases [1][2][3][4]. it was verified in all cases with Congo red stain positivity and apple green birefringence under polarized microscopy, light chain restriction with immunofluorescence microscopy and amyloid fibrils (diameter 8-12 nm) on electron microscopy.
Case 1
the medical history of a 61-year-old female patient included osteomyelitis of the right tibia in 1975, splenectomy due to immune thrombocytopenia, hypertension, treatment for type-2 diabetes mellitus since 2012, neoadjuvant chemotherapy for rectal cancer in 2013, followed by a Hartmann's procedure, colostomy closure, and left knee replacement in 2018. in September 2020, she was referred to a nephrologist because of increasing proteinuria (2 g/day), which began in 2018. She had diabetes that was well controlled with metformin and semiglutide. the kidney size and function were normal, as were routine blood tests and urine sediment. Serum protein electrophoresis (Spep) confirmed an igG lambda monoclonal component (6.3 g/l), which was also present in urine (7 mg/l), and the serum free light chain kappa/lambda ratio (sfLCk/l) was 0.13. Bone marrow flow cytometry revealed 0.45% plasma cells, of which 55% were pathological, showing lambda monoclonality.
A renal biopsy was performed, revealing only one sclerotic glomerulus among the 23 glomeruli, and negligible iftA (less than 10%). immunofluorescence revealed segmental linear igG and lambda deposition along the glomerular basement membrane and the tBM (figure 1). the non-immunological trapping of linear igG and albumin in thickened abnormal basement membranes is common, leading to nonspecific staining in patients with diabetes. Congo red staining was verified in the periodic acid-Schiff (pAS) negative mesangium, peritubula, and tBM, but not in the blood vessels. Among the non-organized diabetic fibrils in the expanded mesangial area and the tBM, electron microscopy (eM) confirmed immunoglobulin-related organized, 8-12 nm wide amyloid fibrils in the tBM also. the patient was diagnosed with diabetic kidney disease and lambda light-chain amyloidosis. there was no restrictive cardiomyopathy and the Nt-proBNp level was normal; however, carpal tunnel syndrome was verified by electroneurography. Chemotherapy was postponed because inflammation in the left knee necessitated the removal of the patient's prosthesis and the reinsertion of the prosthesis a year later. in August 2022, double-J stenting was performed for ureterolithiasis. the patient is on conservative treatment, eGfr is 88 ml/min/1.73 m 2 , proteinuria was unchanged, and urinary protein-to-creatinine ratio (upCr) is 225 mg/mmol now. Table 1. Clinical and histological data of patients with light chain amyloidosis.
Case 2
A 67-year-old man had a history of regular alcohol consumption and anemia requiring transfusion due to a bleeding duodenal ulcer and paroxysmal atrial fibrillation. in January 2022, he was referred to the nephrology department because of a rapidly decreasing eGfr of 17 ml/min/1.73 m 2 and increased proteinuria with a upCr of 596 mg/mmol. Both kidneys were normal in size, but a Ct scan verified the presence of a solid tumor in the liver. Spep revealed monoclonal gammopathy (igA lambda 2.4 g/l) with a normal sfLCk/l of 0.743 and no detectable monoclonal component in the urine. renal histology showed 29 subtotally sclerotic and 4 normal glomeruli, with an iftA of 60%. immunofluorescence staining showed monoclonal lambda light-chain restriction primarily in the mesangial region, blood vessels, and tubules. in these localizations, a pAS-negative, Congo red positive, apple-green birefringent material was detected under polarized microscopy. the eM examination confirmed 8-12 nm wide amyloid fibrillar structures. A subsequent liver biopsy confirmed cholangiocellular carcinoma; therefore, a liver segment resection was performed. During the long postoperative recovery period, the kidney function decreased, and hemodialysis was initiated. Chemotherapy was not initiated for AL amyloidosis, and both oncologists and hematologists recommended conservative treatment. Multiple liver metastases were observed, and the patient died in December 2022.
Case 3
A 69-year-old male patient had a myocardial infarction in 2002, and a history of kidney stones, Lyme disease, prostate hyperplasia, fatty liver, and diabetes mellitus. in february 2016, he was admitted to the nephrology department due to a decrease to 19 ml/min/1.73 m 2 in eGfr and nephrotic syndrome with a upCr level of 1332 mg/mmol. An increased sfLCk/l value of 7 was verified without detectable monoclonal components in the Spep. renal histology revealed one sclerotic glomerulus and 18 glomeruli with extensive kappa restriction with pAS-negative, Congo red positive and apple green birefringent material in the mesangium, efferent, afferent, and smaller arterioles. iftA was 50%. eM confirmed the presence of 8-12 nm wide amyloid fibrillar structure in the tBM also. Bone marrow biopsy revealed a 10% plasma cell infiltration; fiSH was negative, confirming monoclonal gammopathy of renal significance as a cause of kappa LA. the patient had no extrarenal involvement. Chemotherapy was initiated; however, unstable angina developed before the second cycle. urgent coronarography and coronary artery bypass surgery were performed; however, the patient died three days later.
Case 4
in october 2019, 55-year-old female patient presented with a history of kidney stones and cholecystectomy. proteinuria and microscopic hematuria were detected, with a slight decrease in kidney function. one year later, urgent hemodialysis was initiated due to an abrupt deterioration of kidney function (eGfr of 9 ml/min/1.73 m 2 ), and she was referred to our nephrology department. Both kidneys were normal in size, microscopic hematuria and nephrotic proteinuria (upCr:967 mg/mmol) were present. Both ANCA and a-GBM tests were negative. A renal biopsy was performed. Lambda LA involving the glomeruli, interstitium, and blood vessels was verified with advanced glomerulosclerosis and an iftA >60%. eM confirmed amyloid 8-12 nm fibrillar structure in the aforementioned localization and tBM. Spep detected free lambda light chains using only immunofixation. SfLCk/l was 0.078. Bone marrow biopsy revealed a 20% plasma cell infiltration; therefore, the final diagnosis was AL amyloidosis due to multiple myeloma. echocardiography showed restrictive cardiomyopathy, serum Nt-proBNp level was 19,831 ng/l; and there were no other signs of extrarenal amyloid deposition. Chemotherapy was then initiated, but sudden cardiac death occurred after the second cycle.
Discussion
We retrospectively reviewed patients with LA who had been diagnosed by renal biopsy in the last 7 years. Among the 12 LA patients, interstitial amyloid deposition was present in 8, including 4 with amyloid deposition involving the tBM. interstitial involvement appeared to be similar to that observed in large cohorts. fun et al. reported interstitial amyloid deposits in 63.7% (165/259) of LA cases [7]. in a Mayo Clinic cohort of systemic (but 85.9% immunoglobulin amyloid type) amyloidosis, the interstitium was involved in 58% of the cases, whereas tBM deposition was found in only 8% (38/470) of the cases [8]. Compared to patients without tBM involvement, 3 of our patients with tBM amyloid deposition were at a later stage of the disease, presenting with lower eGfr (9-19 ml/min/1.73 m 2 ), and higher rate of both sclerotic glomeruli and iftA. Late diagnosis probably contributed to the higher frequency of tBM involvement because the main serum creatinine was only 1.3 mg% in the Mayo Clinic cohort [8]. in 3 of 4 patients with tBM amyloid deposition (case 2-4), advanced glomerulosclerosis and tubulointerstitial damage were present at the time of diagnosis, and renal failure progressed despite chemotherapy. However, in case 1, sclerosis and fibrosis were negligible, and kidney function remained normal for 2.5 years without chemotherapy, even though the tBM was affected by both amyloid fibrils and non-amyloid fibrillosis. this indicates that amyloid deposition in the tBM alone does not play a significant role in tubulointerstitial damage. in contrast, there are opposing observations. in a case report of a patient with rheumatoid arthritis, tBM amyloid deposition with negligible glomerular and absent vascular involvement was proposed to be responsible for renal failure [9]. Kuruda found a similar frequency of circumferential tBM amyloid deposition in rheumatoid arthritis and LA; however, its influence on renal function remains unclear [10].
in cases 2-4, there was significant amyloid deposition in the arterioles as well as in the smaller arteries, which was absent in case 1. Among patients without tBM involvement, those with blood vessel amyloid deposition (cases 7-12), iftA varied between 10 and 50%, but in those who did not have vascular involvement (cases 5-6), iftA was less than 10%. the narrowing and occlusion of vessels by amyloid deposition may contribute to advanced tubulointerstitial fibrosis. in a report from the Mayo Clinic, 12 of 234 patients with AL amyloidosis had vascular-limited renal amyloidosis. Compared with patients with diffuse renal involvement, patients with vascular-limited amyloidosis have higher serum creatinine levels at the time of diagnosis and significantly shorter survival times [2]. Besides localization, amyloid load is also an independent predictor of AL amyloidosis prognosis. Computer-assisted software is currently used to estimate the amyloid area involved [7,11]. the missing vascular and only focal and segmental glomerular amyloid involvement indicated a limited amyloid load in case 1 which could also explain the favorable prognosis.
Case 1 had long-standing diabetes mellitus with histologically verified diabetic nephropathy; therefore, 'diabetic fibrillosis' emerged in the background as non-amyloid tBM fiber. these fibrils generally have a similar diameter (10 nm) as immunoglobulin-related amyloid fibrils (8-12 nm), especially in the expanded diabetic mesangium [12,13]. they may represent a type of fibrillary, perhaps glycosylated, collagen cross-linked to other proteins, making it impossible for them to be properly catabolized. the N-glycosylation process also plays a major role in LC amyloidogenesis [14]; therefore, the collagen fibrils may have helped in the precipitation of LC in our case. this can also explain the patient's carpal tunnel syndrome, which is common in both diabetes mellitus and LA, but its frequency is higher in LA, and can be the first warning sign of the disease.
Apart from the uncertain effect on renal function, amyloid deposition in the tBM was not associated with other clinical consequences. Glucosuria was observed only in one patient without tBM amyloid involvement. No other symptoms were suggestive of fanconi syndrome. this is typically observed in light chain proximal tubulopathy when the excess free light chain injures proximal tubular epithelial cells, sometimes with crystal formation in the cytoplasm [15]. Very rarely, intracellular amyloid, termed amyloid proximal tubulopathy, has been detected in such cases [16]. one of our patients experiencing LA without tBM amyloid involvement (case 11) had LC cast nephropathy, but no amyloid casts were formed, which is not a rare phenomenon in such cases [6,17]. None of the patients had polyuria or diabetes insipidus, which have been reported in cases with predominant tubulointerstitial amyloid involvement [18]. this is mostly observed when the medullary region is involved. in cases with non-significant glomerular deposition (resulting in less proteinuria), the clinical presentation could suggest acute interstitial nephritis, especially in patients with a predisposition, such as in Sjogren syndrome [19]. rarely, giant cell reactions provoked by interstitial amyloid deposits are observed, typically in AA amyloidosis, but they also occur in LA and LCDD [20,21].
Amyloid formation does not depend on the amount of light chains but on mutations. the lambda light chain is significantly more amyloidogenic than the kappa light chain. the variable region of the lambda light chain sequences has the most extensive non-conservative and total number of mutations compared to all other sequence groups [22]. Lambda LC was also overrepresented in our LA cohort; however, with respect to tBM involvement, the type of LC was not different (kappa/lambda ratio of 1/3 vs. 2/6).
there is substantial information about renal amyloidogenesis in LA; however, amyloid deposition within the tBM is not well established. Amyloid formation begins in the mesangium. the normal smooth muscle phenotype of mesangial cells changes to a macrophage phenotype upon recognition of precursor light chains. Amyloidogenic light chains are internalized in mesangial cells and transported to lysosomes for digestion to form amyloid fibrils [13]. this process is the same in the proximal tubular cells. they are able to reabsorb the immunoglobulin light chain through the cubilin-megalin receptors and by clathrin-dependent endocytosis, and amyloid may form in tubular cells. Accordingly, during the eM examination of the tBM, it was established that in cases 1 and 4, amyloid accumulated more on the cell side of the tBM; however, in case 2, it was dominant on the interstitial side. this may suggest that amyloid deposition can also occur from the interstitium; although peritubular capillaries have a fenestrated endothelium, their basement membrane, like the glomerular basement membrane, is impermeable to amyloid.
Based on this small cohort study, we would like to raise awareness and emphasize the importance of finding amyloid deposition in the tBM. this usually represents a late-stage presentation of the disease with a heavy LC burden. However, it can occur in cases with a smaller amyloid load and does not in and of itself lead to the deterioration of kidney function or tubular dysfunction. Amyloid deposition in the tBM depends not only on the concentration and properties of folding immunoglobulins, but also on the local environment. in the present case (case 1), 'diabetic fibrillosis' may have served as a precipitating factor.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding this study was supported by the university of Debrecen, Kálmán Laki Doctoral School of Biomedical and Clinical Sciences.
|
2023-04-28T15:04:34.354Z
|
2023-04-26T00:00:00.000
|
{
"year": 2023,
"sha1": "3ab4aa8424f9a6b6fdfd0bc92351461c8f27de25",
"oa_license": "CCBYNC",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "235245820f45c1ac4f0da9fc15043b15842b2a79",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
236470345
|
pes2o/s2orc
|
v3-fos-license
|
Associations of Electromyographic Activity of Anterior Temporalis Associations of Electromyographic Activity of Anterior Temporalis Muscles, Sex, and Occlusal Classes in Asymptomatic Young Muscles, Sex, and Occlusal Classes in Asymptomatic Young Adults Adults
Symmetry evaluation of the craniofacial complex generally involves models of mandibular movement and masticatory muscle activity, especially during the growth of the craniofacial complex. Objectives: The aim of the study was to determine what, if any, associations exist between the activity of the masticatory muscles, sex, and occlusal classes in asymptomatic young adults. Methods: 18-year-old volunteers, showing no symptoms of TMD based on an RDC/TMD examination, were included in the study.Surface electromyography (sEMG) recording was used to quantify the activity of masticatory muscles. The occlusal contact points were analyzed using a T-scan III Evolution 7.01 device. Occlusal classes were graded, employing an approach based on plaster study models. Results: In Class I–II subjects, we found significant differences only in the voltage of LTA in correlation with the gender and occlusal Class. Conclusions: Our findings show that the electromyography voltage of LTA significantly differs according to sex and occlusal Class. The voltage is higher in the female occlusal class II group, while the voltage is less in the male Class I and II group. This may be responsible for the symmetry index, which shows the predominance of the right-side muscles in all gender and occlusal groups.
INTRODUCTION
During the development of the craniofacial complex, multidirectional forces of the mastication muscles act on the teeth undergoing eruption. The place in which the tooth erupts depending on where opposing forces are in equilibrium. Interarch tooth alignment, which during the closing of the mandible, determines occlusal contact of the teeth, determines its quality. The contact between the opposing teeth generates reflex stimuli causing muscles to contract. 1 The distribution of the intercuspal contact between opposing teeth may be related to the spread of electrical voltage of paired masticatory muscles during contraction. This symmetry can be investigated with electromyography (EMG). It is indeed difficult to directly examine tension using elementary EMG of the muscles during movement. 2 During the examination, it is necessary to place the unipolar needle electrodes inside the muscles. Surface EMG, is vulnerable to extramuscular factors that can interfere with electrical signals of exact muscular origin, whereas the results obtained with well-standardized protocols closely ref lect intramuscular recordings. 3 A T-scan device allows the percentage distribution of occlusal contact (occlusal force) to be recorded. To the best of our knowledge, the relationships between masticatory muscle activity at maximal occlusal contact by gender have not been reported so far. Using these two methods, and by additionally considering the occlusal classes, we sought a relationship between the asymmetry of the masticatory muscles and occlusal classes in a homogenous group of asymptomatic 18-year-old young adults. Additionally, in the foregoing study was included relatively small number of patients and the groups were not homogeneous in age; they were examined when differences in muscular activity were associated with occlusal trauma.
In Ferrario's study, electromyography activity was measured in 92 subjects (49 male and 43 female), 20-27 years old, and selected from 160 Caucasian dental students with the absence of moderate and severe clinical mandibular disorders. 4 They were not divided into the occlusal class group, and they may have had minor clinical mandibular disorders. In another study, Ferrario analyzed 30 participants (15 women and 15 men) aged 18-19 years of occlusal Class I only. 3 We focused the analysis on subjects aged 18 years when craniofacial development has virtually terminated, and possible pathological influences can be considered minimal.
OBJECTIVE
Our aim was to validate the hypothesis that there exist associations between the activity of the masticatory muscles-namely, the left temporalis anterior (LTA), the right temporalis anterior (RTA), the left masseter muscles (LMM), and the right masseter muscles (RMM)-gender, and I, II, III Angle'a Classes in nonpatients young adults.
METHODS
We invited Caucasian school seniors aged 18 years from two high schools. In the screening phase, 268 out of the 1000 invited subjects who were willing to participate in the project underwent clinical examination. They were examined and completed the RDC/TMDtemporomandibular disorder questionnaire in polish version. [5][6][7][8] All the volunteers had signed informed consent forms. The research was conducted following the Declaration of Helsinki ICH Guidelines for Good Clinical Practice. The research study had been approved by the Ethical Committee of Jagiellonian University. The inclusion criteria were: a full dental arch, no symptoms of TMD according to RDC/TMD the symmetry of facial structures. The exclusion criteria were tooth decay, periodontitis, any fixed or removable denture, parafunction and any pathology in craniofacial structures.
Finally, 145 subjects underwent the study. To the study, we chose surface electromyography (sEMG) an 8-channel BioEMG III BioPAK Measurement System Electromyograph (BioResearch, Inc., Milwaukee, WI, USA) and to examine the occlusal contact points between tooth was used T-scan III evolution 7.01 device (Tekscan Inc., South Boston, MA, USA). The T-scan III / BioEMG has a synchronization module that allows for the simultaneous acquisition of occlusal function and associated muscle activity, which were examined. 9 All examinations were performed in the morning. At first, the volunteers were seated isolated in the exanimation room in a quiet environment listening to relaxing music. At first, the skin was cleaned with alcohol, the bipolar surface electrodes (Bio FLEX, BioResearch Associates, Inc., Brown Deer, WI, USA) were placed bilaterally on the subject's skin overlying the anterior temporalis -vertically along the anterior muscular margin. Electrodes for the masseter muscles were placed parallel to the muscle fibers. 10 A plate ground electrode was secured to the forehead. The sensor of the T-scan was placed in the mouth. During the examination, they seated upright on a chair, and the volunteer was asked to close their jaws as hard as possible. When a force of occlusal contact is exerted on the sensor, the simultaneous record of the distribution of occlusal contacts and voltage of the examined muscle is started. The record is displayed for clinical analysis. 10,[11][12][13] They were asked to hold their teeth hard together for 3 seconds and then open. 14 All registrations were repeated two more times. All details of this protocol are described in an article by Wieczorek et al. 9,15 When the point 100% of possible contact was reached, this same point was selected from the voltage of the anterior temporalis muscle (TA) and the masseter muscle (MM). This was possible thanks T-scan III / BioEMG III integration software, which is showing the voltage of the muscles at any moment of occlusal contacts was reached.
As described previously, measurement variability was assessed by repeated sEMG analysis. 9 Accuracy and precision were represented by the intraclass correlation coefficient (ICC). To describe the asymmetry of the masticatory muscles, an asymmetry index was calculated using the following formula based on the root mean square (RMS). Asymmetry index (AsI) = (RMS right -RMS left )/(RMS right + RMS left ) × 100 The value of AsI can vary from + 100 to -100, with an AsI of +100 or -100 being equivalent to an absolute predominance of right-side (R) or left-side (L) muscle activity, respectively, whereas a perfectly balanced activity of R/L muscles is represented by an AsI of zero. AsI was proposed by Naeije et al. 16 The subjects were classified into three occlusal classes using plaster study models taken during examination by alginate impressions of the maxillae and mandibles of the volunteers.
Statistical analysis
All data were analyzed using the SPSS (Statistical Package for Social Sciences) Statistics 17.0 (2008) for Windows. Agreement with a normal distribution was tested using the Kolmogorov-Smirnov test (with Lillefors' correction) and the Shapiro-Wilk test. For normally distributed variables, the differences between 3 occlusal classes were tested by one-way ANOVA followed by post-hoc Tukey's HSD test for unequal n.
Differences between sex were assessed by Student's t-test or the Mann-Whitney U test. The Intraclass Correlation Coefficient (ICC) was quantified with the F-test. A p-value below 0.05 was taken as significant.
RESULTS
The ICC of measurement variability was calculated to be 0.765. The entire group of 145 volunteers was divided according to occlusal Class, based on analysis of the plaster models created during the examination. Ninety-seven (97) volunteers were classified as occlusal Class I, 36 as Class II, and the remaining 12 subjects presented Class III features. No significant interclass differences in right-side and left-side activity of the MM or TA were found (Table 1). We next excluded from the study those samples with occlusal Class III, because of the number of participants involved.
In the next step, analysis of the voltage of all muscles according to sex and occlusal Class was performed. The only significant finding was an association between LTA voltage and sex and occlusal Class (p = 0.02). Additionally, we observed a lower LTA voltage in men and a higher LTA voltage in women of occlusal Class II ( Figure.1) To determine if this lower voltage inf luences the balance of masticatory muscles, calculation of the AsI, which showed the predominance of the right-side muscles was done. This tendency towards balance was shown only in occlusal Class II in the female group. (Figure. 2)
DISCUSSION
Our findings show that the electromyography voltage of the LTA significantly differs from gender and occlusal Class. The voltage is higher in the female occlusal class II group, while the voltage decreases from the male Class I to class II group. The symmetry index (AsI) shows the predominance of the right-side muscles in all gender and occlusal groups.
We analyzed EMG data recorded at 100% occlusal contact by examining young adult volunteers lacking signs of temporomandibular disorders (according to the RDC/TMD). The age group was chosen to guarantee that the development of the stomatognathic system would have finished and that wear on the masticatory organs would be minimal. Analyzing the asymmetry index, we observed the predominance of the right-side muscles in all occlusal classes. The differences in the asymmetric index were not significant across occlusal Classes.
One interesting finding occurred as a result of dividing occlusal Class groups by gender. Significant differences were observed between gender groups and occlusal classes in the LTA voltage. These differences probably directly influenced differences in the asymmetry index, which show the predominance of the right-side muscles in the subjects of classes I and II. The dominance of right-side muscles was observed in the study by Ferrario et al. 3 , but the participants were selected from 20-27 years old dental students. They were not divided by occlusal Class. Additionally, there were no significant differences in the gender groups between the electrical potentials of LTA or RTA-only between LMM and RMM. A literature search on the relationship between the distribution of occlusal contacts and EMG of the masseter and anterior temporal muscles did not reveal many studies involving a comparable age group.
The studies we did find aimed at understanding the influence of artificial interference on the voltage of the temporalis anterior and masseter muscles 14 or psychological status. 17 Some researchers have examined the asymmetry between the left and right muscles during clenching. 3 Suvinen et al. 18 have reported that, even in the mandibular postural position, asymptomatic muscles are physiologically asymmetrical and that asymmetry and activity indices of 4% and 17%, respectively, should be considered normal. In analyzing sex and occlusal Class, we found that the only muscle that differed was the LTA.
Ferrario et al. showed that healthy individuals have a prevalent side on which they generate higher muscular activity levels during bilateral clenching. 4 However, this should affect the activity of the masseter muscles, and we did not find any association between sex and occlusal class group for these muscles. However, in these two groups, it was also found that the predominance of intensity was on the right side.
Additionally, in our previous study, we found significant differences in RTA voltage between sexes, with the male group showing a significantly lower the voltage than the female group. 15 Based on our results, it may be concluded that, in asymptomatic patients, asymmetry of electromyography potential of mastication muscles tends to affect the right side. It could also be interesting to determine whether the mandible structure affects the electromyography voltage of the LTA and the asymmetry of the masticatory muscles. Lower LTA voltages do affect the ideal balance of the masticatory muscles, and the asymmetry index shows the predominance of the right-side muscles.
The T-scan III can be integrated with the BioEmg III system. In this way, synchronized clinical data can be recorded simultaneously. 13 At present, the T-scan III/ BioEMG integration software is considered the best tool on the market. It allows simultaneous recording and analysis of the sEMG of selected muscles and the distribution of occlusal force while correlating specific occlusal moments with specific electromyographic changes.
CONCLUSION
Our findings show that the electromyography voltage of the LTA differs significantly by sex and occlusal Class, with the voltage being higher in the female occlusal class II group, and decreasing from occlusal Class I to II the make group. This may be responsible for the symmetry index, which, in all gender and occlusal groups, shows the predominance of the right-side muscles.
|
2021-07-27T14:26:06.427Z
|
2021-04-30T00:00:00.000
|
{
"year": 2021,
"sha1": "5716a5579458de7aa6dfdcc1c9412d2e884a6c62",
"oa_license": "CCBYSA",
"oa_url": "https://scholarhub.ui.ac.id/cgi/viewcontent.cgi?article=1171&context=jdi",
"oa_status": "GOLD",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "11a3e432ad96daf13cdfd622330a8ef1da98f858",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
260132983
|
pes2o/s2orc
|
v3-fos-license
|
Disulfiram blocks inflammatory TLR4 signaling by targeting MD-2
Significance In response to pathogen/damage-associated molecular patterns, Toll-like receptor 4 (TLR4) initiates a host innate immune inflammatory response to control infection, tissue repair, and regeneration, while aberrant activation of TLR4 causes tissue damage and various inflammatory diseases. Therefore, TLR4 signaling is an attractive drug target for the treatment of TLR4-driven inflammatory disorders. Here, we showed that the FDA-approved drug disulfiram inhibits TLR4 signaling in both human and mouse cells by modifying a cysteine residue of MD-2, a critical cofactor of TLR4, and found that DSF markedly ameliorates TLR4-mediated pathology in an aggressive mouse model of Parkinson’s disease (PD). This study suggests an approach to treating TLR4-mediated inflammatory diseases.
DSF Specifically Inhibits LPS-Triggered Inflammatory Response.
Both DSF and Bay 11-7082 potently inhibit gasdermin D (GSDMD) pore formation to block inflammatory cell death (31,32).Because Bay 11-7082 also antagonizes IKK and thus inhibits NF-κB signaling (33,34), we assessed whether DSF might also inhibit the NF-κB pathway that primes inflammasome signaling.To examine the effect of DSF on NF-κB activation, we pretreated mouse immortalized bone marrow-derived macrophages (iBMDMs) with DSF and then stimulated them with different NF-κB signaling activators, including tumor necrosis factor α (TNFα, ligand of tumor necrosis factor receptor), peptidoglycan (PGN, ligand of TLR2), LPS (ligand of TLR4), and flagellin (FLN, ligand of TLR5), and assessed NF-κB transcriptional activation.In the absence of DSF, all of these stimuli potently activated mRNA expression of the NF-κBdependent genes, Il1b, Il6, Il12b, Tnfa, Ccl2, Ccl5, and Cxcl10 within 4 h (Fig. 1A and SI Appendix, Fig. S1 A-C).Unexpectedly, DSF specifically blocked TLR4-dependent NF-κBmediated gene expression in response to LPS, but only modestly inhibited Il1b expression and either increased or did not significantly change the expression of the other NF-κBdependent genes tested after TNFR, TLR2, or TLR5 stimulation (Fig. 1A and SI Appendix, Fig. S1 A-C).In parallel, DSF blocked TLR4, but not TLR2 or TLR5, stimulation of interferon (Ifnb) and the interferon-stimulated gene 15 (Isg15) (Fig. 1A, and SI Appendix, Fig. S1 B and C).DSF inhibited LPS-mediated induction of both NF-κB and IRF3 regulated genes in a dose-dependent manner -DSF inhibition was detected at the lowest concentration tested (1.25 µM), and gene induction was completely blocked at 5 and 10 µM (Fig. 1B).DSF was even more active at blocking TLR4 signaling than at blocking canonical and noncanonical inflammasome-induced GSDMD pore formation and pyroptosis, which was previously reported in dose-response experiments to plateau at concentrations >20 µM in the same cells (31).To confirm the effectiveness of DSF to inhibit LPS activation of TLR4 signaling, triplicate libraries prepared from iBMDMs that were pretreated or not with DSF and then challenged or not with LPS were analyzed by RNA-seq (35).LPS-triggered induction and inhibition of downstream genes was blocked in the presence of DSF (Fig. 1C).The genes up-regulated in response to LPS, whose induction was blocked by DSF, were proinflammatory genes known to be induced by TLR4 activation (Fig. 1D).The inhibitory effect of DSF on TLR4 signaling was also confirmed in human monocyte-like THP-1 cells (SI Appendix, Fig. S1D).These data suggest that DSF potently inhibits TLR4 signaling in both human and mouse cells.
DSF Inhibition of TLR4 Signaling Is Independent of GSDMD and
Caspase-1.Because DSF directly targets GSDMD and blocks its pore-forming activity in cells and also inhibits GSDMD and the inflammatory caspases in vitro, we assessed whether the inhibitory effect of DSF on TLR4 signaling was caused by the blockade of the inflammasome-GSDMD pathway by DSF.To this end, Gsdmdor caspase-1-deficient iBMDMs, generated using CRISPR-based gene editing (SI Appendix, Fig. S3 A-C), were stimulated with LPS in the presence or absence of DSF.The protein level of key factors in TLR4-mediated NF-κB activation, including Irak4, IκBα, and p65, was unaltered by Gsdmd or Casp1 knockout (SI Appendix, Fig. S3C).In response to LPS treatment, the induction of Il1b, Il6, and Il12b in Gsdmd-or caspase-1-deficient iBMDMs was similar to that in wild-type (WT) cells, indicating that neither Gsdmd nor caspase-1 is involved in TLR4 signaling (SI Appendix, Fig. S3D).DSF was unimpaired in inhibiting the LPS induction of these inflammatory cytokines in Gsdmdor caspase-1-deficient iBMDMs compared to WT iBMDMs.Thus, DSF blockade of TLR4 signaling does not depend on the inflammasome-GSDMD axis.
To further assess the dual effects of DSF on both TLR4-mediated priming and GSDMD membrane pore formation, we compared WT iBMDMs treated with DSF before and after priming, with Tlr4 −/− and Gsdmd −/− iBMDMs.Induction of Il1b mRNA and mature IL-1β secretion are early-phase (priming) and late-phase (pyroptosis) readouts, respectively.As expected, application of DSF before, but not after LPS treatment, blocked priming (SI Appendix, Fig. S3 E and F).No priming occurred in TLR4-deficient iBMDMs under either condition (SI Appendix, Fig. S3 E and F).In comparison, and as expected, nigericin-triggered IL-1β secretion, which depends on TLR4 for Il1b expression and on GSDMD for release, was abrogated in WT iBMDMs by DSF and in both Gsdmd-deficient and Tlr4-deficient iBMDMs in the absence of DSF (SI Appendix, Fig. S3 G and H).These data highlight the dual roles of DSF in inhibiting both upstream and downstream inflammatory responses.
TLR4 Sensing of LPS Is the Main Target of DSF.
To elucidate the cellular mechanism of DSF acting on TLR4 signaling, we first assessed the effect of LPS on NF-κB expression using a luciferase reporter assay with LPS stimulation or exogenous overexpression of signaling molecules involved in TLR4 signaling.Consistent with DSF suppression of NF-κBinduced gene expression (Fig. 1), DSF suppressed LPS activation of the NF-κBluciferase reporter (Fig. 3A).However, although ectopic overexpression of key TLR4 signaling molecules, including MyD88, TRAF6, TAK1 + TAB1, IKKβ, and p65, robustly induced the NF-κBluciferase reporter (Fig. 3 B and C), DSF did not suppress their activation of the NF-κBluciferase reporter.These results suggest that DSF interferes with LPS sensing by the upstream TLR4 signalosome rather than by inhibiting molecules that mediate downstream signaling.This finding is consistent with the specificity of DSF for TLR4 and not TLR2 or TLR5, which use the same downstream signaling pathway.Next, we examined by immunoblot (Fig. 3D) and immunofluorescence microscopy (Fig. 3E) how DSF affected the key molecular events of TLR4 signaling.LPS triggered the phosphorylation of IKKβ, IκBα, p65, degradation of IκBα, nuclear translocation of p65, and induced expression of IL-1β, which were all blocked with DSF pretreatment.By contrast, DSF had no effect on TNFα-induced nuclear translocation of p65 (SI Appendix, Fig. S4).These data suggested that TLR4 sensing of LPS is the main step inTLR4 activation that is blocked by DSF.
DSF Modifies MD-2 on Cys133 to Inhibit TLR4 Activation.As a thiol-reactive compound, DSF targets and covalently modifies substrate proteins on cysteine residues, which prevents their oxidation.In cells glutathione is the major molecule responsible for reducing oxidized cysteines.Glutathione is depleted during cellular oxidative stress and its levels can be increased by treatment with N-acetyl-L-cysteine (NAC) or antioxidants such as the superoxide scavenger Tiron.To determine whether the inhibitory effect of DSF on TLR4 signaling depends on its thiol-reactivity, iBMDMs were pretreated with Tiron or N-acetyl-L-cysteine (NAC) before they were stimulated with LPS.Both Tiron and NAC markedly reduced DSF inhibition of LPS-mediated proinflammatory cytokine induction (Fig. 4 A and B), suggesting that thiol-reactivity is required for DSF to inhibit TLR4 signaling.MD-2 is an extracellular glycoprotein that binds to the LRR sensing region of TLR4 and enhances TLR4 ligand binding in a process that depends on a reactive cysteine (Cys133), which is evolutionarily conserved (SI Appendix, Fig. S5) (38,39), suggesting that MD-2 might be the target of DSF.To test this hypothesis and identify any Cys that reacted with DSF, nano-liquid chromatographytandem mass spectrometry (nano-LC-MS/MS) was performed on recombinant MD-2 treated or not with DSF and then treated with iodoacetamide to stabilize any free cysteines.Tryptic fragments identified covalent attachment of DTC to MD-2 Cys133 (Fig. 4C).In agreement with these observations, DSF abrogated the cell internalization of LPS with TLR4 and MD-2, but had no effect on the expression of cell surface TLR4/MD-2 complex (Fig. 4 D and E), which is consistent with previous findings that MD-2 Cys133 is critical for the TLR4/MD-2 complex to bind and be activated by LPS (39,40).Addition of untreated recombinant MD-2 to the medium of iBMDMs pretreated with DSF partially rescued LPS-stimulated inflammatory cytokine gene expression (Fig. 4F).Thus, DSF inhibits TLR4 sensing by modifying Cys133 on MD-2.
To determine whether cysteine modification is a more general mechanism to inhibit TLR4 signaling, we tested the effects on TLR4 signaling of two other cysteine-reactive compounds, afatinib and DMF.Like DSF, both afatinib and DMF significantly down-regulated LPS-triggered induction of Il1b, Il6, and Il12b in a dose-dependent manner (Fig. 5 A and B).However, afatinib and DMF inhibited TLR4 signaling less than DSF (Fig. 5 A and B).Both afatinib and DMF inhibited LPS-triggered phosphorylation of IκBα and p65 and induction of IL-1β, IL-6, and TNFα protein (Fig. 5 C and D and SI Appendix, Fig. S6).TLR4 inhibition by afatinib and DMF was mediated by their cysteine-reactivity because NAC reduced their inhibitory effects on inflammatory cytokine expression (Fig. 5 E and F).Collectively, these data highlight the critical role of protein cysteines in TLR4 signaling and show that other cysteine-reactive compounds also inhibit TLR4 sensing of LPS.
DSF Protects Mice in an Aggressive PD Model.TLR4 has been suggested to play a major role in PD pathogenesis, since Tlr4 −/− mice are partially protected in the MPTP-induced mouse model of PD (19,29,30).Tlr4 −/− mice show less aggregated αsynuclein, reduced neuroinflammation, including less NF-κB activation, and less neuronal cell death in the substantia nigra (SN) and a reduction in motor deficits.TLR4's role in PD pathogenesis has been presumed to be activation of TLR4 signaling in microglia and other infiltrating myeloid cells, which highly express TLR4, but this has not been demonstrated.It is worth noting that TLR4 is up-regulated in the SN of MPTP-treated mice and that this region of the brain also has reduced glutathione levels, factors that may contribute to the high sensitivity of this brain region.Thus, to probe the in vivo effect of DSF on TLR4 signaling and inflammatory pathology, mice challenged with MPTP were administrated DSF intraperitoneally around the time of challenge (Fig. 6A).Without DSF, MPTP challenge resulted in marked loss of tyrosine hydroxylase (TH)-positive dopaminergic neurons in the striatum and SN, but DSF on its own did not cause any overt toxicity, consistent with previous studies (Fig. 6 B-F) (41)(42)(43).DSF treatment dramatically rescued loss of dopaminergic neurons in MPTP-challenged mice and the functional motor deficits in this model, as measured by the turnaround time and total time to descend in the pole test (Fig. 6 B-H) Thus DSF alleviates TLR4mediated pathogenesis in an aggressive PD model.
Discussion
This study identified DSF as a potent and selective inhibitor of TLR4-mediated inflammatory signaling.DSF acted by modifying a Cys residue (Cys133) in the TLR-binding partner MD-2, which blocked LPS sensing.We previously showed that DSF also inactivates GSDMD pore formation by binding to a critical Cys residue needed for pore assembly (31).Thus, the innate immune responses to both extracellular LPS and intracellular LPS, generated by intracellular bacterial pathogens or by uptake of outer membrane vesicles produced by gram-negative bacteria, are strongly inhibited by DSF.Moreover, in this study we also showed dramatic protection in a TLR4-dependent and LPS-independent mouse model of PD.
DSF and other Cys-reactive compounds have shown promise for treating both infection-induced and sterile inflammation-related disorders, including sepsis and autoimmune inflammation in mouse models (31,(44)(45)(46).In fact, inflammatory pathways are exquisitely sensitive to cellular redox status (20,(47)(48)(49).Inflammatory immune pathways are fine-tuned through active cysteines by reactive oxygen species and unpaired Cys's can act as sensors of cellular redox status.Many key inflammatory mediators or regulators, including STING, NLRP3, the inflammatory caspases, Cys-reactive compounds are notoriously nonspecific in their targets, although the targets that they modify that are critical for their therapeutic benefit can often be identified (i.e., alcohol dehydrogenase for DSF's use in alcoholism or as here TLR4 and GSDMD for preventing LPS toxicity).Although drug developers are reluctant to develop Cys-reactive drugs because of their lack of specificity, the approved Cys-reactive drugs such as DSF and DMF (used for multiple sclerosis) are remarkably well tolerated.The prominence of reactive Cys's in innate immune protein mediators suggests that it is worth seriously considering Cys-reactive compounds for suppressing inflammation in many settings including autoimmune and autoinflammatory diseases.
DSF and DMF were both previously shown to protect mice from LPS-induced sepsis (31,50,51).DSF not only protected mice from pyroptotic death, but also strongly inhibited serum levels of the inflammatory cytokines IL-1β, IL-6, and TNFα, cytokines whose expression is up-regulated by NF-κB.Extracellular LPS is sensed by TLR4, but intracellular LPS is sensed by the cytosolic noncanonical inflammasome (caspase-4 and -5 in humans and caspase-11 in mice), which when activated cleaves GSDMD to cause pyroptosis.Noncanonical inflammasome activation of pyroptosis dominates the pathology of LPS-mediated sepsis in mice since genetic deficiency of either Gsdmd or Casp11 protects mice from lethality (31).Protection from LPS-induced sepsis by DSF and DMF depended on their ability to inhibit cleaved GSDMD pore formation.However, the effectiveness of these drugs in sepsis likely depends on inhibiting both LPS sensing by TLR4 and GSDMD pores.
Similarly, although neither GSDMD nor caspase-1 deficiency affected DSF inhibition of inflammatory cytokine expression in response to LPS in vitro (SI Appendix, Fig. S3D), inhibiting NLRP3 and/or GSDMD pores may contribute to the strong protection provided by these drugs in the in vivo PD model reported here.In fact, NLRP3, caspase-1, and IL-1β are all activated in the MPTP model of PD and reduced in Tlr4 -/-mice (19,20).Pyroptosis could be a feature of neuronal or microglial cell death in PD.Future studies to examine inflammatory pathways in PD and whether microglia, which express TLR4 and GSDMD, and/ or neurons, which highly express GSDME, are undergoing pyroptosis, and their relative contributions to inflammatory pathology will be important for developing strategies to reduce inflammation as an approach to treating PD.It will also be important to understand what the TLR4/MD-2 complex senses in this PD model and whether TLR4/MD-2 plays a pathogenic role in human PD.
RNA-seq Preparation and Analysis.Biological triplicate RNA-seq libraries were prepared using the VAHTS mRNA-seq V3 library prep kit (Illumina) and sequenced on an Illumina Novaseq 6000 platform (Illumina).Image analysis and base calling were conducted using the Illumina RTA software.Sequencing reads were aligned to the Mus musculus reference genome (GRCm39) obtained from GENCODE.R package DESeq2 was used to analyze differential expression and fold change.The differentially expressed genes were subjected to heat map plotting with R package pheatmap.
Generation of Knockout Cell Lines by CRISPR-Cas9.HEK293T cells were transfected with lentivirus-packaging plasmids (PMD2.G/psPAX2) and lentiviral vector (lentiCRISPRV2-puro) carrying Cas9 and gRNAs (Gsdmd 5′-AGC ATC CTG GCA TTC CGA G-3′; Caspase-1 5′-ATA ATG AAT ACA ACC ACT CG-3′) by the calcium phosphate precipitation method as previously described (52).Lentivirus-containing medium was collected 72 h later and filtered with a 0.22µm Stericup filter unit (Millipore) to remove cell debris.iBMDMs were then infected with lentiviruses in the presence of 10 μg/mL polybrene when cells reached 50 to 60% confluency.Three days after infection, cells were treated with puromycin (5 μg/mL) for stable transfection selection.The surviving cells were further resuspended and diluted to seed as single and isolated cells.Gene editing was verified by genome sequencing and immunoblot analysis.
Immunostaining and Imaging.Cells seeded on coverslips were fixed with 4% paraformaldehyde for 15 min, permeabilized in 0.2% Triton X-100 for 5 min, and blocked with 5% BSA (Sigma Aldrich) diluted with PBS for 1 h.Then, cells were immunostained with an anti-NF-κB p65 antibody (CST #8242) for 2 h followed by an Alexa Fluor 488-conjugated secondary antibody (Invitrogen) for 1 h.Nuclei were counterstained with DAPI (Sigma Aldrich) for 5 min.Cells were extensively washed with PBS for three times between each step.Slides were then mounted using Aqua-Poly/Mount (Dako).Images were captured using an Olympus SpinSR10 Confocal System with 60X objective and Olympus cellSens Dimension software.All images are representative of three independent experiments.
Measurement of Cytokine Secretion.Concentrations of IL-6 and TNFα in cell culture medium were determined using mouse IL-6 (M6000B) and TNFα (MTA00B) Quantikine ELISA kits (R&D Systems) according to the manufacturer's instructions.
Recombinant Protein Purification.Escherichia coli (Rosetta) cells harboring pET28a-6 × His-SUMO-MD-2 were grown in LB broth with 50 μg/mL kanamycin at 37 °C.Overnight cells were diluted with fresh medium in a ratio of 1:100 and grown at 37 °C until the OD600 reached 0.8.IPTG was then added to cells at a final concentration of 0.2 mM to induce the expression of recombinant MD-2 at 16 °C.Cells were collected and homogenized by ultrasonication in lysis buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 5 mM imidazole and 2 mM β-ME).The clarified lysates obtained after centrifugation were incubated with equilibrated Ni-NTA resin (Qiagen) at 4 °C for 4 h and then extensively washed with wash buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 20 mM imidazole and 2 mM β-ME).Recombinant proteins were eluted with elution buffer (50 mM Tris-HCl pH 8.0, 400 mM NaCl and 500 mM imidazole) and treated with SUMO protease ULP1 (ubiquitin-like-specific protease 1) to remove the 6× His-SUMO tag, followed by further purification with Superdex 200 (10/300) gel-filtration chromatography and mono-Q ion exchange.Purified recombinant proteins were stored in (50 mM Tris pH 8.0, 150 mM NaCl and 5 mM DTT) at −80 °C for long-term use.
Liquid Chromatography-Mass Spectrometry (LC-MS) Analysis.Gel bands containing recombinant MD-2 treated or not with DSF were cut into pieces and washed with 50 mM Tris-HCl (pH 8.0)/acetonitrile (1:1, v/v) solution to destain the gel.The gel slices were incubated with 55 mM iodoacetamide for 60 min at room temperature in the dark, followed by treatment with 100% acetonitrile for 10 min.Trypsin digestion solution (15 ng/μL in 50 mM ammonium bicarbonate) was then added to dried gel pieces overnight at 37 °C for digestion.Tryptic peptides were extracted using extraction solution (5% TFA-50% ACN-45% ddH 2 O), acidified with 0.1% formic acid (FA), and subjected to LC-MS analysis using an Easy-nLC1200 High Performance Liquid Chromatography system and Q ExactiveTM™ Hybrid Quadrupole-Orbitrap™ Mass Spectrometer (Thermo Fisher Scientific).Data were analyzed using Byonic and MaxQuant1.6.2.10 software.
Flow Cytometry.For membrane TLR4-MD2 complex staining, single-cell suspensions of iBMDMs were incubated with the indicated fluorescent dyeconjugated antibodies [PE-conjugated rat IgG isotype control (RTK2758), PEconjugated anti-mouse TLR4/MD-2 complex (MTS510)] at 4 °C for 20 min in the dark and then washed three times with FACS buffer (PBS plus 2% FBS), followed by analysis on a BD Celesta cytometer using FlowJo software.For LPS internalization assay, iBMDMs pretreated or not with 20 μM DSF at 37 °C were washed and resuspended in FACS buffer.Single-cell suspensions were incubated with 1 μg/mL Alexa Fluor™ 488-conjugated LPS (Thermo-Fisher L23351) for 2 h at 4 °C and washed three times with FACS buffer, before analysis on a BD Celesta cytometer using FlowJo software.
Animals.WT C57BL/6 mice were maintained at the specific pathogen-free facility at Institut Pasteur of Shanghai (IPS) with a 12-h light/12-h dark cycle in a standard ambient environment (20 to 26 °C and 30 to 70% humidity with ad libitum access to food and water).All mouse experiments were conducted in accordance with the national animal research guidelines, using protocols approved by the Institutional Animal Care and Use Committee of IPS.
Drug Administration and MPTP-Induced Mouse PD Model.For DSF administration, mice were injected intraperitoneally with DSF formulated in sesame oil at a dose of 50 mg/kg on days 0, 2, 4, and 6.PD was induced in male C57BL/6 mice (8 to 10 wk old) by intraperitoneal injection of MPTP hydrochloride (Sigma) using a dose of 18 mg/kg every 2 h for four times on day 1 (43).All mice were subjected to the pole test and killed to harvest brain tissue on day 7.
Pole Test.Mice were placed head up on top of a rough surfaced pole (a bar 50 cm in height, 1 cm in diameter with a ball of 0.3 cm in diameter at the top).The time it took the mouse to completely turn downward was recorded as the turn-around time and the time it took to climb down from the pole was recorded as total time.Before the test, the mice were trained repeatedly until each could climb down from the pole immediately.Records were excluded if the mouse hesitated before climbing.
Histological Analysis.Fresh brain tissues were embedded with optimal cutting temperature compound (OCT) (Sakura Tissue-TekH) and snap-frozen in liquid nitrogen.Embedded tissues were equilibrated to −20 °C and sectioned using a microtome.Tissue sections were subjected to immunohistochemistry (IHC) by staining with monoclonal anti-tyrosine hydroxylase (TH) antibody (abcam ab137869) and the 3,3-diaminobenzidine (DAB) staining kit (Vector sk-4100) and hematoxylin staining according to the manufacturer's instructions before mounting using Aqua-Poly/Mount (Dako).Stained sections were imaged by light microscopy (Olympus BX53) with CellSens software.Densitometric analysis of the TH-positive signal was determined as the percentage of brown staining as measured by Image J software.
Data, Materials, and Software Availability.All study data are included in the article and/or SI Appendix.RNA-seq data are available in the Gene Expression Omnibus (accession number: GSE237028) (35).
Fig. 1 .
Fig. 1.DSF inhibits the LPS-triggered inflammatory response.(A) Mouse iBMDMs were pretreated or not with DSF (5 μM) for 0.5 h before stimulation with LPS (1 μg/mL) for the indicated times.mRNA levels of the indicated inflammatory cytokines were assessed by qRT-PCR, normalized to Gapdh and relative to unstimulated cells.(B) Mouse iBMDMs were pretreated with the indicated concentrations of DSF for 0.5 h before stimulation with LPS (1 μg/mL).mRNA levels of the indicated inflammatory cytokines were assessed by qRT-PCR, normalized to Gapdh and relative to unstimulated cells.(C and D) RNASeq analysis of RNA expression in mouse iBMDMs treated or not with LPS in the presence or absence of DSF (5 μM).(C) Differential gene expression after treatment with LPS in the presence vs. absence of DSF (Y axis) or LPS treatment vs. untreated controls (X axis).The dotted lines represent log (fold change) = 1.(D) Heatmap of genes filtered by adjusted P value < 0.05 and the absolute log2 fold change > 2 after treatment with LPS compared to untreated control.Graphs in A and B show mean ± SD; data are representative of three independent experiments.Data were analyzed using two-way ANOVA.***P < 0.001.
Fig. 2 .
Fig. 2. DSF inhibition is markedly enhanced in the presence of Cu(II).(A) DSF is rapidly metabolized to DTC, which forms a complex with Cu(II).(B and C) Doseresponse curves of inhibition of TLR4-induced proinflammatory cytokine expression by DSF in the absence or presence of Cu(II) (B) or Mg(II) (C) in iBMDMs.Graphs show mean ± SD; data are representative of three independent experiments.
Fig. 3 .
Fig. 3. DSF targets the TLR4 signalosome.(A) RAW 264.7 cells transfected with Firefly 5×κB-luciferase and pTK-Renilla reporters were pretreated or not with LPS in the absence or presence of DSF and analyzed by dual luciferase reporter assay.(B) Schematic diagram of TLR4 signaling pathway.(C) HEK293T cells, transfected with 5×κB-luciferase and pTK-Renilla reporters together with the indicated expression plasmids for 6 h, were treated or not with DSF (20 μM) and analyzed by dual luciferase assay 18 h later.(D) iBMDMs, pretreated with DSF (5 μM) or DMSO for 0.5 h before stimulation with LPS (1 μg/mL), were harvested at the indicated time points for immunoblotting probed for IKKβ, IκBα and NF-κB p65 phosphorylation and IL-1β induction.(E) MEF cells, pretreated or not with DSF (5 μM) for 0.5 h, were stimulated with LPS (1 μg/mL) for another 0.5 h, before staining for NF-κB p65.Representative immunofluorescence images of NF-κB p65 subcellular localization are shown.Graphs in A and C show mean ± SD; data are representative of three independent experiments.Data were analyzed using a two-tailed Student's t test.**P < 0.01.
Fig. 4 .
Fig. 4. DSF modifies Cys133 of MD-2.(A and B) iBMDMs were pretreated with the indicated doses of DSF in the presence or absence of Tiron (A) or N-acetyl-Lcysteine (NAC) (B) for 0.5 h before stimulation or not with LPS (1 μg/mL) for 4 h.mRNA levels of the indicated inflammatory cytokines were assessed by qRT-PCR, normalized to Gapdh and relative to unstimulated cells.(A) Il1b, Tiron 1 mM vs. 0 mM P = 0.0005, 2 mM vs. 0 mM P = 0.0029; Il6, Tiron 1 mM vs. 0 mM P = 0.0012, 2 mM vs. 0 mM P < 0.0001, Il12b, Tiron 1mM vs 0 mM P <0.0001, 2 mM vs. 0 mM P < 0.0001; (B) Il1b, NAC 1 mM vs. 0 mM P = 0.0105, 2 mM vs. 0 mM P = 0.0012; Il6, NAC 1 mM vs. 0 mM P < 0.0001, 2 mM vs. 0 mM P < 0.0001, Il12b, NAC 1 mM vs. 0 mM P < 0.001, 2 mM vs. 0 mM P < 0.0001.(C) Recombinant mouse MD-2 protein was incubated with DSF or DMSO for 1 h before mass spectrometry analysis.Cys133 on MD-2 modification by DSF is highlighted.The mass-tocharge ratio (m/z), nuclear charge (z), and theoretical masses of indicated peptides are shown in red.The spectra at the bottom plot relative abundance vs. m/z ratio.(D) Quantitative analysis of internalization of Alexa Fluor™ 488-labeled LPS by iBMDMs pretreated or not with DSF using flow cytometry.(E) Representative flow cytometry histograms showing TLR4/MD-2 complex expression on the surface of iBMDM cells treated or not with DSF.(F) iBMDMs, pretreated with DSF for 0.5 h and extensively washed to remove DSF, were treated with increasing concentrations of recombinant MD-2 and stimulated with LPS for 4 h before measuring inflammatory cytokine mRNA levels by qRT-PCR.mRNA level was normalized to Gapdh and relative to unstimulated cells.Graphs in A, B, and F show mean ± SD; data are representative of three independent experiments.Data were analyzed using two-way ANOVA (A and B) or a two-tailed Student's t test (F).*P < 0.05; **P < 0.01; ***P < 0.001.
Fig. 5 .
Fig. 5.Other cysteine-reactive compounds, afatinb and DMF, inhibit TLR4 signaling.(A and B) iBMDMs were pretreated with afatinib, DMF, or DSF for 0.5 h before stimulation with LPS (1 μg/mL) for 4 h.mRNA levels of the indicated inflammatory cytokines were assessed by qRT-PCR, normalized to Gapdh and relative to unstimulated cells.(C and D) iBMDMs were pretreated or not with afatinib (10 μM) (C) or DMF (20 μM) (D) for 0.5 h before stimulation with LPS (1 μg/mL).Whole-cell lysates were harvested at the indicated time points for immunoblot analysis.(E and F) iBMDMs, pretreated with afatinib (10 μM) or DMF (20 μM) in the presence or absence of NAC for 0.5 h, were stimulated with LPS (1 μg/mL).mRNA levels of the indicated inflammatory cytokines were assessed by qRT-PCR, normalized to Gapdh.Graphs in A, B, E, and F show mean ± SD; data are representative of three independent experiments.Data were analyzed using a twotailed Student's t test.***P < 0.001.
Fig. 6 .
Fig.6.DSF alleviates MPTP-induced PD. (A-H) Mice (n = 3 mice/gp) were injected intraperitoneally with DSF or vehicle 24 h before and every other day after intraperitoneal challenge with MPTP hydrochloride, followed by histopathological (B-E) and immunoblot (F) analysis of tyrosine hydroxylase (TH) in striatum and substantia nigra biopsies from mice brain tissues or motor ability test (G and H) on day 7. Quantification of TH-positive neurons was performed by optical density measurement using ImageJ software (C and E).Graphs in C, E, G, and H show mean ± SD; data are representative of three independent experiments.Data were analyzed using a two-tailed Student's t test.*P < 0.05; **P < 0.01; ***P < 0.001.
|
2023-07-26T06:15:59.440Z
|
2023-07-24T00:00:00.000
|
{
"year": 2023,
"sha1": "b17f36b4a2d21cd2a2740b268862e41d4a65f9c3",
"oa_license": "CCBYNCND",
"oa_url": "https://doi.org/10.1073/pnas.2306399120",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "9af1284b7f1857855b634490eaf8e212d1e491e5",
"s2fieldsofstudy": [
"Medicine",
"Chemistry"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
253259422
|
pes2o/s2orc
|
v3-fos-license
|
Remnant cholesterol and the risk of cardiovascular disease in type 2 diabetes: a nationwide longitudinal cohort study
Background Elevated remnant cholesterol (remnant-C) is considered a risk factor for cardiovascular disease (CVD); however, whether this notion applies to the East Asian population with type 2 diabetes (T2D) has not been established. This study investigated the association between remnant-C concentrations and the risk of CVD in Korean patients with T2D. Methods By using the Korean National Health Insurance Service database, 1,956,452 patients with T2D and without atherosclerotic CVD who underwent regular health checks between 2009 and 2012 were included. Cox regression analyses were conducted to assess the association between remnant-C concentrations and incident CVD comprising myocardial infarction (MI) and ischemic stroke. Results In total, 50,120 (2.56%) cases of MI and 73,231 (3.74%) cases of ischemic strokes occurred during a median follow-up of 8.1 years. The adjusted hazard ratios for MI and stroke in the highest remnant-C quartile were 1.281 (95% confidence interval [CIs], 1.249–1.314) for MI and 1.22 (1.195–1.247) for ischemic stroke, compared to those in the lowest quartiles. The results were similar, based on stratified analysis by age, sex, use of statin or fibrate, and levels of other cholesterol. The increased risk of CVD in the highest remnant-C quartile was profound in patients who had a longer T2D duration. A remnant-C concentration ≥ 30 mg/dL differentiated patients who were at a higher risk of CVD, compared to patients with a lower concentrations, regardless of whether LDL-C levels were or were not on target at ≤ 100 mg/dL. Conclusion In Korean patients with T2D, remnant-C was associated with CVD, independent of the LDL-C level or other conventional CVD risk factors. Our finding confirmed evidence of the causal role of remnant-C on CVD, as a residual risk of CVD, in East Asian patients with T2D. Supplementary Information The online version contains supplementary material available at 10.1186/s12933-022-01667-6.
Background
Atherosclerotic cardiovascular disease (ASCVD) remains a leading cause of morbidity and mortality for patients with type 2 diabetes (T2D) [1]. However, only few people with T2D have optimal risk factor control and many patients with T2D still have a high residual risk [2]. Therefore, numerous efforts have been made to identify these residual risks for cardiovascular disease (CVD) in patients with T2D.
Remnant cholesterol (remnant-C) is the cholesterol content of TRLs [8]. TRLs are more abundant, larger, and carry more cholesterol than do LDLs in atherogenic dyslipidemia. Previous studies have shown that remnant-C was associated with the risk of myocardial infarction (MI), stroke, and all-cause mortality in the general population [9][10][11]. A recent study demonstrated that remnant-C predicts ASCVD beyond low-density lipoprotein cholesterol (LDL-C) levels, but the investigators suggested that TG or remnant cholesterol-associated ASCVD risk is proportional to LDL-C [9]. Limited information about this issue is available among patients with only T2D who have a higher chance of atherogenic dyslipidemia. Therefore, the aim of the present study was to evaluate the association between remnant-C and the risk of CVD in T2D patients without a known history of CVD in a large pooled primary prevention cohort using the National Health Insurance Service (NHIS) health checkup data. We also investigated the risk of CVD associated with remnant-C concentrations, independent of LDL-C concentrations.
The NHIS database and NHIS Health Checkup Program
We used data from the National Health Insurance Service-Health Screening Cohort (NHIS-HEALS), which is provided by the Republic of Korea. In Korea, 97% of the population is obliged to enroll in this program [12]. The NHIS manages a biennial health checkup program for all insured South Koreans > 40 years of age and an annual checkup that is recommended for employee subscribers who are ≥ 20 years of age. The NHIS health checkup programs include anthropometric measurements; hearing and visual acuity checks; laboratory tests; past family, medical, and surgical history; and social history. Hospitals perform health checkups after being certified by the NHIS, which also regularly qualifies trained examiners. The NHIS-HEALS collects patients' demographic data, such as region, age, sex, medical utilization/transaction information, claims and deduction data, and insurers' payment coverage. The NHIS database has been described in detail in previous studies [13].
Study population
In the NHIS Health Checkup database from 2002 to 2018, we selected adults aged 20 years and older who had a diagnosis of T2D and underwent health examinations from 2009 to 2012 (n = 2,746,079). We followed up with the individuals until the date of incident CVD or until December 31, 2018. We excluded people who is younger than 20 years old (n = 441) and who had received a diagnosis of MI or ischemic stroke before the index date of January 1, 2013 (n = 382,404). To minimize the influence of possible "reverse causation, " we excluded individuals who were developed MI, stroke, or CV deaths within 1 year after the baseline measurements (n = 35,063). Then, we also excluded patients with a TG level ≥ 300 mg/dL (n = 287,482) and who had any missing variables for measurement of remnant cholesterol (n = 84,237). Finally, 1,956,452 total participants were included in this study. (men = 1,159,932 and women = 796,520), and the median observational time was 8.15 (7.02-9.08) years ( Figure S1). This study was approved by the Institutional Review Board of the Hallym University Sacred Heart Hospital (An-yang Si, South Korea; IRB No. HALLYM 2020-05-001), and permission was granted to use the NHIS Health Checkup data (NHIS-REQ000042107-001). The requirement for written informed consent was waived in the present study because data in the NHIS database are anonymized in adherence to strict confidentiality guidelines.
Data collection
Detailed information on the individuals' demographics and lifestyles was obtained through standardized selfreporting questionnaires. Income level was dichotomized at the lower 20%. Smoking status was classified as "nonsmoker, " "ex-smoker, " or "current smoker. " Alcohol drinking was categorized as 0 g/day ("none"), 30 g/day ("mild"), or ≥ 30 g/day ("heavy"). Regular exercise was defined as vigorous-intensity physical activity, performed at least three times per week, or as moderate-intensity physical activity, performed at least five times per week [14]. The health examinations provided by NHIS include anthropometric and laboratory measurements. Height, weight, and waist circumference (WC) were measured, and body mass index (BMI) was calculated by dividing weight (kg) by height (m) squared.
Measurement of the lipid profile
Blood samples for the measurement of the lipid profile and glucose were drawn after an overnight fast. Lipid profiles, which included the levels of total cholesterol, LDL-C, high-density lipoprotein cholesterol (HDL-C), and TGs, were measured by using an enzymatic method. Quality control of laboratory tests was conducted by using the procedures of the Korean Association of Laboratory Quality Control [15]. Remnant-C level was estimated as the total cholesterol level minus the LDL-C level minus the HDL-C level. No established method exists to measure the remnant-C level, although previous studies have frequently used this equation because remnant-C levels can be easily obtained from the standard lipid profile [8]. The non-HDL-C level was calculated as the total cholesterol level minus the HDL-C level [10].
Definition of comorbidities
All participants were categorized into three groups, based on their glycemic status: newly diagnosed T2D, T2D < 5 years, and T2D ≥ 5 years [16]. Patients with newly diagnosed T2D were diagnosed with T2D at the time of national health examinations in 2009-2012. Newly diagnosed T2D was defined as a fasting plasma glucose level ≥ 126 mg/dL during health examinations or at least one claim per year for the prescription of hypoglycemic drugs under International Classification of Disease (ICD-10) codes E11-14 in an outpatient or inpatient setting and a prescription for at least one antidiabetic drug at any time over 1 year to exclude prediabetic or nondiabetic individuals. Patients with T2D < 5 years were diagnosed with T2D within 5 years of the date of the health checkup in 2009-2012. Thus, these patients had been newly diagnosed with T2D from 2004 to 2008. Patients with T2D ≥ 5 years were diagnosed with T2D 5 years before 2009. Thus, these patients had been newly diagnosed with T2D before 2003.
Hypertension was defined as a blood pressure ≥ 140/90 mmHg or at least one claim per year for antihypertensive medication prescription under the International Classification of Disease, 10th Revision (ICD-10) codes I10-I15. "Dyslipidemia" was defined as a total cholesterol level ≥ 240 mg/dL or at least one claim per year for the prescription of lipid-lowering agents under ICD-10 code E78 [16]. "Chronic kidney disease" was defined as a glomerular filtration rate of < 60 mL/min/1.73 m 2 and was based on a combination of ICD-10 codes (i.e., N18-19, Z49, Z94.0, and Z99.2). "Metabolic syndrome" was defined, based on the modified criteria of the National Cholesterol Education Program Adult Treatment Panel III (NCEP-ATP III) [17]. "Obesity" was defined as a BMI ≥ 25 kg/m 2 , using the Asia-Pacific criteria of the World Health Organization guidelines [17]. We defined a statin user as a person who had been prescribed statins during 2009-2012 and a fibrate user as a person who had been prescribed fibrates during 2009-2012.
Outcome ascertainment
The primary endpoints of the study were newly developed MI or stroke. The diagnosis was based on the ICD-10 codes. MI was determined, based on the recording of ICD-10 code I21 or I22 during hospitalization for ≥ 4 days or the recording of these codes at least twice [18]. Ischemic stroke was determined, based on the recording of the ICD-10 code I63 or I64 during hospitalization for ≥ 4 days with claims for brain magnetic resonance imaging or brain computerized tomography [13]. The study participants were censored at date of the occurrence of MI or stroke, or end of follow-up (December 31, 2018), whichever came first.
Statistical analyses
The baseline characteristics of the participants are expressed as the mean ± the standard deviation for continuous variables and as the number (percentage) for categorical variables. Values between groups were compared using the independent t-test for continuous variables and the chi-squared test for categorical variables. To compare each group, we performed one-way analyses of variance (ANOVA), and Chi square test, as appropriate. The incidence rates (IRs) of MI and ischemic stroke were calculated by dividing the number of events by 1000 person-years. Cox proportional hazard analysis was used to evaluate the association of remnant-C with MI and ischemic stroke and to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs). Multivariable fully adjusted Cox proportional hazards models were applied. We assessed PH assumption using Schoenfeld residuals analysis and Log-log plot in our cox proportional hazard model. In the present study, subgroup analyses were also conducted using multivariable Cox proportional hazard models stratified by potential modifiable clinical factors through stratified analysis and interaction testing using the likelihood ratio test. Values of P < 0.05 were statistically significant. Statistical analysis was conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA) and R 3.1.0 (R Foundation for Statistical Computing, Vienna, Austria).
Baseline characteristics of the study population
The baseline characteristics of the study participants are presented in remnant-C quartiles in Table 1. Participants in the higher remnant-C quartiles were more likely to be male and younger than participants in the lower remnant-C quartiles, and they had worse metabolic traits. They had a higher proportion of current smokers and heavy alcohol drinkers, and they were more likely to exercise less regularly. Hypertension and dyslipidemia were more prevalent in patients in the higher remnant-C quartiles. Participants in the higher remnant-C quartiles had a low prevalence of longer duration of T2D (i.e., ≥ 5 years) and a lower proportion of insulin users.
Baseline lipid concentrations and major cardiovascular events
During the median (interquartile range) follow-up duration of 8.15 (7.02-9.08) years, 123,351 individuals experienced the new development of MI or stroke (50,120 cases of MI and 73,231 cases of stroke). The association between lipid concentrations and risk of MI and stroke is presented in Table S1. The serum concentrations of LDL-C, TG, and non-HDL-C were associated with a 1.4%, 1.5%, and 1.6% higher risk of MI per every 10 mg/ dL increase, respectively. The serum concentrations of HDL-C were associated with a 2% lower risk of MI per every 5 mg/dL increase, whereas the concentrations of remnant-C was associated with a 3.4% higher risk of MI per 10 mg/dL increase. The serum concentrations of LDL-C, TG, and non-HDL-C were likewise associated with a 1.1%, 1.2%, and 1.2% higher risk of stroke per every 10 mg/dL increase, respectively. The serum concentration of HDL-C was associated with a 1.4% lower risk of stroke per every 5 mg/dL increase and the concentration of remnant-C was associated with a 3.2% higher risk of stroke per 10 mg/dL increase. We further examined the association between the remnant-C quartile and the risk of MI and ischemic stroke ( Table 2). The
Association between remnant-C level and the risk of CVD, based on the duration of diabetes
We examined the risk of MI and stroke, based on the remnant-C quartiles, stratified by the duration of T2D (Fig. 1). Among patients who had newly developed T2D, the IRs of MI and stroke tended to be incrementally higher in the upper remnant-C quartiles than in the lowest quartile However, IRs of stroke were highest in the third quartile of the newly developed T2D group. However, in the groups with T2D < 5 years or T2D ≥ 5 years, the IRs of MI and stroke were incrementally higher in the upper remnant-C quartiles than in the lowest quartile, and it increased with the progression of diabetes. This linear trend between remnant-C levels and incident CVD was more prominent in patients with a longer duration of T2D (i.e., ≥ 5 years). The risks of MI and stroke gradually increased with the progression of the diabetes status and increased remnant-C concentrations, using newly developed T2D with the lowest remnant-C quartile as the reference (for the fully adjusted HRs of MI in the highest remnant-C quartile: HR = 1.356 for newly diagnosed T2D, HR = 1.515 for T2D < 5 years, and HR = 2.085 for T2D ≥ 5 years; for the fully adjusted HRs of stroke in the highest remnant-C quartile: HR = 1.238 for newly diagnosed T2D, HR = 1.485 for T2D < 5 years, and HR = 2.016 for T2D ≥ 5 years). It indicates that a higher risk of CVD in higher remnant-C concentration was more profound in patients with a longer duration of T2D.
Risk of CVD, based on the remnant-C quartile in the various subgroups
We conducted subgroup analyses, stratified by age, sex, obesity status, absence or presence of hypertension, chronic kidney disease, metabolic syndrome, hypertriglyceridemia (TG, ≥ 150 mg/dL), low HDL-C (< 40 mg/dL in men and < 50 mg/dL in women), and use of statin or fibrate drugs ( Fig. 2 and Table S2). The higher risk of MI and stroke in the upper remnant-C quartiles was sustained consistently across all subgroups of patients with T2D. The higher risk of MI and stroke in the upper remnant-C quartiles persisted notably, regardless of the TG or HDL-C level and the use of medication for dyslipidemia. The association between remnant-C level and MI risk was stronger in younger (i.e., < 65 years) patients with T2D, and in patients without hypertension. The association between the remnant-C level and ischemic stroke risk was stronger in women, statin users, and in patients without obesity or abdominal obesity.
Contribution of the remnant-C level to the development of CVD, based on the LDL-C level
Based on definitions used in a previous study [10], we categorized our patients into four groups, using the combinations of each remnant-C and LDL-C cutoff level (Table 3). We found that 446,210 patients with T2D (i.e., 22.8% of the entire cohort) were at the target for both lipid values. However, 14.5% of patients were classified with a high remnant-C level and a target level of LDL-C. Even in patients with controlled LDL-C levels,
Discussion
In this nationwide population-based cohort study, we investigated the association between remnant-C concentrations and the risk of CVD in Korean patients with T2D, who were followed for more than approximately 8.15 years, by using data from the Korean National Health Insurance Service database. We noted the following findings: (1) the remnant-C level was significantly associated with CVD events consisting of MI and stroke in T2D, regardless of whether lipid-lowering agents were or were not used; (2) elevated remnant-C levels were associated with the risk of incident CVD, independent of traditional CVD risk factors, including LDL-C concentrations, in patients with T2D; in addition, elevated remnant-C levels increased the risk of CVD, even in patients who were at the target range of LDL-C concentrations for primary prevention in T2D; and (3) the linear association between elevated remnant-C levels and the higher risk of ASCVD was more prominent in patients with a more advanced stage of T2D. These findings indicated that the remnant-C level may be a significant residual risk factor for CVD in East Asian patients with T2D. The increased CVD risk associated with a high remnant-C level in T2D is due, in part, to diabetes-related atherogenic dyslipidemia, which is characterized by high circulating TG and/or cholesterol content within TRLs and low concentrations of HDL-C with relatively normal concentrations of LDL-C. In particular, the relationship between hypertriglyceridemia and the risk of CVD in patients with T2D remains inconclusive [19]. TG, per se, is unlikely to directly cause CVD [20]; however, Nordestgaard et al. [5] demonstrated that the harmful component in TRLs is cholesterol, not TG itself. Furthermore, a recent systematic review demonstrated that, in patients with T2D, an elevated TG concentration is not an independent marker for an increased risk of ASCVD events [21], and increasing HDL-C levels fail to reduce ASCVD [6,7]. These findings suggest that TG is a just marker of elevated remnant-C levels. Therefore, an estimation of remnant-C level may be more important than the TG or HDL-C level itself in patients with T2D [3]. However, Fig. 1 Incident rate of myocardial infarction (A) and stroke (B) per 1000 person-years (bar) and adjusted hazard ratios of myocardial infarction and stroke (line) by quartiles of remnant cholesterol according to diabetes status. * adjusted for age, sex, body mass index, smoking status, alcohol drinking status, regular exercise, low income, hypertension, statin treatment, and fibrate use T2D, type 2 diabetes; HR, hazard ratio; IR. incident rate; PYRS, person years few data exist on the role of remnant-C in the primary prevention of CVD, especially in patients with T2D. Therefore, this study has clinical implications showing that elevated remnant-C concentrations may be an independent residual risk factor for patients with T2D, which may provide new information on the necessity of monitoring remnant-C for the primary prevention of ASCVD in T2D.
The strong association between remnant-C concentrations and CVD in T2D patients can be explained by the following mechanisms. Cholesterol is the major lipid that accumulates in atherosclerotic lesions. Remnant-C has a greater cholesterol content per particle than does LDL-C, perhaps making remnant-C even more atherogenic than LDL-C [2]. It also passes into the arterial wall and is taken up by macrophages and smooth muscle cells without modification, unlike LDL, which needs to be modified before uptake. However, in diabetes the catabolism of chylomicron-derived remnant-C particles is decreased, thereby resulting in the accumulation of remnant-C particles in plasma [22]. Some early studies in which radiolabeled chylomicrons were injected in diabetic rats, demonstrated the accumulation of cholesteryl ester, but not TG components, in plasma, which suggested that chylomicron-derived remnant-C accumulates in diabetes [22]. Another study similarly showed that the retention of chylomicron remnants by the arterial intima was associated with hyperglycemia in diabetic rabbits and rats [23]. Hence, the atherogenic effects of remnant-C in elevated glucose levels may explain the associations between remnant-C and CVD, especially in the advanced stage of T2D, as demonstrated in the present study. We demonstrated that remnant-C was the major cholesterol fraction contributor to the development of MI and stroke in patients with T2D without a history of ASCVD. Participants in our study had nearly normal or slightly increased TG concentrations (TG < 300 mg/dL), and a considerable percentage of patients (approximately 25%) were taking statin drugs. We found that individuals with high remnant-C levels (> 30 mg/dL) had a greater risk of MI and ischemic stroke, regardless of whether their LDL-C level was or was not at optimal levels (< 100 mg/dL). Based on the current findings, increased concentrations of remnant-C contribute to atherogenic risk, even in patients with T2D, independent of TG or LDL-C concentrations and statin use. Moreover, an implication is that the treatment of residual risk, based on measurements of the remnant-C concentration, may be more beneficial than further reducing LDL-C concentrations in patients with T2D for primary prevention if they are already at the target LDL-C concentration by taking a statin drug. This finding is consistent with findings from previous studies which was conducted in specific ethnic and disease groups on a small scale. Castaner et al. [10] reported that high remnant-C levels among obese and overweight patients were associated with a 2.6 times higher risk of ASCVD, although LDL-C levels being in the optimal range, Cao et al. [30] also demonstrated that remnant-C was associated with a 62% higher risk of ASCVD in T2D patients with a history of coronary artery disease. The absolute values of HRs in the present study seemed to be relatively lower than the values reported in previous studies. This finding could be explained by the fact that our cohort comprised a primary prevention cohort and other clinical conditions and confounding factors associated with T2D may attenuate the association between remnant-C and CVD. Considered together, we suggest that our findings provide more concrete evidence that measuring plasma remnant-C levels may be clinically relevant in primary prevention to identify T2D patients who are at high risk of CVD.
By using subgroup analysis, we found that higher remnant-C levels were consistently associated with an increased risk of MI and stroke in all clinical subgroups. The risk of CVD associated with remnant-C was more prominent in patients who had a relatively lower CVD risk such as patients with a younger age, absence of hypertension, and women. The risk of CVD associated with remnant-C was more prominent in patients who had a relatively lower CVD risk such as patients with a younger age, absence of hypertension, and women. We are not able to explain why individuals at relatively lower risk of CVD are more susceptible to elevated remnant-C levels. We assume that as individuals with multiple CVD risk factors would be influenced by other strong CVD risk factors rather than remnant-C. For example, because the proportion of men who currently smoke was higher than women, the independent risk of remnant-C on CVD in men would be attenuated by smoking.
Of interest, the association between the remnant-C level and CVD risk was more prominent in statin-treated patients. This finding indicated that, even when treated with statins, patients with high remnant-C levels were at high risk of CVD. European dyslipidemia guidelines recommend lowering the LDL-C and apolipoprotein B (apo B) levels in high-risk patients such as patients with T2D [24]. Another recent study [25] has also shown that elevated non-HDL cholesterol and apo B levels, but not LDL-C levels, are associated with the residual risk of MI and overall mortality in statin users. Given the 1:1 association between apoB and atherogenic lipoprotein particles, apoB concentrations represent the number of the lipoprotein particles. While, remnant-C represents cholesterol contents carried by the apolipoprotein B-containing lipoprotein particles. Therefore, these two factors are closely associated and both of two factors can be used to assess CV risk in individuals. Our findings confirmed the evidence supporting that the remnant-C level explains the residual risk of ASCVD in patients with T2D. We also suggest that the remnant-C level should be considered for use as an additional prognostic lipid measure, along with LDL-C in patients with T2D, even if they were already taking antidyslipidemic agents.
The higher CVD risk with higher remnant-C concentrations was interestingly more profound in patients with a more advanced stage of T2D in this study. A longer duration of diabetes increases the risk of ASCVD or CV deaths in patients with T2D [26]. Furthermore, another study [27] demonstrated that high plasma residual cholesterol was overproduced in the insulin-resistant status, and it may have a key role in the etiology of coronary artery disease in patients with diabetes. Several in vitro and in vivo studies [28] have also suggested that the macrophage uptake of remnant-C in diabetic patients was correlated with the degree of glycemic control in patients with diabetes. We assume that factors associated with an advanced stage of T2D may act as mediating factors that enhance intimal permeability or the capture of remnant-C within the arterial intima in the process of atherosclerosis [29]. These facts indicate that remnant-C is additionally interacting with a worsening glycemic status and combining effect of these two factors consequently have a stronger atherogenic potential in patients with T2D. Therefore, evaluating the combined effect of high remnant-C levels and a worsening glycemic status in T2D may provide new insight into cardiovascular events and metabolic risk estimation.
This study has several strengths. First, it is the first study to examine the association between remnant-C and MI or stroke, based on the duration of T2D status, in the Korean population. Second, this study had a large sample size, consisting of > 2,500,000 patients with T2D, and used a well-validated national database. However, several limitations of this study should be addressed. First, potential known confounding variables were adjusted for in multivariable Cox regression analysis; however, we could not exclude all possible residual bias because we did not collect all data about metabolic parameters related to CVD or lifestyle factors. Moreover, because NHIS-HEALS did not measure glycated hemoglobin, a well validated surrogate marker for the degree of glycemic control, we were unable to assess the association between remnant-C and CVD stratified by glycemic control status based on glycated hemoglobin. The remnant-C level calculated in the current study may have over-or under-estimated its value, compared to direct measurements. However, the calculation of the remnant-C level is an affordable method that could provide valuable data for clinical management. Therefore, many recent studies [30] have measured remnant-C by a calculation, as in our study. Third, we did not consider the effects of antidiabetic medications, which might have potential effects on the development of CVD ((e.g. sodium-glucose cotransporter-2 [SGLT-2] inhibitors or glucagon like peptide-1 receptor agonists (GLP-1 RA)). However, these kinds of anti-diabetic medications are not widely availabe in Korea due to reimbursement restriction [31]. Finally, the present study was observational, and the causal role of remnant-C on ASCVD risk should be determined in further studies.
Conclusion
This large-scale cohort study, which had a long-term follow-up, showed that remnant-C was a risk factor for MI and stroke in patients with T2D without a history of ASCVD, independent of traditional risk factors and serum LDL-C concentrations. Furthermore, we demonstrated that the higher CVD risk in patients with a higher remnant-C concentration was more profound in patients with a longer duration of T2D. This finding indicated that determining remnant-C levels in patients with T2D would be useful in terms of cardiovascular risk stratification and future interventions.
|
2022-11-03T18:13:38.686Z
|
2022-11-02T00:00:00.000
|
{
"year": 2022,
"sha1": "7dbd0c94b92e677b972231dd7c8f44efca41a307",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Springer",
"pdf_hash": "0b27eeaa555451f56e063e21c870cff014d5ac3e",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
86740992
|
pes2o/s2orc
|
v3-fos-license
|
Factor Xa Inhibitory Profile of Apixaban, Betrixaban, Edoxaban, and Rivaroxaban Does Not Fully Reflect Their Biologic Spectrum
The currently available oral anti-Xa agents are claimed to produce their anticoagulant and antithrombotic effects solely by the inhibition of factor Xa. This study profiled various anti-Xa drugs in routinely used laboratory assays to demonstrate that their effects are not solely related to the anti-Xa activities. Apixaban, betrixaban, edoxaban, and rivaroxaban were obtained commercially. Native and citrated whole blood was used for the activated clotting time (ACT) and thromboelastography (TEG). Citrated plasma was used for monitoring the prothrombin time (PT), activated partial thromboplastin time (aPTT), Heptest, and prothrombinase-induced clotting time (PiCT) tests. An amidolytic method was used for the determination of anti-Xa effects. Thrombin-induced fibrinokinetics was monitored optically. Thrombin generation studies were carried out using the calibrated automated thrombogram. All of the anti-Xa agents produced concentration- and assay-dependent effects. In the ACT at 2.5 μg/mL and TEG at 1.0 μg/mL, edoxaban exhibited the strongest anticoagulation effect. In the PiCT, PT, and aPTT assay at 1 μg/mL, edoxaban showed stronger effects than other agents. The half maximal inhibitory concentration of these agents for the inhibition of factor Xa ranged from 340 to >1000 ng/mL. In the thrombin generation inhibition assay, apixaban showed the strongest activity. In the fibrinokinetics, different anti-Xa agents produced varying degrees of inhibition. These results demonstrate that the measured anti-Xa activity alone does not fully reflect the overall biologic spectrum of these agents.
Introduction
The introduction of direct oral anticoagulants (DOACs) has added a new dimension to the management of thrombotic and cardiovascular disorders. 1 Owing to better safety and efficacy, these drugs are widely used in agent-specific indications, which include venous thromboembolism and nonvalvular atrial fibrillation. 2,3 These drugs are currently being developed for other indications, including acute coronary syndrome, peripheral arterial disease, and cancer-associated thrombosis. Currently, there is one oral antithrombin agent, namely, dabigatran, and four anti-Xa agents, namely, apixaban, betrixaban, edoxaban, and rivaroxaban, approved for specific indications. Although these drugs are orally administered, their dosages for different indications vary widely. The anti-IIa drug dabigatran can be readily differentiated from the currently available anti-Xa agents. Differences in the currently available anti-Xa agents are also notable in terms of their inhibitory profile in factor Xa and thrombin generations assays.
Both the chemical and the enzymatic methods are used to synthesize new oral anticoagulant agents. Minor microchemical differences in their structure contribute to their pharmacological identity. 4 Table 1 lists the currently available oral anti-Xa drugs. All of these exhibit similar characteristics in term of molecular weight (400-600 Da). All these agents mediate their antithrombotic actions via inhibiting factor Xa. The relative inhibitory potency of each of these agents differs. The dosage of these agents also varies widely and ranges from 2.5 to 160 mg/d with certain adjustments. Apixaban is used at a relatively lower dosage, whereas betrixaban is used at a relatively higher dosage. Moreover, the pharmacokinetic and pharmacodynamic profiles of these agents widely differ. Each of these agents is approved for specific indications at a fixed dosing regimen. A direct comparison of these agents in terms of their biochemical and pharmacological properties is not available. The therapeutic effects of these agents are mainly attributed to their inhibitory effects on factor Xa. As these drugs represent synthetic heterocyclic agents, it may be possible that besides inhibiting factor Xa, these agents may have additional effects on the coagulation network. In this study, the effects of these drugs on various laboratory assays used to assess the anticoagulant effects of drugs were studied. It is hypothesized that the overall anticoagulant and antiprotease effects of these drugs may not be directly proportional to their anti-Xa potency.
Materials and Method Whole Blood Analysis
Thromboelastographic (TEG) studies were carried out using TEG 5000 Hemostasis System (Haemonetics Corp, Massachusetts). Blood samples were drawn from healthy donors (n ¼ 5-7) and placed into tubes containing 3.2% sodium citrate. To the TEG cup, CaCl 2 , the Xa inhibitor at a final concentration of 1.0 mg/mL, and citrated human whole blood were added. The TEG profile was monitored for 45 minutes. The parameters such as r-time, k-time, maximum amplitude (MA), and angle were measured.
Whole blood activated clotting time (ACT) was measured by using Hemochron, Whole Blood Coagulation System (Accriva Diagnostics, Inc, California). Whole blood was supplemented with the factor Xa inhibitors at a final concentration of 2.5 mg/mL and transferred into celite ACT tubes. Results were recorded in seconds.
Clot-Based Assays
Whole blood supplementation studies. Citrated whole blood and retrieved plasma studies were carried out. Citrated whole blood from healthy donors was supplemented with the factor Xa inhibitors in a concentration range of 1.0 to 0.61 mg/mL. Samples were analyzed using 1-stage prothrombinase-induced clotting time (PiCT), activated-partial thromboplastin time (aPTT), and prothrombin time (PT) assays. All reagents were reconstituted according to the manufacturer's instructions. The PiCT assay was obtained from PentaPharm (Basel, Switzerland). The TriniCLOT aPTT reagent was obtained from Diagnostica Stago (Parsippany, New Jersey). The PT, HemosIL reagent was obtained from Instrumentation Laboratory (Bedford, Massachusetts). All assays were run on the ST4 from Diagnostica Stago. Whole blood supplemented with the factor Xa inhibitors was centrifuged at 3000 Â g for 20 minutes to obtain retrieved plasma. Theses plasma samples were further analyzed for PiCT, aPTT, and PT using the same methodologies. Results were compiled in terms of mean (standard deviation; n ¼ 3).
Citrated plasma studies. Factor Xa inhibitors were supplemented in citrated plasma in a concentration range of 1.0 to 0.062 mg/ mL. The PT, aPTT, and Heptest (Haemachem, St. Louis, Missouri) were measured using an ACL-Elite (Instrumentation Laboratory, Bedford, Massachusetts). The PiCT was measured on the ST4. Results were compiled in terms of mean (SD; n ¼ 5).
Chromogenic assay
Anti-factor Xa activity was measured by a kinetic amidolytic method using ACL-Elite instrument. The factor Xa inhibitors were supplemented in normal human plasma/blood bank plasma in a concentration range of 1.0 to 0.62 mg/mL. Bovine factor Xa (Enzyme Research Laboratories, South Bend, Indiana) and factor Xa substrate (BioMedica Diagnostics, Connecticut) were used in this assay. The inhibitory potency was calculated in terms of half maximal inhibitory concentration (IC50) as ng/mL or mM.
Inhibition of Thrombin Generation
Inhibition of thrombin generation was measured by fluoroskan ascent fluorimeter, calibrated automated thrombogram (CAT; Diagnostica Stago). Drugs were supplemented in normal pooled plasma to obtain concentrations of 1.0 to 0.062 mg/mL. Reagents used in this assay included the Fluo-Substrate and tissue factor high reagent (mixture of tissue factor and phospholipids) and a thrombin calibrator were obtained from Diagnostica Stago. The thrombin generation was carried out in 96-well Immulon 2HB transparent round bottom plates. The thrombin generation potential was measured by using the peak thrombin, lag time, endogenous thrombin potential /area under the curve (AUC), % peak thrombin, and % endogenous thrombin potential. The potency was measured in terms of IC 50 in ng/mL and mM. Results were compiled in terms of mean (SD; n ¼ 5).
Fibrinokinetic Measurements
Fibrinokinetics was carried out using an optical kinetic method, where thrombin was used to trigger clot formation. Fibrinokinetics profile was measured in a microtiter plate using Spec-traMax Plus 384 microplate reader (Molecular Devices, San Jose, California). The factor Xa inhibitors were supplemented into normal human plasma in a concentration range of 1.0 to 0.062 mg/mL. Thrombin (5U) and 0.025 M CaCl 2 were added to the plasma supplemented with the inhibitor, and the clot density was measured in terms of increased in the absorbance, over 15 minutes. All results were compiled in terms of mean (SD; n ¼ 3-5). Figure 1 shows the composite thromboelastogram of various factor Xa agents compared at 1 mg/mL. Apixaban, edoxaban, and rivaroxaban exhibited similar effects on the TEG. However, betrixaban produced a stronger anticoagulant effect in this assay system. The TEG parameters including r-time, k-time, angle, and MA were compiled from three individual donors for each of these agents as shown in Table 2. Betrixaban produced the strongest effects and the r-time values (32.7 [10.9]) were higher than with other inhibitors. Edoxaban was marginally lower than betrixaban (31.7 [8.6]) followed by rivaroxaban (28.2 [3.5]), whereas apixaban showed relatively weaker effects (23.2 [0.5]) than the other factor Xa agents. The rank order for the k-time followed a slightly different pattern, betrixaban > rivaroxaban > edoxaban > apixaban. The ranked order in the angle was found to be apixaban > betrixaban > edoxaban > rivaroxaban. The MA values for all four agents were comparable and ranged from 51.5 to 54.4 mm. Figure 2 shows a comparison of the ACT results with the different factor Xa inhibitors. In the ACT test, at 2.5 mg/mL, apixaban produced a marginal effect on the ACT (150 seconds) in comparison to saline control (144 seconds). The other agents produced stronger effects prolonging the ACT in the range of 169 to 254 seconds. The difference between edoxaban and rivaroxaban was minimal. A comparison between factor Xa agents in whole blood and retrieved plasma is shown in Figures 3 and 4, respectively. In the whole blood PiCT test, edoxaban showed maximum response followed by betrixaban and edoxaban, which demonstrated comparable effects. Apixaban showed a relatively weaker response. The effects of drugs in aPTT assay for whole blood and retrieved plasma were similar: Betrixaban and edoxaban showed stronger and comparable anticoagulant effects followed by rivaroxaban and apixaban which showed lesser effects. The PT test for whole blood revealed that edoxaban was the stronger anti-Xa inhibitor, followed by betrixaban, rivaroxaban, and apixaban. In contrast to the whole blood, results obtained in retrieved plasma showed slightly different trends ( Figure 4). Edoxaban showed stronger anticoagulant effects, followed by rivaroxaban, betrixaban, and apixaban. In the retrieved plasma, edoxaban showed stronger anticoagulant effects whereas betrixaban and rivaroxaban showed comparable effects, which were slightly weaker than edoxaban. Apixaban produced relatively weaker effects in this test. Figure 5 shows a composite of the effects of various anti-Xa agents when supplemented to citrated normal plasma in the PiCT, Heptest, aPTT, and PT assays. All drugs produced concentration-dependent anticoagulant effects in these tests with the exception of betrixaban, which did not have any effects on Heptest. Assay-dependent variations in the anticoagulant activities of these agents were noted. In the PiCT test, rank order of the anticoagulant activity was edoxaban > rivaroxaban > apixaban ¼ betrixaban. In Heptest, apixaban, rivaroxaban, and edoxaban produced comparable anticoagulant effects with minor variations. Betrixaban did not produce any anticoagulant effect on this assay. In the aPTT assay, the rank order followed betrixaban > edoxaban > rivaroxaban > apixaban. In the PT test, the rank order was rivaroxaban > edoxaban > betrixaban > apixaban. Figure 6 shows the effect of various anti-Xa inhibitors in the amidolytic anti-Xa assay in citrated normal plasma. All agents produced a concentration-dependent inhibition of factor Xa. The IC 50 values followed the ranked order edoxaban < apixaban < rivaroxaban < betrixaban. The anti-Xa activity of edoxaban and apixaban was comparable. Figure 7A to D shows the thrombokinetograms obtained in the thrombin generation (CAT) assay. All of the factor Xa inhibitors produced a concentration-dependent inhibition of thrombin generation. Apixaban, betrixaban, and edoxaban exhibited comparable inhibitory effects; however, rivaroxaban produced relatively weaker effects. At a 1.0 mg/mL concentration, the peak thrombin values were 9.5 (4.4) nM for apixaban, 10.3 (4.2) nM for betrixaban, and 11.1 (6.2) nM for edoxaban, whereas rivaroxaban showed relatively higher generation of thrombin 16.0 (6.7) nM in comparison to the control values 104.6 (19.3) nM. Figure 8 and Table 3 shows the composite of the individual thrombin generation parameters for the four anti-Xa agents as measured by peak thrombin, AUC, and lag time. As shown in Figure 8A, the peak thrombin values show comparable inhibitory patterns with apixaban, betrixaban, and edoxaban, whereas rivaroxaban produced weaker effects. Figure 8B shows endogenous thrombin potential in terms of AUC. Apixaban showed relatively stronger inhibition in this parameter, followed by betrixaban and edoxaban that were comparable. Rivaroxaban produced weaker values at all concentrations in comparison to the other agents. Figure 8C shows the effects of these agents on lag time values. Edoxaban and betrixaban showed delayed lag time in a comparable manner, which was different when compared to apixaban and rivaroxaban. Rivaroxaban produced the most pronounced effects on the lag time followed by apixaban. Figure 9 shows the effect of various anti-Xa drugs on thrombin-induced fibrin formation as measured by the fibrinokinetics method. At a 1 mg/mL concentration, rivaroxaban produced the most pronounced inhibitory effects followed by edoxaban. Apixaban also inhibited fibrin formation though not as strongly as rivaroxaban and edoxaban. Betrixaban produced relatively weaker effects on this assay. The clot density at 15 minutes range from 0.28 to 0.7 in the presence of these agents. Saline-supplemented system showed stronger clot formation with the higher rate of the formation of a clot in 15 minutes. Clot density was much higher than those observed with anti-Xa agents.
Discussion
The relative potencies of the anticoagulant drugs of different classes are usually expressed in terms of their effects on a laboratory assay. The potency of unfractionated heparin is expressed in terms of units per milligram. The lowmolecular-weight heparins are potency standardized in terms of anti-Xa units. However, a reliable study relating the anti-Xa potency to their biological activity is not available. The newly developed parenteral and oral anti-IIa and anti-Xa drugs are usually characterized in terms of their global anticoagulant effects and their inhibitory effects toward thrombin and factor Xa. [5][6][7][8][9] Several studies have compared the anticoagulant effects of these DOACs in global anticoagulant and other assays; however, none of these studies have shown any relevance of the antiprotease potency (Xa and IIa) with other test. The in vitro inhibitory profile and antiprotease effects of these agents may not be proportional to their pharmacological and clinical effects. This is primarily due to the differences in their pharmacodynamic profiles. The currently developed anti-Xa agents include apixaban, betrixaban, edoxaban, and rivaroxaban. These drugs are approved for specific indications, which are largely based on the clinical trial data. 10,11 The structures of these drugs are similar to molecular weights ranging from 435 to 548 kDa. 3 All of these drugs inhibit factor Xa in a competitive reversible manner. All of these drugs are used orally, and their dosages vary widely. Other pharmacological differences are also known and are product specific. These differences may be responsible for the clinical profiles of these newer drugs. [10][11][12] A study to compare these four agents in standardized biochemical and pharmacological assays is not available. This study represents the first report where all four agents are compared in standardized assay systems to relate their activities with the determined anti-Xa actions.
The use of TEG for the detection of the newer oral anticoagulants has been reported previously. 13 In the whole blood clotting assay, the anticoagulant behavior of these drugs was assay dependent. In the TEG studies, all of these drugs produced mild and concentration-dependent anticoagulant effects. At a 1 mg/mL concentration, betrixaban exhibited relatively stronger anticoagulant effects as measured by r-time, k-time, angle, and MA. The other three drugs exhibited a comparable anticoagulant profile. Once again, the TEG profile in terms of anticoagulant parameters was not proportional to the anti-Xa potency of these agents.
In the ACT assay, all of these drugs were capable of producing varying degrees of anticoagulation in whole blood. In these studies, edoxaban produced the strongest anticoagulant effects followed by rivaroxaban. Betrixaban and apixaban exhibited relatively milder anticoagulant effects. The relative anticoagulant effects followed the rank order: edoxaban > rivaroxaban > betrixaban > apixaban. Apixaban exhibited much weaker anticoagulant effects than other drugs. The anticoagulant potency as measured by this assay was not proportional to anti-Xa effects. In addition, in the whole blood assays, the relative anticoagulant effect produced in the TEG was different from the rank order in the ACT. These results are consistent with the previously reported study. 14 This may be due to the differences in the mechanisms of thrombogenesis involved in the two systems.
In the current studies, the anticoagulant profile of all of these drugs was also studied in the whole blood obtained from normal human volunteers. This study was designed to investigate the relative anticoagulant effects in the whole blood and the retrieved plasma from the same systems to differentiate the effects of cells on the anticoagulant profile of the four anti-Xa agents. The PiCT, aPTT, and PT assays were used.
In these assays, all four drugs produced a concentrationdependent anticoagulant effect. In whole blood, PiCT was found to be the most sensitive to demonstrate the anticoagulant effects of these drugs. The aPTT was more sensitive compared to the PT test. Apixaban consistently produced milder anticoagulant coagulant effects compared to the other agents in all assays. In the PiCT test, edoxaban produced the strongest anticoagulant effects, whereas betrixaban and rivaroxaban produced comparable anticoagulant effects. In the aPTT assay, edoxaban and betrixaban were almost comparable, whereas rivaroxaban was much weaker than these two agents. In the PT assay, edoxaban produced the strongest effect followed by betrixaban, whereas rivaroxaban showed much weaker effects.
The same assays were performed on the plasma retrieved from the whole blood samples. In the retrieved plasma, PiCT test once again was found to be much more sensitive than aPTT and PT. Apixaban consistently produced relatively weaker anticoagulant effects in the retrieved plasma systems. Consistent with the results of the whole assay, edoxaban produced relatively stronger anticoagulant effects in the PiCT assay in the retrieved plasma. Although betrixaban and rivaroxaban showed comparable anticoagulant effects in the whole blood PiCT assay, in the retrieved plasma, rivaroxaban produced stronger effects than betrixaban. In the aPTT assay, edoxaban and betrixaban produced similar effects to those observed in the whole blood. Rivaroxaban produced milder anticoagulant effects in comparison to betrixaban and edoxaban. In the PT test, in the retrieved plasma, edoxaban produced the strongest anticoagulant effects while betrixaban and rivaroxaban were comparable, unlike the whole blood results where these two drugs produced differential effects. Interestingly, the anticoagulant effects of these drugs were found to be stronger in the PiCT test as measured in the plasma in comparison to the whole blood; however, in the aPTT and the PT tests, the anticoagulant profiles were stronger in the whole blood in comparison to the retrieved plasma.
These studies underscore the effect of cellular components on the anticoagulant responses as measured by different assays. Moreover, these observations also underscore the multiple mechanisms that may be involved in mediating the anticoagulant responses of the anti-Xa agents. On the basis of these collective results as shown in Table 4, once again the anti-Xa potency is found not to be relevant to the in vitro anticoagulant effects as measured in the whole blood and plasma systems.
The relative anticoagulant activities of apixaban, betrixaban, edoxaban, and rivaroxaban were also studied by directly supplementing these drugs to citrated normal plasma in order to compare their anticoagulant effects. These studies were carried out in the concentration range of 0 to 1 mg/mL using PiCT, aPTT, Heptest, and PT tests. These assays showed varying sensitivities to the anticoagulant effects of these drugs. The PiCT test was found to be the most sensitive test, while aPTT, Heptest, and PT showed relatively weaker responses. In the PiCT assay, edoxaban produced stronger anticoagulant effects than rivaroxaban, edoxaban, and betrixaban. At 1 mg/mL, all inhibitors showed >100 seconds clotting time. In the aPTT assay, apixaban produced marginal anticoagulant effects. The other three agents produced relatively weaker anticoagulant effects in comparison to PiCT test. The anticoagulant effects ranked betrixaban followed by edoxaban, rivaroxaban, and apixaban. In the PT assay, all Xa inhibitors produced relatively weaker anticoagulant effects, with apixaban showing the weakest effect on PT. At concentrations up to 0.5 mg/mL, betrixaban, edoxaban, and rivaroxaban produced comparable anticoagulation; however, at 1 mg/mL, rivaroxaban produced relatively stronger anticoagulation followed by edoxaban and betrixaban. In the Heptest assay, betrixaban did not produce any anticoagulant effect; however, the other three agents showed a concentration-dependent effect. Interestingly, in this assay, apixaban was relatively stronger in producing anticoagulant effects. The observed difference in the anticoagulant profile in the PiCT, aPTT, PT, and Heptest assays may be due to the differences in the activation mechanisms triggering the clot formation in these assays. The PiCT test seems to be the most sensitive in comparison to all of the other methods and may be useful in the monitoring of global anticoagulant effects of these drugs. The results obtained with Heptest assay are somewhat paradoxical, in particular, the lack of response with betrixaban. Moreover, these agents cannot be differentiated based on the anticoagulant activity measured in this assay. Once again, results of these supplementation studies underline the nonrelevance of the anticoagulant activities of these agents with the anti-Xa effects.
Standardized coagulation and amidolytic assays are used to assess the potency of the anti-Xa effects of these agents. [15][16][17][18]6 In the anti-Xa studies carried out, the IC 50 of the currently available anti-Xa agents varied widely. Betrixaban was found to have a relatively higher value for the IC 50 at 1300 ng/mL (2.88 nM) and rivaroxaban exhibited an IC 50 value of 670 ng (1.5 nM), whereas apixaban exhibited a value of 490 ng (1.06 nM) and edoxaban was relatively stronger than the other 3 drugs exhibiting an IC 50 of 430 ng (0.78 nM). Therefore, a large variation in the in vitro potency for inhibiting factor Xa is evident. The current dosages for edoxaban are not proportional to the observed IC 50 for the anti-Xa effects. Apixaban is used at relatively lower dosages in comparison to other drugs, which are considerably higher for various indications. Thus, the in vitro potency and the clinically effective dosing for these drugs are not proportional.
The comparative thrombin generation inhibitory profile of these agents was investigated in normal plasma samples in a concentration range of 0 to 1 mg/mL, which represents the therapeutic range of these agents. The thrombin generation profile was measured in terms of peak thrombin, AUC, and the lag time. In terms of peak thrombin, all agents produced comparable inhibitory effects with the exception of rivaroxaban, which showed somewhat weaker inhibition of thrombin at equivalent concentrations. Although IC 50 values for apixaban, betrixaban, and edoxaban were in the range of 50 to 60 ng/mL, rivaroxaban showed a relatively higher value of 100 ng/mL. On a molar basis, this is approximately twice as high as for the other agents. In terms of AUC, all agents produced concentration-dependent decreases in the AUC, with apixaban exhibiting the strongest effect, and minor differences in AUC values noted with other agents. In terms of lag time, all agents produced a concentration-dependent increase. Apixaban and rivaroxaban produced relatively milder prolongation of the lag time, whereas edoxaban and betrixaban produced more pronounced effects on lag time. A previous report has compared the effects of apixaban and rivaroxaban on the thrombin generation inhibition assay. 19 These thrombin generation parameters were product specific and did not relate to their anti-Xa activities.
The mechanistic basis for the differential effects of rivaroxaban and apixaban on global anticoagulation tests has been reported recently. 20 A complex mathematical analysis on the associated rate constant and projected pharmacodynamics of apixaban and rivaroxaban provides theoretical rationale without any relevance to biologic outcomes. 21 These complex studies may not have any bearing on the observed anticoagulant effects of the anti-Xa agents. A more plausible way is to develop logistic regression models incorporating PT, aPTT, and the PiCT test along with the anti-Xa and thrombin generation assays. The cumulative result then can be compared with the measured factor Xa inhibitory activities of these drugs. The fibrinokinetic study represents an approach where thrombin-induced fibrin formation is measured in citrated plasma by monitoring clot density over time. All of these agents were found to produce an inhibition of fibrin formation at a concentration of 1 mg/mL in this system. While rivaroxaban produced the strongest inhibition of fibrin formation, edoxaban and apixaban produced moderate effects for the formation of fibrin and betrixaban only produced a mild effect. These studies show that the anti-Xa agents are capable of inhibiting fibrin formation in a varied manner. Although thrombin generation inhibition is a key factor in this process, interestingly the thrombin generation inhibitory potency doesn't correlate with their fibrinokinetic inhibitory effects. For example, although rivaroxaban is the strongest inhibitor in this assay, its potency in the thrombin generation inhibition assay is weaker than that of other agents. In regard to the relevance of fibrinokinetic inhibition by these agents with anti-Xa activity once again, a direct correlation is not observed; therefore, additional factors may contribute to the fibrinokinetics inhibitory potential of these agents.
Evidently, the factor Xa inhibitory profile as measured utilizing the standardized amidolytic assay to designate the potency of these drugs doesn't reflect their biologic effects in the in vitro studies carried out in whole blood, plasma, and defined biochemical systems. Each agent has a distinct clotting and biochemical profile that cannot be predicted on the basis of their anti-Xa potency. The studies reported on the anti-Xa profiling were carried out in a bovine factor Xa system, which is commonly used to characterize their anti-Xa spectrum. It is likely that similar outcomes will be observed if human factor Xa is used. Based on these findings, each of the factor Xa agents should be considered as a distinct drug whose biochemical and pharmacological actions are independent of its measured anti-Xa activity. Furthermore, beside the reported activities, these drugs may produce their biologic effects through other mechanisms that are not completely understood at this time. This includes their effects on blood cells, endothelial cells, and platelets. Finally, it must be underscored that the pharmacodynamic profile of these drugs may also be dependent on the differential nature of the pharmacophores in these molecules. Despite their simple structure, these drugs may exhibit complex interactions with enzymes and other binding sites to exert their effects.
An antidote, andexanet alfa representing a decoy protein as recently, becomes available to neutralize the anti-Xa effects of apixaban and rivaroxaban. 22,23 This antidote is likely to neutralize other anti-Xa drugs, such as betrixaban and edoxaban. However, it remains unclear that solely neutralizing the anti-Xa activities will antagonize many of the other pharmacodynamic effects of the anti-Xa agents. The recently observed thrombogenicity and other adverse events along with the high mortality rate associated with the use of andexanet may be due to residual actions of anti-Xa agents and other undetermined effects of andexanet.
In summary, the reported studies in this manuscript clearly suggest that despite their central action targeting factor Xa inhibition, the oral anti-Xa drugs exhibit markedly different anticoagulant properties as measured by different assay systems. Since these assays are designed to measured distinct target sites in the coagulation cascade, the cumulative biologic effects of each of these agents may not be directly related to their observed anti-Xa potencies. Therefore, the relevance of their clinical spectrum and potential neutralization by specific antidotes such as andexanet alfa may not be proportional to their anti-Xa effects. These observations warrant further investigations in both the laboratory and clinical setting to generate data on each of the individual anti-Xa agents for developing paradigms for their optimal clinical use.
|
2019-05-16T13:03:48.824Z
|
2019-01-01T00:00:00.000
|
{
"year": 2019,
"sha1": "e782a6d9cdb1af3457ff75d06d8cba0fa58ba5ca",
"oa_license": "CCBYNC",
"oa_url": "https://journals.sagepub.com/doi/pdf/10.1177/1076029619847524",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "1ede5bd5da890461d6efcad4e6d6c3aa5736e97a",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
53448847
|
pes2o/s2orc
|
v3-fos-license
|
Distribution of fishery benefits and community well-being : a review of increased access to the Eastern Nova Scotia snow crab fishery
An expanding fish stock offers a rare opportunity to support fishing enterprises whose traditional fisheries have diminished or failed. The Eastern Nova Scotia snow crab fishery is one example, where in 2005, a growing stock allowed benefit-sharing among more than 700 harvesters. As a contributing case study of social and institutional aspects of sustainability, we review the background of that fishery and the outcomes of the redistribution of fishery benefits. Based on more than 50 semistructured interviews, the case study demonstrates how conflict has settled into cooperation, with the fishery remaining biologically sound and highly beneficial to individuals and their communities. However, the method chosen to manage the distribution in Eastern Nova Scotia has not guaranteed that benefits will remain in local communities. In other jurisdictions, alternative approaches developed in conjunction with broad-based harvester organizations demonstrate better benefit retention in local communities. When compared with the Canada Fisheries Research Network Sustainability Framework, this case study offers insights into the benefits that thoughtful resource redistribution can provide, illustrating that fishery policy decisions must anticipate long-term implications and should apply a definition of fisheries sustainability that includes community well-being, in this case, as evidenced in local licence retention.
INTRODUCTION
A property rights-based approach (primarily through Individual Transferable Quota [ITQ]) to management dominates globally in industrial fisheries (FCC 1994, EDF 2015).This is also true in Canada, but since the 1970s, Canadian fisheries management has included multiple economic and social as well as biological objectives (Matthews 1993).The benefit distribution under ITQs thus generated significant debate, given how initial allocations affected both those who obtained and those who were denied access (Wiber 2000).Both the historic participation that determined original quota allocations (Arnason 1996, Shotton 2001, Anderson and Libecap 2014) and the subsequent redistribution of allocations has generated conflict (Government of Western Australia 2011, Harvey 2013).The impact of climate and other environmental change has been a confounding factor (Fulton 2011).In the Scotian Shelf ecosystem of eastern Canada, for example, Frank et al. (2005:1623) have noted how shrimp and snow crab landings now "far exceed the groundfish fishery it replaced."Such changes in abundance have required new access decisions and benefit redistribution among Eastern Nova Scotia (ENS) fishermen.
The federal Canadian Sustainable Fisheries Framework requires that biological and socio-economic consequences of management measures must be considered (DFO 2013) but does not specify how socio-economic consequences are to be assessed.Canada is not unique in this regard (Brooks et al. 2014).The Canadian Fisheries Research Network (CFRN) addresses this gap through the CFRN Comprehensive Fisheries Sustainability Framework (hereafter referred to as the CFRN Framework, see Table 1), which emphasizes the importance of considering all four elements necessary to build sustainability: the ecological, economic, social, and institutional (CFRN 2012, Stephenson et al. 2017).Distribution of benefits affects all domains within the CFRN Framework, but the ENS redistribution highlights economic, social, and institutional factors, particularly as they relate to financial viability, intergenerational equity, sustainable communities, and transparent and democratic decision-making.
We examine the results of a 2005 decision to permanently expand access to snow crab on the advice of the Advisory Panel on Access and Allocation, Eastern Nova Scotia Snow Crab Fishery (referred to hereafter as the Panel).While the Panel's recommendations affected snow crab Fishing Areas (CFAs) 20 through 24, [1] this case study focuses on CFAs 23 and 24, where high rates of participation, production, and access transactions better illustrate subsequent trends.The results from this case study highlight the need for indicators that evaluate the long-term effects of the approach to such redistribution of benefits.
METHODS
To better understand how the permanent expansion of the ENS snow crab fishery was allocated and to compare the results with other snow crab fisheries in the Atlantic region, 53 semistructured interviews were conducted after purposive sampling to reach those who had been active in the presentations to the Panel, in negotiations with Fisheries and Oceans Canada (DFO), and in establishing snow crab fishing collectives.Snowball sampling expanded the sample to represent those both for and against the allocation process.The sample included snow crab fisheries participants (N = 28), federal and provincial fisheries bureaucrats (N = 16), and industry consultants/representatives (N = 9) across Atlantic Canada.Interviews were conducted until topic saturation was reached.Fishermen were interviewed using a semistructured interview schedule (Appendix 1), while other industry players were involved in informal discussions or responded to specific questions.The first author, who is a fisherman in ENS, undertook all interviews and data analysis.A literature review included academic, grey, and government literature, government data, and media sources.Analysis involved hand coding the interview transcripts and literature for themes, with three emerging that are the focus here: how the allocation process came about, the implications of the approach taken by DFO, and alternative approaches not taken.In what follows, we provide background before exploring these themes and then turning to conclusions.
RESULTS
In what follows, we explore the three themes that emerged from the data and their connection to key CFRN Framework dimensions, including the DFO process followed in making snow crab allocations (institutional process and outcomes), the implications of that approach (equity and fairness), and alternative approaches not taken (livelihood sustainability).
The Eastern Nova Scotia allocation process
Approximately 120 permanent and about 700 temporary participants had to be considered in the 2005 allocation process.Panel member Beaton observed that temporary participants had to "be accommodated somehow," as temporary access had contributed greatly to sustainable fishing communities, and "the strength of the resource gave us some wiggle room" (personal communication, 11 June 2014).Three groups were involved: displaced groundfishermen who had been granted access first, core licence holders who lived adjacent to the management zones, and core nonadjacents.Each had received different allocations during the temporary access period, and the Panel was charged with addressing this (Gardner et al. 2005).This generated tension among temporary participants, and between them and the permanent fleet (Burke and Patterson 2005).This situation represented a significant challenge to equity and fairness in the allocation of resource benefits (see CFRN Framework, Table 1, Socio-Economic Domain).One informant reported, "it was bitter -it was really bitter.People were wanting nonadjacents eliminated and all three groups had their own agendas." It was bitter because of the importance of snow crab income to enterprise and community survival.As one informant noted, "Some just got a few thousand dollars the first couple of years [from crab], but it was enough to keep them going.And then when they started making a bit more on it, they started to get better gear and better boats -you wouldn't have half the good boats you got around here now if it weren't for the crab.You wouldn't have the life you have now -none of us would." While "multilicenced enterprises" (DFO 1996) were "promoted" in licensing policy, "fostered" in the Atlantic Fisheries Policy Review (DFO 2004b), and recommended to improve enterprise sustainability and management flexibility (Charles 2005), for many ENS fishermen, "multilicenced" had been winnowed down to lobster and crab.Eastern Nova Scotia enterprises vary greatly in financial performance (LeBreton 2012), making it difficult to estimate snow crab contributions, but informants reported between 20 and 50% of annual net income from crab, a significant contribution to livelihood sustainability.This heavy dependence on crab may explain why members of industry organizations had successfully collaborated on managing temporary snow crab allocations through democratic institutions, with a focus on community sustainability.Temporary snow crab access was often managed by fishermen's associations, which sometimes divided community allocations among members to fish themselves, and sometimes hired boats to fish the pooled allocation.In the latter case, after administration costs, all income was divided annually among members.Respondents felt this system had worked well, with management options decided both democratically and in a transparent fashion.
The Panel recommended that DFO make access permanent for all current participants, a decision that stabilized access to resource benefits.But the Panel made two further recommendations that had far-reaching consequences.First, they ignored the creativity of industry management institutions that had managed crab allocations to promote community benefit.https://www.ecologyandsociety.org/vol23/iss2/art25/Instead, the Panel followed the advice of an earlier DFO discussion paper (DFO 2004a) and treated snow crab participants with their small individual allocations as "quota-holders."Second, the Panel recommended pooling individual allocations into "some form of legal entity in which they [held] shares," which would then be issued a licence (Gardner et ), the final Panel report did not directly refer to community welfare, equity, and fairness values.Among fishermen informants, however, sustaining communities and regions through stability of access to resource benefits remained an important value.The question arises then, did the allocation process result in sustainable coastal communities as defined by fishermen?
Consequences of the Eastern Nova Scotia process
One example typifies the problems of maintaining community benefits from snow crab over time.In Canso, the Canso Trawlermen's Co-operative Limited (CTCL) was formed in 1997, acquired a boat to fish shrimp and groundfish, and later received a licence for snow crab.The Canso Trawlermen's Co-operative Limited supported community economic development (Perry 2003), but over time, membership declined and CTCL began hiring boats to fish its snow crab.Some local fishermen tried to arrange financing to acquire the company assets, only to discover the licences and associated allocations had been abruptly sold in 2013 to a company from southwest Nova Scotia (SWNS), a nontransparent decision reached without full consultation of members.When asked whether CTCL's assets might have been kept in Canso, one informant said that would have been difficult, since few locals were qualified to operate the vessel and remaining members wanted fair market value because they viewed the assets as their retirement fund.He noted that the snow crab is still fished from and landed in Canso.But other informants felt strongly that resource allocations should remain tied to communities.
In other communities, interviewees expressed concerns about the internal operation of Core Companies, including no effective control over contracted skippers, little sense of ownership in that individual allocations were not recorded by DFO, inappropriate influence being exerted by outsiders, and low levels of engagement of fishermen in company operations.Numerous fishermenshareholders indicated they never attended company meetings.One respondent, who was the secretary for the company, had not attended a meeting since the company's first year.He simply signed papers when required by the company lawyer.Two respondents anticipated pursuing legal action to solve problems within their Core Companies, and another had already gone to court seeking compensation for his allocation, the ownership of which was in dispute (Cape Breton Post 2015).
The Panel had suggested providing legal expertise to help set up companies and develop Shareholders' Agreements (Gardner et al. 2005).When queried about failure to do this, one retired DFO manager protested that "most of these guys were already incorporated."His response ignores the contrast between an individually owned company and one comprised of numerous novice shareholders.As one fisherman observed, "What do I know about running a company?I can run my own outfit alright, but this is different."One lawyer who has worked with Core Companies observed that they were rushed into existence with no education of shareholders or directors, no provision for arbitration should internal disagreements become problematic, and no clear-cut process to wind the company up.Clearly, these institutional arrangements lacked good structure, rules, and overall strategies for shared decision-making, and resources were not provided to ensure they were in place (CFRN Framework, Table 1, Institutional Domain).
As shareholders approach retirement, other implications of Core Companies emerge.Usually, a retiring fisherman will sell his fishing assets and operation to another fisherman, but Core Companies offer options to retain snow crab benefits after retirement.Fisheries and Oceans Canada requires that all Core Company members are core fishermen, but a retiring fisherman who holds several core licences could retain one and remain in https://www.ecologyandsociety.org/vol23/iss2/art25/his Company, or could place his share in trust with another shareholder.Although DFO has moved to eliminate trust and controlling agreements where captains cede control of their enterprise to the processing sector (DFO 2015), trust agreements related to company shares are subject to different legal oversight.
It is unclear whether DFO has the authority, capacity, or interest to scrutinize such internal financial or legal arrangements, perhaps leading to outcomes that are noncompliant with DFO policy (CFRN Framework, Table 1, Institutional Domain).
Other consequences arise from the growth in value of individual allocations, and their mobility.Informants reported that allocations initially sold for as little as Can$22,000 but have since increased to more than Can$250,000.This value was driven by consistent returns from snow crab, by acquisitions by Aboriginal groups supported by federal funding (Wiber and Milley 2007), and by the lucrative SWNS lobster fishery that generates capital to invest in other fisheries.The number of ENS Core Company shareholders declined from 715 in 2005 to 533 in 2014.The 182 who divested could have done one of the following: sold their allocation to a Core Company or an individual shareholder, sold it to a traditional snow crab licence holder, or transferred it, in trust, to a shareholder in a Core Company (thereby benefiting an inactive harvester).The first two options result in benefit concentration but retain benefits within the area unless the licence itself has been sold outside the community.Such external transfers are on the rise as more SWNS fishermen seek both ENS crab allocations and licences.Recently, the company the first author belongs to received four purchase offers, three of which were from SWNS.Eastern Nova Scotia licence holders report that they cannot compete with the high prices offered by outsiders to the region; thus, ENS crab allocations continue to leave the area.
Interviewees also expressed concerns about intergenerational equity.Fisheries and Oceans Canada records suggest there are few opportunities for new entrants to acquire a snow crab allocation.Privacy requirements with Core Company shareholder records make it difficult to track ownership changes, so we relied on informant interviews.While more than 140 lobster licences changed hands in Lobster Fishing Area 27 over the past 10 years, we could locate only six new harvesters who acquired a snow crab allocation along with a lobster licence.Of these, only one was part of a complete enterprise; all others involved some form of family support.It is rare to see a snow crab allocation sold with the licence to which it was originally attached.
Despite these problems, many Core Companies are operating well.Members of three Core Companies reported their Shareholders' Agreements largely accomplished their intended goals.All three companies have acquired allocations, either from retiring members or outside harvesters, and one retains a portion of annual revenues to fund further purchases.But along with this success have come new attitudes and behaviors.While some respondents deplore the increasing use of hired boats, one licence holder observed ironically that "leasing out their quota and hiring people from away, it's just bad business -well it's bad stewardship, but it's probably good business."Numerous respondents would have preferred receiving allocations directly so they might fish it either alone or in groups.But if that was not possible, they were willing to become "investors" in snow crab allocations.In Newfoundland, Davis and Korneski (2012) observed similar changes in attitudes and language use, which reflected more entrepreneurial and individualistic perspectives.As in ENS, language reflects a change from valuing a community access to fish to valuing individual ownership of fish.
The range of identities identified herein-fishermen, shareholders, allocation holders, quota holders-is not mere semantics.The Panel report uses "quota-holder" to describe those whose allocations were combined within an incorporated company, whereby they became shareholders.But many fishermen relate to Core Companies more like members of an organization rather than as engaged shareholders.One frustrated president of a Core Company explained that fishermen "don't own anything except a share in a company that owns the licence, that owns the allocations."While possessing an allocation made fishermen part of a Core Company, when they retire, they typically sell the allocation rather than the company share, giving rise to confusion.Several fishermen, despite having sold their allocation, believed they were still company shareholders with voting privileges.In another Company, the President tried to exert control by temporarily transferring several allocations into the Company in his own name.The lack of sound institutional arrangements around Core Companies has provided opportunities for problematic practices.
Fisheries and Oceans Canada's Owner Operator policy requires that licence holders operate the boats to which licences are attached, and fishing industry organizations strongly support this.The Core Company approach contradicts this policy, and shareholders recognize that they are benefiting from the erosion of Owner Operator policy.Shareholders also recognize that there are community and industry implications when allocations leave enterprises to which they were originally granted.The question thus arises as to whether a different approach might have produced greater community and enterprise sustainability.opposed to ITQs because, as one interviewee put it, "we take the position that the fish should be there for the guys who fish it, not the guys who peddle it."With no transferability, a licence holder must fish their quota or those crab stay in the water.Since the initial expansion, economic pressures have necessitated changes such as the "buddy-up system," where two licence holders can combine a maximum of three Individual Quotas on one boat.These changes have been developed with varying levels of support from the fishermen's union, but the organization has consistently protected the principle that the one who is fishing should get the benefits.On the west coast of Cape Breton Island, the snow crab fishery has seen several iterations of resource sharing (Loucks 2005), resulting in a reasonable price of entry, no excessive quota concentration, and benefits that have not migrated far afield.In these cases, good governance has allowed for transparency and accountability, which in turn improves sustainable communities.
CONCLUSIONS
The CFRN's Framework incorporates social, economic, institutional, and ecological considerations in assessing fishery sustainability (Table 1).According to informants, the ENS snow crab rates quite highly on several measures.Stock productivity has remained strong, and other ecological aspects are accounted for within the Integrated Fisheries Management Plan.Under the socio-economic domain of the Framework, snow crab allocation greatly improved local economic stability for individual fishing enterprises and communities.Institutionally, a well-functioning industry advisory board has been created and contributes to operational details of the ENS fishery.Management is generally considered effective.Economically, the strong market for snow crab allocations indicates satisfaction with financial returns and confidence in future prospects.
However, this case study also illustrates how institutional arrangements (structure and process) may allow benefits to exit communities, thus affecting equity and fairness and access stability over the long run.The longevity of such benefits can also be undermined when left to unfettered choice, especially when associated with individual quasi-property rights.The pooling of ENS snow crab allocations in Core Companies has led to problems as fishermen age out of the industry, with the company structure facilitating the separation of benefits from active fishing enterprises.Finally, unfettered allocation transferability contributes to escalating prices, with few new entrants acquiring snow crab allocations.This affects intergenerational equity and the right to a livelihood.
Attitudes and actions of individual harvesters can be significantly shaped by fishery policy.Many interviewees would prefer to fish crab allocations themselves but were ready to invest in additional allocations, an option that might not have existed under different incentives.Responding to policy incentives is seen simply as adapting to the rules of the game, as is arranging to retain benefits from snow crab upon retirement.Further, poorly designed company structures contributed to the questionable actions of a few shareholders who seek to control all Core Company assets.These problems illustrate the significance of the Framework's inclusion of institutional process, particularly related to rules, collaboration, cooperation, and transparency (see Table 1).
The institutional domain in the CFRN Framework also highlights the importance of how decisions are reached; democratic deliberations within industry organizations often resulted in more sustainable agreements that are comprehensive and enforceable (Wiber et al. 2004).Temporary ENS snow crab allocations had been managed successfully by industry organizations, and the Gulf of St. Lawrence-based organizations continue to manage such that benefits go to active harvesters.In Newfoundland, expanded access to inshore snow crab through individual licensing and nontransferable quotas has resulted in harvesters who both receive the benefits and have an incentive to participate in management decisions.In ENS, local contestation and limited organizational capacity set the stage for an imposed rather than a negotiated solution.
Consideration of the CFRN Framework's explicit focus on regional economic community and livelihood sustainability might have avoided shortcomings identified in this case.As asserted by many interviewees, creating permanent allocations in the ENS snow crab fishery provided timely, significant, and worthwhile benefits for fishing enterprises and the communities in which they operate.The most serious concern, according to interviewees, is how long benefits will continue to accrue to the communities to which they were first distributed.The choice of a property rights-based approach has allowed benefits to flow out of the immediate area, and rising prices for snow crab allocation makes acquisition by new entrants to the fishery very difficult.Had snow crab allocations been tied to the original communities, or to the core licences to which they were initially distributed, these problems could have been minimized.
The decision to introduce property rights to fish harvesters has affected incentives, attitudes, and behaviors of those harvesters.Such incentives and behaviors have implications for the social and economic objectives outlined in fishery policy.Responsible management should consider these broader implications, just as it accounts for narrow ecological aspects.The various approaches to managing the expanded ENS snow crab fisheries offer important evidence with long-term management implications.
Collectively developed and managed approaches demonstrate better benefit retention within local communities, as compared to individually allocated quotas.When a decision to share public fishery resources is made, both the means by which that decision is made and the duration and location of benefits should be considered, and this case strongly supports a collective approach as a means of implementing established Canadian fisheries policy objectives. [1] See https://marine.rutgers.edu/~cfree/wp-content/uploads/dfo_east_snow_crab_mgmt_areas.gif Responses to this article can be read online at: http://www.ecologyandsociety.org/issues/responses.php/10137 Ecology and Society 23(2): 25 https://www.ecologyandsociety.org/vol23/iss2/art25/
Table 1 .
Canadian Fisheries Research Network Framework for Sustainable Fisheries, Version 2.1 with domains, dimensions, elements, indicators, and attributes. 3 5 3 10 • Proportion of sensitive [Benthic Species] 11 subject to [Anthropogenic Activity] 10 3 3 3 3 (Matthews 1993)chaud 1981, GardnEagles 2008)05Public Inf, 2011)ture] 44 in [Human Geographic Region] 22 • Value of fisheries related [Fisheries-Related Private Infrastructure] 43 in [Human Geographic Region] 22 • [Benefit Axis] 45 by [Socio-economic Distribution Axis]46• [Cost Axis] 47 by [Socio-economic Distribution Axis]46• Distribution of [Value Type] 48 by [Value Chain Element] 49 • Distribution of [Value Type] 48 by [Operator Type] 42 Level and duration of [Support] 67 for [General Management Activity] 68 and/or [Fisheries Management Activity] 69 among [Stakeholder Group] 38 and/or [Human population] 21 at [Human Geographic Region] 22 • Types of [Conflict Resolution Approaches] 70 available to deal with disputes Qualitative] 35 evidence of [Stakeholder Group] 38 and [Human Population] 21 perception of appropriateness • Presence/absence of role for [Stakeholder Group] 38 in the development, establishment, and enforcement of rules at the [Rule Level] 82BACKGROUNDIn ENS, small inshore boats pursuing a multispecies fishery first began harvesting snow crab in 1966.By the mid-1970s, DFO was managing the fishery as supplementary to other fisheries(Elner and Robichaud 1981, Gardner et al. 2005, DFO 2007, 2011).The snow crab fishery expanded slowly until 1989, with new exploratory licences issued, and was later converted to regular commercial status(DFO 2000).Fisheries and Oceans Canada now views the ENS snow crab fishery as "a primary fishery"(Barrow et al. 2001:49).Thus, the entire ENS snow crab fishery lies within the living memory of harvesters, and has grown into a large and profitable industry, the benefits of which many believe should be more widely shared.The Canadian fishing industry changed greatly after the introduction of the 200-mile (322-km) Exclusive Economic Zone.Fisheries and Oceans Canada's "regulatory interventions mushroomed" (Parsons 2010:393), including introduction of seasonal total allowable catches, allocation of access among fleet sectors (inshore, midshore, and offshore), limited-entry licensing, and introduction of quotas in some sectors.Inthe 1980s, DFO licensing policy began to distinguish between part-and full-time fishermen(Matthews 1993), aiming to eliminate the former in order to enhance the livelihoods of the latter.Core fisherman status required owning a boat and fishing licences and operating the enterprise.Gien such regulatory pressures, individual fishermen sought every opportunity to protect their full-time, core status.Ecnoic pressures were also increasing, as the 1992 groundfish moratorium had eliminated 30-60% of the overall catch in the Maritimes and Quebec (DFO 1993). Eaern Nova Sotia lobster landings were declining at the same time (FRCC 1995, Annand and Peacock 2001,Barrow et al. 2001, Peacock andEagles 2008). Snow rab represented an opportunity for income replacement (DFO 2001), but expanding the crab fishery required favorable stock assessments.
• Change in [Biodiversity Indices] 9 over [Time Period]5Non-native species: extent and impact of non-native species • Degree of impact of introduced species on [Food-web Stability]17• Probability of introduction of new species to ecosystem • [Quantification] 12 of introduced species in ecosystem • Probability of ability to extirpate introduced species, proportional to the degree of impact to[Food-web Stability]17• [Quantification] 12 of extirpation of introduced species, proportional to the degree of impact to [Food-web Stability] 17 • [Quantification] 12 of aquaculture escapes • [Quantification] 12 of introduction and proliferation of disease/pathogens • [Risk Axis] 50 by [Socio-economic Distribution Axis] 46 Livelihoods: sustainability of livelihoods • [Livelihood Index] 51 applied at [Human Geographic Region] 22 • Unemployment rate in fishery-dependent [Human Geographic Region] 22 Economic and financial Human capital: development and maintenance of human capital • [Human Demographic Axis] 52 by [Occupational Axis] 53 • [Quantification] 12 of [Time Period] 5 in the industry by [Occupational Axis] 53 • [Quantification] 12 of generations of fishing history of current participants in the fishery • [Quantification] 12 of fishermen meeting [Certification Standards] 29 Efficiency: maximization of harvest value relative to waste • Realized catch relative to potential target harvest • [Quantification] 12 of [Resource Demographic Category] 2 discard waste • Market price relative to private marginal cost of production • Cost of output for [Economic Unit] 37 by [Fishery Category] 14 relative to the lowest possible average total cost • Output obtained from a given quantity of inputs relative to the maximum output obtainable from that given quantity of inputs • [Productivity] 54 of [Economic Unit] 37 by [Fishery Category] 14 • [Efficiency] 55 of [Economic Unit] 37 by [Fishery Category] 14 DOMAIN: INSTITUTIONAL Structure Rules: legal, regulatory, and policy framework is appropriate • Proportion of [Anthropogenic Activity] 10 covered by [Institutional Arrangement] 65 and subject to [Legislation/Regulation] 61 and/or [Management Plan] 66 • [Qualitative] 35 evidence of support for the [Institutional Arrangement] 65 and/or [Legislation/Regulation] 61 and/or [Management Plan] 66 among [Stakeholder Group] 38 • [Qualitative] 35 evidence of consistency between the [Institutional Arrangement] 65 and [Legislation/Regulation] 61 and [Human Population] 21 norms and values • [Qualitative] 35 evidence of consistency in [Institutional Arrangement] 65 between [Stakeholder Group] 38 • [Qualitative] 35 evidence of [Stakeholder Group] 38 and [Human Population] 21 perception of inclusivity • Degree to which [Inclusivity Criteria] 75 exist • [Quantification] 12 of [Inclusivity Criteria] 75 • [Quantification] 12 of [Stakeholder Group] 38 participation in [General Management Activity] 68 and/or [Fisheries Management Activity] 69 • Degree to which [Accredited Organization Criteria] 83 was consulted in the development, establishment, and enforcement of rules at the [Rule Level] 82 • Degree to which [Stakeholder Group] 38 role in the development, establishment, and enforcement of rules at the [Rule Level] 82 is commensurate with impact of rule on the [Stakeholder Group] 38 • Degree to which there is [Flexibility Criteria] 78 Trade-offs: explicit consideration of trade-offs in decision-making • [Qualitative] 35 evidence of [Stakeholder Group] 38 and [Human Population] 21 perception of trade-off • Degree to which [Trade-off Criteria] 84 are identified and implemented • [Quantification] 12 of [Trade-off Criteria] 84
In Newfoundland and Labrador, a strong resource allowed significant expansion in the snow crab fishery (DFO 2011).Access became permanent in 2003, and the 2009 Integrated Fisheries Management Plan lists 2683 inshore licences out of a total of 3455.This inshore fishery operates with Individual Quotas as https://www.ecologyandsociety.org/vol23/iss2/art25/
|
2018-11-01T17:56:44.761Z
|
2018-05-14T00:00:00.000
|
{
"year": 2018,
"sha1": "70f3958f2e3efbea7443e1d1af517113ef16a913",
"oa_license": "CCBY",
"oa_url": "https://www.ecologyandsociety.org/vol23/iss2/art25/ES-2018-10137.pdf",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "70f3958f2e3efbea7443e1d1af517113ef16a913",
"s2fieldsofstudy": [
"Environmental Science",
"Sociology"
],
"extfieldsofstudy": [
"Economics"
]
}
|
21136578
|
pes2o/s2orc
|
v3-fos-license
|
Expression Profiling of Mitogen-Activated Protein Kinase Genes Reveals Their Evolutionary and Functional Diversity in Different Rubber Tree (Hevea brasiliensis) Cultivars
Rubber tree (Hevea brasiliensis) is the only commercially cultivated plant for producing natural rubber, one of the most essential industrial raw materials. Knowledge of the evolutionary and functional characteristics of kinases in H. brasiliensis is limited because of the long growth period and lack of well annotated genome information. Here, we reported mitogen-activated protein kinases in H. brasiliensis (HbMPKs) by manually checking and correcting the rubber tree genome. Of the 20 identified HbMPKs, four members were validated by proteomic data. Protein motif and phylogenetic analyses classified these members into four known groups comprising Thr-Glu-Tyr (TEY) and Thr-Asp-Tyr (TDY) domains, respectively. Evolutionary and syntenic analyses suggested four duplication events: HbMPK3/HbMPK6, HbMPK8/HbMPK9/HbMPK15, HbMPK10/HbMPK12 and HbMPK11/HbMPK16/HbMPK19. Expression profiling of the identified HbMPKs in roots, stems, leaves and latex obtained from three cultivars with different latex yield ability revealed tissue- and variety-expression specificity of HbMPK paralogues. Gene expression patterns under osmotic, oxidative, salt and cold stresses, combined with cis-element distribution analyses, indicated different regulation patterns of HbMPK paralogues. Further, Ka/Ks and Tajima analyses suggested an accelerated evolutionary rate in paralogues HbMPK10/12. These results revealed HbMPKs have diverse functions in natural rubber biosynthesis, and highlighted the potential possibility of using MPKs to improve stress tolerance in future rubber tree breeding.
Introduction
The rubber tree (Hevea brasiliensis) is the most important commercially cultivated natural rubber-bearing plant. It is a member of the spurge family (Euphorbiaceae) and is closely related to cassava (Manihot esculenta) and castor bean (Ricinus communis). Natural rubber (cis-1,4-polyisoprene) is strategically important, and no other synthetic alternatives possess the exact same physicochemical properties. Therefore, rubber tree has substantial economic value as the only commercially cultivated
Identification and Genomic Location of HbMPK Genes
Previously reported MPKs of Arabidopsis [15] and rice [16] were retrieved and subjected to a Basic Local Alignment Search Tool for protein query (BLASTP) against the Hevea genome database, in order to comprehensively annotate MPKs [4]. Candidate HbMPK gene sequences were then submitted to InterProScan [30] to assess the MPK conserved motif (IPR003527) and protein kinase domain (IPR000719). A careful manual review of the resultant HbMPK candidate genes was performed to correct any potential mistakes in the genome database.
The genomic locations of these identified HbMPK genes were further determined based on the results of a BLAST nucleotide (BLASTN) query against the Hevea genome sequence. MapInspect software was used to draw the locations of the HbMPK genes.
Phylogenetic Tree, Motif Distribution and Alignment of the Mitogen-Activated Protein Kinase Family
The amino acids that compose MPKs from Hevea, two model plants Arabidopsis and rice, and a closely related cassava species [31] were identified, after which a phylogenetic tree was constructed by MEGA5.0 [32] using neighbour-joining (NJ) method. Bootstrap tests were performed and consisted of 1000 replicates.
Multiple sequence alignment was performed using ClustalW [33]. The molecular weight and isoelectric points of predicted HbMPK proteins were estimated using the ExPASy proteomics server [34]. The subcellular localization of HbMPKs was predicted using Softberry [35]. Conserved motifs of the HbMPK proteins were analysed using the Multiple Expectation Maximization for Motif Elicitation (MEME) program [36].
Gene Structure, Duplication Events and Syntenic Analyses
Gene Structure Display Server (GSDS) software [37] was used to analyse the exon-intron distribution based on the comparison of the open reading frame (ORF) sequence and genomic coding region of each HbMPK gene.
Gene duplication events were determined by multiple sequence alignment and collinear analyses. Homologous HbMPK genes were considered as paralogues only when their nucleotide identities were >90% [38,39]. The Ka (nonsynonymous substitution rate) and Ks (synonymous substitution rate) were calculated using DnaSP 5.0 software [40]. The Ka/Ks ratios for the HbMPK genes were calculated to assess the selection pressure on duplicated genes; a Ka/Ks ratio >1, <1, or =1 indicates positive, negative, or neutral evolution, respectively [41]. Tajima relative rate tests [42] were performed using the amino acid sequences of the duplicated HbMPK pairs. The syntenic relationships of paralogues and/or orthologues among Arabidopsis, cassava, and rubber tree were analysed using the Circos program [43].
Plant Materials and Abiotic Stress Treatments
Three rubber tree cultivated varieties, Reyan8-79 (high yielding), Reyan7-33-97 (medium yielding), and PR107 (low yielding), were used in this study. Different tissues (roots, stems, leaves and latex) were collected from the ten-year-old mature trees, and immediately frozen in liquid nitrogen. One-year-old Reyan7-33-97 seedlings were used for abiotic stress treatments. For osmotic and salt stress treatments, the seedlings were irrigated with 20% polyethylene glycol (PEG) 6000 and 1 M NaCl, respectively. For oxidative stress, the leaves were sprayed with 2% (v/v) H 2 O 2 . For cold stress, the seedlings were placed in a growth chamber at 4 • C. For RNA extraction, the leaves were collected at 3 h, 12 h, and 24 h after stress treatment. Untreated rubber tree leaves were used as 0 h control.
RNA Extraction and Expression Profiling of HbMPKs
The total RNA from different tissues (roots, stems, leaves and latex) of three cultivated varieties and the leaves of Reyan7-33-97 following abiotic stress treatments were extracted using an RNAprep Pure Plant Kit (Tiangen, Beijing, China). First-strand complementary DNA (cDNA) was generated using a reverse transcription system (Takara, Kusatsu, Japan) in accordance with the manufacturer's instruction.
The qRT-PCR was conducted using the gene specific primers listed in Supplementary Table S1. The reactions were performed using a Stratagene Mx3005P Real-Time Thermal Cycler (Agilent, Santa Clara, CA, USA) and SYBR Green Master Mix Reagent (Takara, Kusatsu, Japan), in accordance with previously described PCR conditions [44]. The specificity of the reaction was verified by melting curve analysis. The 2 −∆∆CT method was used to calculate the relative expression level of the target genes. The HbActin (GenBank Acc. HQ260674.1) served as a reference gene. Three independent biological replications were performed for each gene. Heat maps were generated using Multi Experiment Viewer (MeV) software as described [45]. Semi-quantitative PCR assays were performed using the same cDNA templates and primers used for qRT-PCR. The template concentrates were normalised using the HbActin gene as a reference. The expression levels of each HbMPK gene were visualised using agarose gel electrophoresis of the corresponding PCR products.
Cis-Element Distributions in HbMPK Promoter Regions
To investigate the cis-regulatory elements in the promotor regions of HbMPKs, a 1500 bp region of the genomic sequence up stream of the start codon of each gene was retrieved from the Hevea genome database. These sequences were then submitted to online PlantCARE software [46] in order to predict putative cis-elements.
Identification and Characterization of MPK Members in Hevea
A total of 45 MPK candidates were identified by submitting the A. thaliana MPK (AtMPK) and O. sativa MPK (OsMPK) sequences to BLASTP queries against the Hevea genome database. Furthermore, 20 candidates that had both a protein kinase domain and a conserved MPK active site were determined to be HbMPKs after sequence analysis ( Table 1). The 20 putative HbMPKs were carefully reviewed, after which the coding sequences (CDS) of HbMPK15 (scaffold0949_294805) were manually corrected by investigating the downstream genomic sequences. The ORF lengths of the 20 HbMPKs ranged from 894 bp (HbMPK17) to 1848 bp (HbMPK12) and their encoded proteins consisting of 298 to 616 amino acids. The calculated molecular weights of these proteins ranged from 34.85 to 69.95 kDa, and their isoelectric points ranged from 4.81 to 9.28. These proteins were predicted to localise in the nucleus (16 members) or cytoplasm (4 members).
Previously obtained isobaric tag for relative and absolute quantitation (iTRAQ) based proteomic data of the Reyan7-33-97 rubber tree latex [47] were submitted to the Hevea genome database to determine the expression patterns of HbMPK proteins. As a result, 4 out of the 20 HbMPKs were validated by high-throughput tandem mass spectrometry (MS/MS) data using a threshold of 95% confident peptides ≥2; two of them are induced by ethylene stimulation (Table 2). The raw spectra of isobaric tag reporters ( Figure 1) and identified peptides that mapped to the HbMPKs are provided (Supplementary Figure S1). These data validated the above 20 putative HbMPKs at the proteomic level. Previously obtained isobaric tag for relative and absolute quantitation (iTRAQ) based proteomic data of the Reyan7-33-97 rubber tree latex [47] were submitted to the Hevea genome database to determine the expression patterns of HbMPK proteins. As a result, 4 out of the 20 HbMPKs were validated by high-throughput tandem mass spectrometry (MS/MS) data using a threshold of 95% confident peptides ≥2; two of them are induced by ethylene stimulation (Table 2). The raw spectra of isobaric tag reporters ( Figure 1) and identified peptides that mapped to the HbMPKs are provided (Supplementary Figure S1). These data validated the above 20 putative HbMPKs at the proteomic level. Figure 1. Proteomic validation of Hevea brasiliensis mitogen-activated protein kinases (HbMPK) expression patterns in rubber tree latex before and after ethephon treatment. Representative mass spectra indicate the signal intensities of isobaric tags for the validated MPKs. The spectra correspond to the peptides of "APELCGSFFSK", "DYVHQLR", "DLKPSNLLLNANCDLK" and "SFDFEQNALTEEQMK" for HbMPK12, HbMPK14, HbMPK16 and HbMPK19, respectively.
Genomic location analysis showed that HbMPK family members are distributed onto 20 individual scaffolds (Figure 2). The rubber tree was considered to have undergone two rounds of genome duplication events [4]. The expansion of the MPK gene family in Hevea was further investigated by analysing duplication events in MPK genes. Four duplication events involving 10 paralogues were identified (HbMPK3/HbMPK6, HbMPK8/HbMPK9/HbMPK15, HbMPK10/HbMPK12, and HbMPK11/HbMPK16/HbMPK19), which indicates that segmental duplication events play significant roles in MPK gene expansion in the Hevea genome. Genomic location analysis showed that HbMPK family members are distributed onto 20 individual scaffolds ( Figure 2). The rubber tree was considered to have undergone two rounds of genome duplication events [4]. The expansion of the MPK gene family in Hevea was further investigated by analysing duplication events in MPK genes. Four duplication events involving 10 paralogues were identified (HbMPK3/HbMPK6, HbMPK8/HbMPK9/HbMPK15, HbMPK10/HbMPK12, and HbMPK11/HbMPK16/HbMPK19), which indicates that segmental duplication events play significant roles in MPK gene expansion in the Hevea genome.
Multiple Sequence Alignment and Motif Distribution of HbMPKs
To investigate the sequence conservation of HbMPK proteins, multiple sequence alignment and conserved motif distribution analyses were performed. The results of the multiple sequence alignment showed that all identified HbMPKs contain the previously reported MPK-specific conserved domains: an N-terminal ATP-binding domain (P-loop), a phosphorylation catalytic loop (C-loop), and an activation loop (TEY or TDY). Notably, the HbMPKs that contained the TEY-type ( Figure 3A) activation loop also have an additional evolutionarily conserved C-terminal common docking domain (CD-domain), whereas the TDY-type ( Figure 3B) subfamily proteins do not, indicating a different evolutionary origin and functional diversity among these proteins.
Furthermore, 12 conserved motifs were recognised by the MEME program [36]. The distributions of the 12 conserved motifs for all 20 HbMPKs were highlighted ( Figure 4A). The MPKs were phylogenetically ordered to better understand their evolutionary relationships. Motifs 2, 5, 6 and 9 are conserved in each HbMPK and are organised in the exact same order. Motif 8 is specific to the MPKs of groups A and B, which have the conserved CD-domain. The conserved amino acid sequence information of these 12 motifs was also shown ( Figure 4B). Motifs 1, 3, 6, and 8 correspond to the conserved C-loop, P-loop, activation loop, and CD-domain, and were identified in the multiple sequence alignment (black boxes in Figure 4B).
Multiple Sequence Alignment and Motif Distribution of HbMPKs
To investigate the sequence conservation of HbMPK proteins, multiple sequence alignment and conserved motif distribution analyses were performed. The results of the multiple sequence alignment showed that all identified HbMPKs contain the previously reported MPK-specific conserved domains: an N-terminal ATP-binding domain (P-loop), a phosphorylation catalytic loop (C-loop), and an activation loop (TEY or TDY). Notably, the HbMPKs that contained the TEY-type ( Figure 3A) activation loop also have an additional evolutionarily conserved C-terminal common docking domain (CD-domain), whereas the TDY-type ( Figure 3B) subfamily proteins do not, indicating a different evolutionary origin and functional diversity among these proteins.
Furthermore, 12 conserved motifs were recognised by the MEME program [36]. The distributions of the 12 conserved motifs for all 20 HbMPKs were highlighted ( Figure 4A). The MPKs were phylogenetically ordered to better understand their evolutionary relationships. Motifs 2, 5, 6 and 9 are conserved in each HbMPK and are organised in the exact same order. Motif 8 is specific to the MPKs of groups A and B, which have the conserved CD-domain. The conserved amino acid sequence information of these 12 motifs was also shown ( Figure 4B). Motifs 1, 3, 6, and 8 correspond to the conserved C-loop, P-loop, activation loop, and CD-domain, and were identified in the multiple sequence alignment (black boxes in Figure 4B).
Evolution and Exon-Intron Organization of HbMPKs
To investigate the evolutionary relationships between HbMPKs and the MPKs from other plants, a phylogenetic tree was constructed using the protein sequences of 17 MPKs from the monocot model plant rice (O. sativa), 20 MPKs from the dicot model plant A. thaliana, 16 MPKs from the rubber tree relative cassava (M. esculenta), and 20 from Hevea in this study. These MPKs were divided into four groups according to phylogenetic analysis: TEY-type MPKs compose groups A, B and C, and TDY-type MPKs compose group D ( Figure 5). No species-specific clades were identified; however, groups A and B contain many more dicot MPK members than monocot MPK members. This finding indicates that duplication events may have occurred after the divergence of monocots and dicots. At the same time, group D contains 9 OsMPKs, 6 AtMPKs, 6 M. esculenta MPKs (MeMPKs), and 8 HbMPKs, which suggests multiple evolutionary and functional diversities for these MPK members. An A. thaliana-specific duplication event was identified in group C, and two
Evolution and Exon-Intron Organization of HbMPKs
To investigate the evolutionary relationships between HbMPKs and the MPKs from other plants, a phylogenetic tree was constructed using the protein sequences of 17 MPKs from the monocot model plant rice (O. sativa), 20 MPKs from the dicot model plant A. thaliana, 16 MPKs from the rubber tree relative cassava (M. esculenta), and 20 from Hevea in this study. These MPKs were divided into four groups according to phylogenetic analysis: TEY-type MPKs compose groups A, B and C, and TDY-type MPKs compose group D ( Figure 5). No species-specific clades were identified; however, groups A and B contain many more dicot MPK members than monocot MPK members. This finding indicates that duplication events may have occurred after the divergence of monocots and dicots. At the same time, group D contains 9 OsMPKs, 6 AtMPKs, 6 M. esculenta MPKs (MeMPKs), and 8 HbMPKs, which suggests multiple evolutionary and functional diversities for these MPK members. An A. thaliana-specific duplication event was identified in group C, and two potential Hevea-specific duplication events were identified in groups B (HbMPK11, 16 and 19) and D (HbMPK8, 9 and 15) by phylogenetic analysis (Figure 5). potential Hevea-specific duplication events were identified in groups B (HbMPK11, 16 and 19) and D (HbMPK8, 9 and 15) by phylogenetic analysis ( Figure 5). Gene structure divergence plays important roles in the evolution of gene families and can be used to assess phylogenetic relationships [48]. Analyses of the phylogenetic relationships and exon-intron distributions of both AtMPKs and HbMPKs revealed that most members in the same group shared very similar exon-intron structures ( Figure 6). This similarity provides evidence of MPK classification and evolutionary relationships between Arabidopsis and Hevea. Group A and B MPKs contain 4-7 exons of similar length, and 3-6 introns of variable size, whereas group C members have 2 parallel exons and only 1 intron. Group D shows a complex distribution of exons and introns. We also observed that the gene structures are conserved between HbMPK and AtMPK homologues. Additionally, similar gene structures were observed not only among paralogues, but also among homologues, which indicates the possible origin of these MPKs and evolutionary duplication events (AtMPK6 compared with HbMPK11/HbMPK16/HbMPK19, AtMPK16 compared with HbMPK8/HbMPK9/HbMPK15, and AtMPK20 compared with HbMPK10/HbMPK12). Moreover, we determined the Hevea-specific duplication relationships in the abovementioned paralogues by comparing the gene structure of MPKs to those of the related spurge family plant M. esculenta, and the result indicates that these duplicated MPKs are potentially involved in rubber tree-specific Gene structure divergence plays important roles in the evolution of gene families and can be used to assess phylogenetic relationships [48]. Analyses of the phylogenetic relationships and exon-intron distributions of both AtMPKs and HbMPKs revealed that most members in the same group shared very similar exon-intron structures ( Figure 6). This similarity provides evidence of MPK classification and evolutionary relationships between Arabidopsis and Hevea. Group A and B MPKs contain 4-7 exons of similar length, and 3-6 introns of variable size, whereas group C members have 2 parallel exons and only 1 intron. Group D shows a complex distribution of exons and introns. We also observed that the gene structures are conserved between HbMPK and AtMPK homologues. Additionally, similar gene structures were observed not only among paralogues, but also among homologues, which indicates the possible origin of these MPKs and evolutionary duplication events (AtMPK6 compared with HbMPK11/HbMPK16/HbMPK19, AtMPK16 compared with HbMPK8/HbMPK9/HbMPK15, and AtMPK20 compared with HbMPK10/HbMPK12). Moreover, we determined the Hevea-specific duplication relationships in the abovementioned paralogues by comparing the gene structure of MPKs to those of the related spurge family plant M. esculenta, and the result indicates that these duplicated MPKs are potentially involved in rubber tree-specific tissues and development processes (Supplementary Figure S2). However, the last duplication events, involving HbMPK3/HbMPK6, were observed in all of the investigated plant species investigated in this study (AtMPK1/AtMPK2 and MeMPK1/MeMPK2). We are interested in the potential functional diversity of Hevea-specific duplicated MPKs. events, involving HbMPK3/HbMPK6, were observed in all of the investigated plant species investigated in this study (AtMPK1/AtMPK2 and MeMPK1/MeMPK2). We are interested in the potential functional diversity of Hevea-specific duplicated MPKs.
To further investigate the expansion of the MPK gene family in Hevea, syntenic analysis of Hevea, M. esculenta, and A. thaliana MPKs was performed. The results were visualised using Circos software. We identified eight segmental duplication pairs in Hevea MPKs, but no tandem duplication events were detected, as our results were partly limited by the lack of chromosomal assembly information for rubber tree genome (Figure 7). Again, the duplication events described above were determined to be syntenic genes: HbMPK3/HbMPK6-MeMPK2, HbMPK10/HbMPK12-MeMPK20, HbMPK11/HbMPK16/HbMPK19-MeMPK6, and HbMPK8/HbMPK9/HbMPK15-MeMPK16-2. Modes of evolutionary selection can be estimated by the Ka/Ks ratio [41]. A Ka/Ks ratio > 1 indicates a positive selection, a Ka/Ks ratio < 1 indicates a purifying selection, and a Ka/Ks ratio = 1 indicates a neutral selection. The Ka/Ks ratios of the duplicated HbMPKs indicated that they all were subjected to purifying selection (Table 3). Furthermore, Tajima relative rate tests were conducted to investigate whether the HbMPK duplicates evolved at an accelerated rate following the duplication events. A statistically significant increase in evolutionary rate occurred between the HbMPK10/HbMPK12 duplicated pairs (Table 4), which indicates a potential functional divergence of these duplicated paralogues. To further investigate the expansion of the MPK gene family in Hevea, syntenic analysis of Hevea, M. esculenta, and A. thaliana MPKs was performed. The results were visualised using Circos software. We identified eight segmental duplication pairs in Hevea MPKs, but no tandem duplication events were detected, as our results were partly limited by the lack of chromosomal assembly information for rubber tree genome (Figure 7). Again, the duplication events described above were determined to be syntenic genes: HbMPK3/HbMPK6-MeMPK2, HbMPK10/HbMPK12-MeMPK20, HbMPK11/HbMPK16/HbMPK19-MeMPK6, and HbMPK8/HbMPK9/HbMPK15-MeMPK16-2.
Modes of evolutionary selection can be estimated by the Ka/Ks ratio [41]. A Ka/Ks ratio > 1 indicates a positive selection, a Ka/Ks ratio < 1 indicates a purifying selection, and a Ka/Ks ratio = 1 indicates a neutral selection. The Ka/Ks ratios of the duplicated HbMPKs indicated that they all were subjected to purifying selection (Table 3). Furthermore, Tajima relative rate tests were conducted to investigate whether the HbMPK duplicates evolved at an accelerated rate following the duplication events. A statistically significant increase in evolutionary rate occurred between the HbMPK10/HbMPK12 duplicated pairs (Table 4), which indicates a potential functional divergence of these duplicated paralogues. a The Tajima relative rate test was used to examine the equality of the evolutionary rate between rubber tree paralogues; b Mt is the sum of the identical sites in all three sequences tested; c M1 is the number of unique differences in the first paralogue; d M2 is the number of unique differences in the second paralogue; e If p < 0.05, the test rejects the equal substitution rates between the two duplicates and infers that one of the two duplicates has an accelerated evolutionary rate. paralogues; b Mt is the sum of the identical sites in all three sequences tested; c M1 is the number of unique differences in the first paralogue; d M2 is the number of unique differences in the second paralogue; e If p < 0.05, the test rejects the equal substitution rates between the two duplicates and infers that one of the two duplicates has an accelerated evolutionary rate.
Expression Profiling of MPKs in Different Tissues from Three Cultivated Varieties of Rubber Tree
To gain insights on the tissue-specific expression patterns of the 20 HbMPKs, the expression patterns of HbMPKs were visualised as a heat map and separated into three clusters ( Figure 8A). Members of cluster TI, including HbMPK11/19 (these paralogues could not be distinguished by gene-specific primers), HbMPK13, HbMPK6, HbMPK16, and HbMPK20, showed the highest expression level in latex, and the members in cluster TI were distributed in all groups. Members of cluster TII showed universal expression patterns in all four tissues; the highest expression levels usually occurred in photosynthetic tissues. Most members in cluster TII were distributed in Group D, only HbMPK3 was distributed in Group A. The remaining HbMPKs were barely expressed. The cluster TIII did not have the members in Group A. The semi-quantitative PCR results were also shown by the electrophoresis analysis of the PCR products ( Figure 8B). Notably, the MPKs in cluster TI (HbMPK11/19, HbMPK13, HbMPK6, HbMPK16, and HbMPK20), which are mostly expressed in the latex, also demonstrated different expression patterns among different rubber tree varieties, indicating that these genes might be involved in latex production processes in Hevea.
Expression Profiling of MPKs in Different Tissues from Three Cultivated Varieties of Rubber Tree
To gain insights on the tissue-specific expression patterns of the 20 HbMPKs, the expression patterns of HbMPKs were visualised as a heat map and separated into three clusters ( Figure 8A). Members of cluster TI, including HbMPK11/19 (these paralogues could not be distinguished by gene-specific primers) , HbMPK13, HbMPK6, HbMPK16, and HbMPK20, showed the highest expression level in latex, and the members in cluster TI were distributed in all groups. Members of cluster TII showed universal expression patterns in all four tissues; the highest expression levels usually occurred in photosynthetic tissues. Most members in cluster TII were distributed in Group D, only HbMPK3 was distributed in Group A. The remaining HbMPKs were barely expressed. The cluster TIII did not have the members in Group A. The semi-quantitative PCR results were also shown by the electrophoresis analysis of the PCR products ( Figure 8B). Notably, the MPKs in cluster TI (HbMPK11/19, HbMPK13, HbMPK6, HbMPK16, and HbMPK20), which are mostly expressed in the latex, also demonstrated different expression patterns among different rubber tree varieties, indicating that these genes might be involved in latex production processes in Hevea.
Expression Profiling of HbMPKs in Reyan7-33-97 under Different Abiotic Stresses
The genes of the MPK family are crucial abiotic stress responsive genes [49]. The expression patterns of HbMPKs under different abiotic stresses were analysed to further understand the functional diversity of HbMPK members. As shown in Figure 8, the 20 HbMPKs were grouped into three clusters according to their response to different abiotic stresses. The MPKs in cluster AI, especially HbMPK6, showed a weak or slightly reduced response to the abiotic stress treatments. Cluster AII included MPK family members that were barely expressed in leaves, either before or after abiotic stress. Cluster AIII contained the most HbMPK members, and showed different degrees of increased expression levels after abiotic stress ( Figure 8C,D). Most members in cluster AIII were distributed in Group D, which means the members in Group D may play roles in diverse abiotic stresses.
Expression Profiling of HbMPKs in Reyan7-33-97 under Different Abiotic Stresses
The genes of the MPK family are crucial abiotic stress responsive genes [49]. The expression patterns of HbMPKs under different abiotic stresses were analysed to further understand the functional diversity of HbMPK members. As shown in Figure 8, the 20 HbMPKs were grouped into three clusters according to their response to different abiotic stresses. The MPKs in cluster AI, especially HbMPK6, showed a weak or slightly reduced response to the abiotic stress treatments. Cluster AII included MPK family members that were barely expressed in leaves, either before or after abiotic stress. Cluster AIII contained the most HbMPK members, and showed different degrees of increased expression levels after abiotic stress ( Figure 8C,D). Most members in cluster AIII were distributed in Group D, which means the members in Group D may play roles in diverse abiotic stresses.
These expression pattern analyses of HbMPKs in different tissues and under different abiotic stresses suggested that MPK genes might have multiple functions in Hevea to adapt to various environmental changes (Figure 8). In addition, the duplicated paralogues showed very different expression profiles, indicating functional and evolutionary divergence emerged following the duplication events.
Analysis of cis-Elements in the Promotor Regions of MPKs in Hevea
The distribution of cis-elements in the promoter regions of HbMPKs was further investigated to demonstrate the regulatory patterns of the MPK family members, especially duplicated paralogues ( Figure 9). To facilitate this understanding, cis-elements were symbolised by capital letters in different colour; detailed classification and sequence information are listed (Supplementary Table S2). No significant positive correlations were observed between the distribution patterns of cis-elements and the expression patterns of HbMPKs, partly because of the lack of promoter sequence information for several genes. However, we did observe different cis-element organisational patterns in paralogues that exhibited different expression patterns in various tissues and/or under different abiotic stresses (HbMPK3-HbMPK6, HbMPK10-HbMPK12, and HbMPK11/19-HbMPK16). These results suggest the existence of different regulatory patterns and potential evolutionary fates in these paralogues. (Figure 8). In addition, the duplicated paralogues showed very different expression profiles, indicating functional and evolutionary divergence emerged following the duplication events.
Analysis of cis-Elements in the Promotor Regions of MPKs in Hevea
The distribution of cis-elements in the promoter regions of HbMPKs was further investigated to demonstrate the regulatory patterns of the MPK family members, especially duplicated paralogues ( Figure 9). To facilitate this understanding, cis-elements were symbolised by capital letters in different colour; detailed classification and sequence information are listed (Supplementary Table S2). No significant positive correlations were observed between the distribution patterns of cis-elements and the expression patterns of HbMPKs, partly because of the lack of promoter sequence information for several genes. However, we did observe different cis-element organisational patterns in paralogues that exhibited different expression patterns in various tissues and/or under different abiotic stresses (HbMPK3-HbMPK6, HbMPK10-HbMPK12, and HbMPK11/19-HbMPK16). These results suggest the existence of different regulatory patterns and potential evolutionary fates in these paralogues. Table S2).
Identification and Characteristics of HbMPK Genes and TheirAssociated Proteins
MPKs in various higher plants, including Arabidopsis, rice, maize, tobacco, poplar, and cassava, etc., have been systematically investigated [15][16][17][18][19][20][21][22][23][24]. However, for the first time, we verified 20 MPK genes in the most important natural rubber-bearing plant. Rubber tree MPK family members share three conserved domains and could be separated into four known groups; similar results were observed in other plants. Three of these groups have a TEY-type activation loop, and 1 has a TDY-type activation loop [14]; the TEY-type MPKs also contain additional C-terminal CD-domains (Figures 3-5). The genome of rubber tree is believed to have undergone two rounds of duplication events, resulting in similar numbers and characteristics of HbMPKs compared with those of Arabidopsis and M. esculenta.
To further verify the identified MPKs in rubber tree, previous iTRAQ data [47] obtained from rubber tree latex were introduced (Table 2; Figure 1). This is the first time that proteomic data were combined with genome-wide gene family studies for rubber-bearing plant research. The results support the gene family identification data and provide insights into the proteome patterns of MPK members in Hevea. Similar to genome research projects, proteomic data should be increasingly introduced in future gene-wide gene family studies. Although the proteomic data shown here did not provide the expression information about MPK proteins under abiotic stresses, we still believe that it is necessary to add proteomic evidence in the gene family investigations.
Evolutionary Divergence of Rubber Tree MPK Genes
The expansion of a gene family within a genome mainly occurs by three kinds of duplication events: tandem duplication of individual genes, segmental duplication of multiple genes, and background duplications that cannot easily be classified [50]. Given the limited chromosome assembly information of Hevea, the present study revealed that there is no or little contribution from tandem duplication to the expansion of the rubber tree MPK family (Figures 2 and 7).
One important purpose of the evolutionary study of gene families is to determine whether there are species-specific family members. This information can aid in understanding the functional divergence of a specific gene family [51]. Although no rubber tree-specific groups were determined by phylogenetic, gene structure, or gene syntenic analyses of the HbMPK family, we identified rubber tree-specific duplication events (Figures 5-7). Three Hevea-specific duplication events could be determined, but they were not observed for Arabidopsis or the spurge family plant cassava. These duplication events are HbMPK11/HbMPK16/HbMPK19-AtMPK6, HbMPK8/HbMPK9/HbMPK15-AtMPK16, and HbMPK10/HbMPK12-AtMPK20. The remaining paralogues demonstrate no species specificities, and HbMPK3/HbMPK6 duplication was observed in all of the investigated plants. These results indicate that all these duplication events occurred at different time points of evolution ( Figure 6 and Supplementary Figure S2).
Functional diversity caused by gene duplication might result in altered expression profiles and protein properties, and gene duplication is a major evolutionary driver for increasing the fitness of plants to new environment [52]. Furthermore, these paralogous pairs show tissue-and variety-specific expression patterns (Figure 8) as well as abiotic stress-specific patterns (Figure 8). One of the duplicated genes of three paralogous pairs shows global expression patterns in most tissues (HbMPK6, HbMPK8/15, HbMPK12, HbMPK11/19, and HbMPK16), whereas the other paralogue copies are barely expressed in root, stem, leaf, and latex tissues (HbMPK3, HbMPK9, and HbMPK10). Notably, the paraloguesHbMPK11/HbMPK16/HbMPK19 show universal expression patterns in different tissues (roots, stems, leaves, and latex) and under different abiotic stresses (osmotic, oxidative, salt and cold); however, HbMPK11/19 are mainly expressed in rubber latex (Figure 8). These results suggest that HbMPK16 might play similar role as its AtMPK6 homologue in Arabidopsis (which is universally expressed in various tissues, and is strongly associated with numerous abiotic stresses) [53] and that HbMPK11/19 somehow acquired new functions during the evolution of Hevea, and thus, are involved in nature rubber biosynthesis.
The selection pressure on HbMPK paralogues was estimated by calculating the ratio of Ka/Ks of rubber tree MPK paralogues. All Ka/Ks ratios were less than 0.1, which suggests that these paralogues were subjected to purifying selection, and mutations that alter amino acid sequences were not accepted during evolution (Table 3). These results are consistent with MPK family members playing very important roles during plant development, which requires highly conserved amino acid sequences to maintain protein functionality. Furthermore, the Tajima relative rates of Hevea gene pairs were calculated to evaluate the evolutionary rate between paralogues (Table 4). Notably, the duplicated gene pair HbMPK10/HbMPK12 has a p-value of 0.00555, which means that one of the two duplicates has a significant accelerated evolutionary rate. After combining and analysing the different expression patterns between HbMPK10 and HbMPK12 both in various tissues and under abiotic stresses, we assumed that the two genes are undergoing an accelerated separation in Euphorbiaceae plants, which suggests that they potentially play specific roles in Hevea.
Expression Characteristics of HbMPKs in Various Tissues and Cultivated Varieties under Different Stresses
MPK genes in groups A and B are more like the mainly expressed MPKs [54]. We observed different expression patterns of MPKs in vegetative tissues (roots, stems, and leaves) and in specialised rubber tree tissue (latex); these MPKs mainly belong to groups A and B.
Five MPK genes (HbMPK6, 11/19, 13, 16 and 20) were expressed in latex, according to the qRT-PCR and semi-quantitative PCR results ( Figure 8). Notably, HbMPK11/19, HbMPK13, and HbMPK6 showed significantly different expression levels across the high (Reyan8-79), medium (Reyan7-33-97), and low (PR107) yielding varieties of rubber tree. In particular, the expression levels of HbMPK11/19 and HbMPK13 in latex showed similar change patterns across the cultivated varieties, whereas HbMPK6 showed opposite patterns, which indicates that these genes may participate in the induction or repression of the biogenesis of natural rubber. The phosphorylation of functional proteins, including rubber elongation factor (REF), small rubber particle protein (SRPP), and osmotin, plays a very important role in natural rubber biosynthesis [55][56][57], and thus provides potential targets for HbMPK6, HbMPK13, and HbMPK11/19. In addition, two other MPKs, named HbMPK20 and HbMPK5, have shown very different expression patterns across the three cultivated varieties, the reason for which remains unclear.
The majority of the HbMPKs can respond to various abiotic stresses in this study ( Figure 8). The expression patterns under abiotic stress do not match the cis-element distributions in their promoter regions; however, the duplicated paralogues that have various cis-element distributions really exhibit completely different expression patterns both in different tissues and in different cultivated varieties under various abiotic stresses (Figures 8 and 9). This phenomenon supports the hypothesis that gene loss occurs very rapidly following whole-genome duplication, and that functional divergence can explain why duplicated genes escape extinction [58].
Conclusions and Prospects
In summary, based on our proteomic data and the released Hevea genome, a total of 20 Hevea MPK family members were characterised. Phylogenetic and syntenic analyses determined that 10 of these HbMPKs are paralogues. Evolutionary analysis showed that all the paralogues were subjected to purifying selection, and that only one duplicated gene pair (HbMPK10/HbMPK12) has an accelerated evolutionary rate. Furthermore, expression profiling of the 20 HbMPKs in the roots, stems, leaves, and latex of three rubber tree cultivated varieties showed very different expression patterns among the paralogues. The paralogues of HbMPK11/19 and HbMPK16 showed very different expression levels in the latex of various rubber tree varieties, which indicates functional divergence between these gene pairs. Significantly different cis-element distributions were observed in the promoter regions of the duplicated paralogues. Our data provided the first systematic and evolutionary aspects of the MPK gene family of the rubber-bearing plant Hevea, and suggest that members of duplicated gene pairs, such as HbMPK11/19 and HbMPK13, might play crucial roles in the biogenesis of natural rubber.
However, the limitation of transgenic technology in rubber tree and the lack of Hevea MPK-specific antibodies lead to a deficiency in functional validation in the present study. Moreover, additional investigations of several HbMPK paralogues (HbMPK8/HbMPK9/HbMPK15 and HbMPK10/HbMPK12) that showed interesting expression patterns should be conducted in the near future.
|
2017-11-27T08:16:15.514Z
|
2017-10-01T00:00:00.000
|
{
"year": 2017,
"sha1": "46a8e3f772e33c9f1d6d9fc5a8c86538b1bee680",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2073-4425/8/10/261/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "46a8e3f772e33c9f1d6d9fc5a8c86538b1bee680",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Biology",
"Medicine"
]
}
|
35824744
|
pes2o/s2orc
|
v3-fos-license
|
Effect of Helix aspersa extract on TNFα , NF ‐ κB and some tumor suppressor genes in breast cancer cell line Hs 578 T
© 2017 Pharmacognosy Magazine | Published by Wolters Kluwer ‐ Medknow 281 ABSTRACT Background: The garden snail, Helix aspersa, is a big land snail widely found in the Mediterranean countries. It is one of the most consumed species and widely used in zootherapy. Objective: The present study was carried out to investigate for the first time the first time the antitumor activity of an aqueous extract from Helix aspersa. Materials and Methods: The effect of H. aspersa extract was studied on a triple negative breast cancer cell line Hs578T. Firstly, the morphological changes and the mode of cell death induced by the extract have been evaluated by microscopy and acridine orange/ethidium bromide staining. The effect of the extract at dilution 0.1% and 1% was then tested on some genes, regulators of cell death and proliferation like tumor necrosis factor α (TNFα), NFκB, and the tumor suppressor genes P53 and PTEN. Results: Data demonstrate that the extract induces necrosis in tumor cells. It enhances significantly the expression of TNFα; mRNA levels were 20 and 10 times more important in treated cells compared to nontreated cells. NF-κB and PTEN were inhibited with the dilution 1% after 8 and 24 hours of treatment. P53 expression was further inhibited but only with the highest dose, after 4, 8, and 24 hours. Conclusion: Our results show that H. aspersa extract has an antitumor activity against Hs578T cells; it is a potent stimulator for TNFα and a good inhibitor for NF-κB.
INTRODUCTION
After cardiovascular disease, cancer is the second leading cause of death in the world. [1] The treatment is basically on synthetics and chemotherapeutics; these kinds of treatments cause various harmful side effects to human beings. [2] For this reason, the search for new natural and safe drugs is the aim of the different laboratories in the world. Plant and marine species are the major source of natural drugs; few researches have been interested in purifying bioactive molecules from other terrestrial consumed species like snails. Snails are a member of mollusk, the second largest phylum in the animal kingdom with about 100,000 living species. One of the most consumed species is the garden common snail, Helix aspersa; it is a small nutrient with high contents of proteins and minerals and low contents of fat and cholesterol. [3,4] Snails have been used in medicine since antiquity and are prepared by several methods; they are recommended for stomach pain, vertigo, nephritis, and respiratory and cardiovascular diseases. [5] Nowadays, lectins from snails are used as a marker for metastatic tissues in breast and colon cancers. [6,7] Initially, we have observed that the Helix aspersa extract had an antitumor effect on breast cancer cell line Hs578T. [8] This finding led us to investigate its mode of action and understand the mechanism involved. In particular, we focused on the expression of TNFα, a stimulator of the extrinsic pathway of apoptosis; [9] NF-κB, a cell proliferation regulator [10] and the tumor suppressor genes P53 and PTEN. [11] TNF α is a key cytokine that plays an important role in inflammation, immunity, and diverse cellular events. It modulates cell proliferation, necrosis, and apoptosis. [12] The major source of TNFα is activated macrophages; it can also be produced by other cells including fibroblasts, astrocytes, kuppfer cells, smooth muscles, keratinocytes, and tumor cells. It contributes to the metastatic process, invading the host tissues, penetrating to blood vessel's endothelium and establishing tumor cell growth at secondary sites. [13] NF-KB, a transcription factor that plays important roles in cancer development, appears to be involved in the regulation of cell proliferation, apoptosis control, angiogenesis promotion, and invasion/metastasis stimulation. [14] P53 is encoded by the TP53 gene, located at 17p13; it plays an important role in mediating cell response to various stresses, [15] mainly inducing or repressing a number of genes involved in cell cycle arrest, senescence, apoptosis, DNA repair, and angiogenesis. [16] Indeed, P53 mutation is Effect of Helix aspersa extract on TNFα, NF-κB and some tumor suppressor genes in breast cancer cell line Hs578T associated with more aggressive disease and worse overall survival, like breast and prostate cancers. [17] Next to P53, PTEN is the most common tumor suppressor to be lost or inactivated in human cancers like glioblastoma, prostate, and breast cancers. [18] The PTEN gene encodes a dual specificity lipid and protein phosphatase. PTEN modulates cell growth, migration, and survival by antagonizing the phospho-inositol 3 kinase (PI3K)/AKT signaling. [19]
MATERIALS AND METHODS
Cell culture: Human breast cancer Hs578T cell line was obtained from the laboratory of functional genomics and experimental pathology of the oncologic institute "Ion Chiricuta" (Cluj Napoca-Roumania). The cells were cultured at 37°C and humidified atmosphere of 5% CO 2 , in Dulbecco's Modified Eagle's Medium high glucose supplemented with 1% nonessential amino acids, 1% L-glutamine (200 mM), 1% gentamicin (10 mg/ml), 1% insulin, and 10% fetal bovine serum. All reagents were purchased from Sigma Aldrich, Germany.
Preparation of H. aspersa extract:
The H. aspersa snails were maintained fasting during at least 2 weeks in order to empty their gut. Before use the snails were washed in 3 or 4 bath of 10% Nacl solution (to completely eliminate their mucus). After removing the shell, the snails were homogenized. Three volumes of water per one volume of wet tissues were then added to the homogenate. The crude extract was filtered and centrifuged for 10 minutes at 5000 g. The supernatants were collected and sterilized using 22 μm Millipore filter and served as extract to treat cells. The final concentration of the extract was 0,33 g/l. Acridine orange/ Ethidium bromide staining: Cells were seeded for 24 hours in the presence of H. aspersa extract; at 0.1% and 1% dilutions of crude extract. Cells were stained by Acridine orange (AO)/ethidium bromide (EB), the dye mix for staining was 100 μg/ml AO and 100 μg/ ml EB in PBS (pH=7). [20]
RT-PCR
To study gene expression, cells were treated with the extract at different times 4, 8 and 24 hours. Total cellular RNA was isolated using TRiagen (Sigma, Germany) and converted to cDNA with The Random Hexamer Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnosis, Germany) following manufacturer's instruction. Quantitative RT-PCR was performed using TaqMan Master Kit and light cycler. All samples samples were run in triplicates and the β actin was amplified as an internal control. Primers used are presented in Table 1. Quantification of relative gene expression was done using the competitive threshold cycle C T method. C T values were averaged for triplicate wells and subtracted from corresponding C T of internal reference RNA to obtain ΔC T values. The averaged control ΔCT was subtracted from the experimental ΔC T to yield ΔΔC T . The fold change was calculated as 2 -ΔΔCT for experimental versus control. [21] ELISA Assay: The concentration of TNFα in culture medium was determined by ELISA assay and using TNFα human Boster immunoleader kit (Cliniscience France).
STATISTICS Data were expressed as mean ± SD from at least three separate experiments performed on triplicate samples. The differences between experimental conditions and controls were analyzed using test t (p.<.0.05 is considered statistically significant). Statistical analyses were carried out using Graph Pad Prism software (free trial).
Morphological identification of cell death
Cells were treated with two doses of H. aspersa extract, the first dose corresponded to the 1% dilution (1:100 dilution of filtered extract) which is approximately equal to IC 50 . [8] The second dose was 10 times less than the first one.
To determine if H. aspersa extract induces necrosis or apoptosis in breast cancer cell line Hs578T, cells were stained by AO/EB. AO/EB staining provides a reliable method to measure cells in different compartments of cell death; live, apoptotic, and necrotic cells were differentiated using fluorescence microscopy. AO (excitation: 502 nm; emission: 525 nm) permeates all cells and makes the nuclei appear green. It will also enter acidic compartments such as lysosomes and become protonated and sequestered. In low pH conditions such as this, AO will emit orange fluorescence (emission ~590 nm) when excited by blue light (475 nm). EB (exitation: 360 nm, emission: 590 nm) is only taken up by cells when cytoplasm membrane integrity is lost and stains nuclei in red. Therefore, live cells (showed in Figure 1A by a green arrow) have normal green nuclei. Apoptotic cells (showed in Figure 1C by a yellow arrow) have a bright green nucleus with condensed or fragmented chromatin and cells that have died from necrosis have an orange/yellow nucleus (showed in Figures 1B and 1C by a red arrow). The analysis of the fluorescence and comparison between the two doses [ Figures 1B and 1C] indicates that the extract induces necrosis. At 1%, the presence of some apoptotic cells has been also noted. Thus, the extract induces necrosis rather than apoptosis in Hs578T cell line.
Effect of H. aspersa extract on TNFα expression
TNFα mRNA (messenger RNA) levels were clearly elevated, after 4 and 8 hours, in Hs578T treated samples as indicated in Figure 2. After 4 hours, TNFα expression was 20 and 10 times more important with dilutions 0.1% and 1%, respectively, after treatment. Statistically and compared to controls, the increase of TNFα mRNA was highly significant (p < 0.001). After 8 hours, mRNA levels were decreased, but they remained significantly higher than those registered in control cells, especially with the 1%.dilution. After 24 hours of treatment with the two doses, TNFα expression has been significantly inhibited (p < 0.05).
Results from ELISA assay [ Figure 3] showed that TNFα concentrations rise after 8 hours of exposition to H. aspersa extract, and still higher after 24 hours. This data confirms that the extract stimulates TNFα production. In addition, the dose 0.1% has a more stimulatory effect.
Effect of H. aspersa extract on NF-κB expression
Treatment of Hs578T cells by H. aspersa extract at 0.1% decreases NK-κB expression after 4, 8, and 24 hours compared to control. At 1%, the effect of the extract was observed after 8 and 24 hours, we noted a significant decrease of NF-κB expression compared to nontreated cells [ Figure 4]. The highest dose (1%) had more inhibitory effect on NK-κB.
Effect of H. aspersa extract on PTEN expression
Treatment of Hs578T cells with H. aspersa extract resulted in the diminution of PTEN expression levels with the two doses 0.1 and 1%. A modest increase in PTEN expression was observed after 4 hours of treatment with the lowest dose (0.1%). The highest dose (1%), however, had more inhibitory effect (p < 0.05) [ Figure 5].
Effect of H. aspersa extract on P53 expression
From Figure 6, we note that, no significant change has been registered with the 0.1% dilution. At 1% the extract significantly decreases p53 mRNA levels after 8 and 24 hours of treatment.
DISCUSSION
The main aim of this study was to determine the mode of action of the aqueous extract from H. aspersa on breast cancer cells. The fluorescence staining shows that H. aspersa induces necrosis in Hs578T cells; at the same time it enhances the TNFα expression significantly. This cytokine had been demonstrated to stimulate both apoptosis and necrotic cell death, depending on cell type and treatment conditions. [22] The necrosis induced by our extract seemed to be mediated by high levels of TNFα. The extract has stimulated the expression and secretion of TNFα, in culture medium. Therefore, the cytokine stimulates its specific receptor and act by autocrine mode. The TNFα induced necrosis is a form of cell death called "necroptosis" that resulted from an incomplete execution of apoptosis program, in which caspases and NF-κB are inhibited. These characteristics are in accordance with our results because H. aspersa extract blocked the apoptotic process and the complete apoptosis has been revealed only in some cases. TNFα enhancement in our experiments may have resulted from the effect of H. aspersa lectins, and change of glycozylation at the surface of tumor cells. A lectin from Helix pomatia, a similar species of snails, has been shown to recognize diverse epitopes on tumor cells like O-linked α-N-acetylgalactosamine on Tn epitopes [6][7], integrin α6, transcription factors heterogeneous nuclear ribo-nuclear protein (HnRNPs), heat shock protein 27 (Hsp27) and enolase 1 (ENO1) in breast and colon tumor cells. H. pomatia lectin binding was also observed in the golgi apparatus in the T47D and MCF7 cells. [23] Indeed, a similar augmentation of TNFα expression had been registered with lectin extracted from mushroom. [24] Another explanation is that H. aspersa extract has a pro-oxidant activity and may act by generation of reactive oxygen species (ROS). A high level of ROS can lead to necrotic cell death while, a low level leads to apoptotic cell death. [25] Moreover, Sakon et al., 2003 have reported that TNF stimulation leads to accumulation of ROS, which is essential for prolonged mitogen-activated protein kinase (MAPK) activation and cell death. [26] TNF receptor-1 (TNFR1) has been shown to initiate necrotic cell death [27] and leads to the generation of ROS, which functions as second messengers in the necrotic cell death pathway. [28] Furthermore, our results show that H. aspersa extract is a good inhibitor of NF-KB. Recent studies have reported that inhibition of NF-KB alone or in combination with cancer therapies leads to tumor cell death or growth inhibition. [14] Li et al., 2003 have reported that the increased ROS generation might contribute to the induction of apoptosis and revealed that NF-KB activity was almost completely inhibited by preventing the degradation of IkBα. Additionally, the level of bcl-2 was decreased. [29] The P53 and PTEN downregulation induced by H. aspersa extract is the result of TNFα overexpression. Different contradictory effects of TNFα have been reported on P53 expression. It was shown to increase P53 expression with HT29 cells and to decrease or minimally change its expression in HCT116 cells. [30] Treatment of cancer cell lines with TNFα decreases PTEN expression and the overexpression of TNFα lowers PTEN expression via TNFα/ NIK/NF-kB pathway. [31] Currently, anticancer drugs are developed with the aim to maximize apoptosis. However, cancer cells eventually become resistant to these therapeutics and multiple mechanisms may contribute to resistance to apoptosis. So, necrosis is an alternative pathway with fewer mechanisms of resistance compared to apoptosis. Targeted necrosis has potential clinical utility because this cell death mechanism retains the cancer cell specificity of apoptosis and bypasses the apoptotic resistance by redirection into necrosis. [32] In addition, Olofsson et al., 2007 have shown that chemotherapy induces more necrotic than apoptotic cell death in breast cancer patients, and this necrotic response is associated with a better survival. [33] So the induction of necrosis by H. aspersa extract may be an interesting characteristic and can open a new perspective in cancer therapy.
CONCLUSION
Our results demonstrate for the first time that H. aspersa extract has a high cytotoxicity against breast cancer cells. The extract can be used as an associated treatment, in chemotherapy that enhances tumor sensitivity by downregulating BcL2 [8] and NF-κB; because tumor resistance is usually correlated with overexpression of these factors. Although the propriety of H. aspersa extract as an inducer of necroptosis has to be further clarified, the present study opens new perspectives in the search for new natural anticancer drugs.
|
2018-04-03T05:10:57.210Z
|
2017-04-01T00:00:00.000
|
{
"year": 2017,
"sha1": "26c880311ee1d9f33e88c3d486bf2ac2abe23280",
"oa_license": "CCBYNCSA",
"oa_url": "https://europepmc.org/articles/pmc5421427",
"oa_status": "GREEN",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "b9ea479b155a41a276101b0c2a9ec72fec231778",
"s2fieldsofstudy": [
"Biology",
"Medicine"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
250243129
|
pes2o/s2orc
|
v3-fos-license
|
Agricultural big data and methods and models for food security analysis—a mini-review
Background Big data and data analysis methods and models are important tools in food security (FS) studies for gap analysis and preparation of appropriate analytical frameworks. These innovations necessitate the development of novel methods for collecting, storing, processing, and extracting data. Methodology The primary goal of this study was to conduct a critical review of agricultural big data and methods and models used for FS studies published in peer-reviewed journals since 2010. Approximately 130 articles were selected for full content review after the pre-screening process. Results There are different sources of data collection, including but not limited to online databases, the internet, omics, Internet of Things, social media, survey rounds, remote sensing, and the Food and Agriculture Organization Corporate Statistical Database. The collected data require analysis (i.e., mining, neural networks, Bayesian networks, and other ML algorithms) before data visualization using Python, R, Circos, Gephi, Tableau, or Cytoscape. Approximately 122 models, all of which were used in FS studies worldwide, were selected from 130 articles. However, most of these models addressed only one or two dimensions of FS (i.e., availability and access) and ignored the other dimensions (i.e., stability and utilization), creating a gap in the global context. Conclusions There are certain FS gaps both worldwide and in the United Arab Emirates that need to be addressed by scientists and policymakers. Following the identification of the drivers, policies, and indicators, the findings of this review could be used to develop an appropriate analytical framework for FS and nutrition.
INTRODUCTION
Global hunger due to climate change (Schmidhuber & Tubiello Francesco, 2007), pandemics (Kuehn, 2020), rapid population growth, dietary changes, as well as limited natural resources (Dutilleul, 2012;Janssens et al., 2020) has increased, which will make accessing nutritious and affordable food difficult in the future (Janssens et al., 2020). Cleaner production is essential for sustaining this expanding need for food production, but all the natural resources are under threat (Wunderlich & Martinez, 2018). Big data is required to investigate food and nutrition security due to population growth, global hunger, and widespread food demand (Jin et al., 2020). Big data is a term that can be defined in various ways but always refers to a large amount of data. Data is frequently produced in large quantities from various sources, necessitating the development of new tools and methods, such as powerful processors, algorithms, and software, to handle it (Marvin et al., 2017). Big data applications can be found throughout the food supply chain (FSC), from farm to fork, and can maximize production and ensure all food security (FS) dimensions and measurements are included. Online databases, omics profiling, the internet, sensors (Bouzembrak et al., 2019), mobile phones, social media (SM), video monitoring (Subudhi, Rout & Ghosh, 2019), portable devices, and sensors using Internet of Things (IoT) tools (Pal & Kant, 2019), geographic information system (GIS), remote sensing (RS) (Strawn et al.), survey (Delphi rounds), and blockchain method (George et al., 2019;Shaikh et al., 2019) are just some examples of the global data sources. The source type varies by region based on availability, type of required data, relative experts, and scientific background. Furthermore, rapid population growth and rising food demand necessitate the use of quick, digital, and reliable data sources to ensure all FS dimensions and measurements are included (Fritz et al., 2019). However, these sources still require global attention to promote and enable their use in various environments and cultures. Consequently, stakeholders have identified five key challenges impeding the effectiveness of agri-food system collaborations (OECD, 2021): (1) a lack of political visibility and prioritization, (2) a lack of long-term investment in statistics and data, (3) challenges in political economy, (4) limited skills and experience in using such technologies, and (5) access gaps to new data sources.
Apart from this, the collected dataset should consider all FS dimensions and measurements (i.e., availability, access, stability, and utilization) (Jones et al., 2013) (Fig. S1). The data should include, but not be limited to, agricultural production, food loss and waste, food supply sufficiency, agricultural infrastructure, population growth and Democratic Domestic Product (DDP), agricultural food costs, household income, water availability and quality, soil properties and biodiversity, human health and diet, consumer behavior, climate change scenarios, demographic changes and stress, market access, imports, and common crops in the specific area. The data should cover FS based on the food system analytical framework of the Food and Agriculture Organization (FAO), which is dependent on food and nutrition security (FNS) economic, environmental, and social factors (Food and Agriculture Organization, 2018). However, most studies on FS have focused solely on availability and affordability, resulting in gaps in data collection and limitations in FS quantification (Nkunzimana et al., 2018). Various countries, including the Gulf Cooperation Council (GCC) and poor and low-income countries, suffer from a lack of agricultural and FS statistics, even though sound decisions are based on accurate data and information. Despite their efforts, such countries continue to face several limitations, including a lack of household and farm survey data, large and long-term data, and data analysis and processing (Food and Agriculture Organization, 2021). Current data collection is primarily focused on national sources with varying degrees of coverage and accuracy, using surveys and operational records such as trade data. That information is frequently disseminated as statistical output, with little or no interpretation or analysis. As previously stated, one significant gap in food and nutrition security data is the lack of indicators relating to the quantities of various foods consumed to determine the adequacy of nutrient intake at both household and individual levels.
Understanding the long-term drivers of FS and how they interact is necessary for policymakers to make informed decisions about today's policies for tomorrow's FS (Van Meijl et al., 2020). Model-based scenario analysis is widely regarded as the appropriate tool given the complexity and uncertainty of multi-dimensional FS (Godfray & Robinson, 2015). This article stated that more than 91 household models classified as statistical, optimization, Computational General Equilibrium (CGE), simulation integrated, and simulation biophysical models were used for FS and considered only the first dimension (availability) and did not cover the other dimensions (i.e., affordability, stability and utilization) (Nicholson et al., 2021a). Such lack of inclusion is primarily due to various factors, including dataset availability, the power and type of model used (dynamic, statistical, etc.), and the purpose of the study. A multi-model ensemble can solve this problem and may capture all FS dimensions (Kheir et al., 2021a;Kheir et al., 2021b;Martre et al., 2015), but this approach has received less attention thus far, resulting in a global gap in the use of modeling to address FS issues. System dynamics (SD) refers to a scientific framework for dealing with complex, nonlinear feedback systems. The book entitled 'Limits to Growth', published in 1972(Meadows et al., 1972 modeled for the first time the long-term risk of FS that would arise from the complex relationship between capital and population growth within the planet's limits, using the World3 System Dynamics model (Nicholson et al., 2021b). Furthermore, machine learning (ML) models can work well with large datasets and have many advantages not found in other models (Abiodun et al., 2018;D'Amore et al., 2022), but they have received little attention thus far and require much attention in global FS studies (Chamara et al., 2020). The ML techniques can be used to automatically collect data using statistical or computational models, which can aid in accurately identifying factors and improving performance (Okori & Obua, 2011). In various languages, ML has a significant impact on sentiment analysis and text classification (Marie-Sainte et al., 2019). Opinion mining and sentiment analysis are techniques for analyzing people's opinions, evaluations, sentiments, attitudes, and emotions from textual datasets (Onan, 2020c). There are numerous methods for text classification, opinion mining, and sentiment evaluation available in the literature (Onan, 2020a;Onan, 2020b;Onan, Korukoğlu & Bulut, 2016;Onan & Toçoğlu, 2021). However, the most widely used text classification techniques are lexicon-based, ML-based, and rule-based methods (Onan, 2021), with deep learning approaches not being used for feature selection or sentiment analysis, which necessitates much attention in FS studies. Consequently, while the integration of statistic, dynamic, ML, and deep learning models is very important in big data assessment and global FS studies, it has received less attention thus far, creating a gap that needs to be filled.
Regarding the position in the United Arab Emirates (UAE), unfortunately, there is a significant gap in large data sources, collection, and analysis. Furthermore, long-term investment in data statistics, digital skills, and sufficient data in science, technology, engineering, and mathematics (STEM) has been lacking. In addition, UAE lacks FS access, stability, and utilization of big data and methods and models based on FS dimensions. Therefore, a review study is required to identify the detailed background of big data and data analysis methods and models for FS on a global and regional scale to quantify the related gap and provide policy recommendations to fill it. Thus, this review screens the global and local big data and methods and models for FS analysis, highlights related challenges in the UAE, and suggests potential solutions.
SURVEY METHODOLOGY
Many authors have emphasized the importance of conducting a literature review because they consider it a valuable and qualified source of information that summarizes and adds to the body of knowledge in a particular field of study (Denyer & Tranfield, 2009;Knopf, 2006;Tranfield, Denyer & Smart, 2003). The best literature review should reveal, assess, and structure the relevant literature on the intended topic, as well as combine it with a critical analysis of various arguments in the literature (Denyer, Tranfield & Van Aken, 2008;Tranfield, Denyer & Smart, 2003). The goal of this study was to conduct a literature review to identify the global big data, methods, and models for FS and investigate how they can be used to improve FS levels globally, as well as in UAE (Denyer, Tranfield & Van Aken, 2008;Tranfield, Denyer & Smart, 2003). The least level of bias was ensured via a comprehensive literature inspection of the available published studies to provide an audit pathway from the decisions of the reviewers to the actions and conclusions (Munn et al., 2018;Tranfield, Denyer & Smart, 2003). Furthermore, selecting the research methodology required identifying, analyzing, and synthesizing the selected secondary data sources related to FS big data, methods, and models across a wide range of contexts and disciplines to provide a comprehensive understanding based on the fit to the review's specified questions. According to (Denyer & Tranfield, 2009;Munn et al., 2018;Tranfield, Denyer & Smart, 2003), producing good and comprehensive systematic reviews is crucial for driving research, developing new research baselines, and opening multiple pathways for future research. As a result, a systematic literature review research method was selected to achieve the research objectives. Based on the approaches described by Denyer & Tranfield (2009), Munn et al. (2018), Tranfield, Denyer & Smart (2003, we conducted the review through five steps to ensure replicability and transparency, as detailed in Fig. 1. The research began with the formulation of a research questions with specific characteristics, such as being purposeful and specific. The scope and focus of the review were then defined. The goal of this study was to conduct a systematic literature review to identify the big data, methods, and models of FS in the global and UAE contexts. To answer the main research questions, this article provides a critical review of the existing literature published in Scopus and Web of Science databases (Martín-Martín et al., 2018). The following topics have been explored in this review of literature: (1) data extraction tools, (2) data format and infrastructure, (3) potential and limitations of agricultural big data (AgBD), and (4) FS methods and models. Thus, this article provides a reference for policymakers and practitioners, as well as a roadmap for future research, by highlighting the concerns in the areas mentioned above. The second step involved creating a specific research criterion to ensure that the research sources chosen were sufficient and comprehensive enough to capture all the major points that adequately answer the research questions (Denyer & Tranfield, 2009). The necessity of understanding big data for FS in both the global and UAE contexts, with a strong emphasis on avoiding any source of bias during the selection process, was the key research gap that drove this study. As a result, the databases Scopus and Web of Science were used (Martín-Martín et al., 2018). Big data, methods, models, FS availability, FS access, FS stability, FS stability, and food infrastructure were among the keywords used. The keywords were chosen after a thorough examination of the most relevant concepts in the literature that affect each of the four FS dimensions. In July 2021, the research sources were chosen, and the title, abstract, and full-text searches for keywords were enabled. To find the available literature, several keywords were identified (Cooper et al., 2018). Primary and secondary keywords were used in the search strings. The purpose of using multiple strings was to cover as many articles as possible that dealt with the topic of FS or any of its four dimensions. The review was then subjected to specific exclusion and inclusion criteria to produce high-quality evidence (Tranfield, Denyer & Smart, 2003). To ensure that the review has a high quality, a reasonable number of articles were selected for in-depth analysis based on a set of exclusion and inclusion criteria (Fig. 1). Within the time frame (2010-2021), only peer-reviewed journal articles written in English were included in the review. Strict selection criteria were applied to the first search pool to maintain research transparency and ensure the selection of relevant material that answers the research questions (Kelly, Sadeghieh & Adeli, 2014;Xiao & Watson, 2017). After removing duplicated articles from both databases, a total of 130 articles were chosen for the review.
The fourth step entailed analyzing the selected 130 articles individually, summarizing and listing all big data, methods, and models for FS analysis, then synthesizing the extracted information from all sources to create new knowledge (framework), listing the similarities between all resources, and extracting the major insights globally and within the UAE context (Denyer & Tranfield, 2009;Tranfield, Denyer & Smart, 2003). The models and methods for FS analysis in each of the 130 articles were summarized using Microsoft Excel. The aggregative approach was then used for synthesis. The findings section includes a detailed report of answers to the following research questions: (1) data extraction tools, (2) data format and infrastructure, (3) potential and limitations of AgBD, and (4) FS methods and models. After that, the synthesis process was used to create a comprehensive framework that models big data and FS methods and models.
DATA EXTRACTION TOOLS Data extraction tools (global context)
The literature review showed different sources of data extraction, including online databases, smartphones, the internet, sensors, omics, social media (SM), Internet of Things (IoT), geographic information system (GIS), satellite images, web mining, the Food and Agriculture Organization Corporate Statistical Database (FAOSTAT), governmental dataset, statistical yearbooks as well as blockchain technology (Bouzembrak et al., 2019;George et al., 2019;Marvin et al., 2017;Pal & Kant, 2019;Strawn et al.;Subudhi, Rout & Ghosh, 2019). Online databases used widely in food safety (Jin et al., 2021;Marvin et al., 2017) covered only one dimension of FS (utilization) and neglected the other dimensions.
Smartphones are widely used in agriculture due to their ability to collect data, ease of mobility, which corresponds to the nature of farming, and low cost (Mendes et al., 2020;Pongnumkul, Chaovalit & Surasvadi, 2015;Thar et al., 2021). Nowadays, more than two billion people worldwide use smartphones, and this number is rapidly increasing, allowing the use of smartphones as important data sources in agriculture and FS (eMarketer, 2014). The numerous built-in sensors are among the factors that improve the smartphone's ability to assist users with various tasks. Cheap smartphones may be a viable option for farmers who lack access to current agricultural information (e.g., market, weather, and crop disease news) and assistance from agricultural experts and government extension workers (Wolfert et al., 2017). Smartphones have recently been used in agriculture for various purposes, including food safety (Alfian, Syafrudin & Rhee, 2017;Shan et al., 2020;Ye et al., 2020), protein content determination (Silva & Rocha, 2020), food contaminant detection (Liu et al., 2017), weather and climate change reporting (Caine et al., 2015), as well as for agricultural and rural development (Donovan, 2017). Smartphones are the most important tools for receiving and recording terminal data (Guo et al., 2019). However, based on the literature, we observed that very little attention had been paid to understanding the various types of information communicated via smartphones, how farmers access this information, and the possible factors influencing the use of smartphones. Furthermore, smartphone applications did not cover all dimensions of FS (i.e., availability, access, utilization, and stability), necessitating a great deal of attention on the global, national, and individual levels.
SM sites are websites that allow users to create profiles (Hether, Murphy & Valente, 2014), share content, and engage in discussions to facilitate communication and community engagement (Bertrand et al., 2021). SM has been widely used to collect food safety data (Wang et al., 2016). Making intelligent decisions based on social big data refers to the techniques, technologies, systems, and platforms that help organizations better understand their data and make better decisions (Wang et al., 2016). FS-related discussions, opinions, and online questionnaires can be collected using SM platforms such as Facebook, Twitter, and YouTube (Soon, 2020). Web mining is a popular method for collecting and analyzing SM data. By analyzing customer sentiments and opinions, SM data could be used to improve client behaviors, raise public awareness, and understand public perceptions of FS (Yuan et al., 2020).
RS data can be used in agriculture for monitoring crop growth, development, and harvesting, and improving the existing monitoring systems, all of which contribute to improved agricultural product quality (Friedl, 2018;Singh et al., 2020). Data of RS images in the European Union (EU) could be accessed by Sentinel-2 satellites for various applications in agriculture and FS (Kussul et al., 2020;Phiri et al., 2020). The FAO GeoNetwork and RS database include grids and layers for classifying soil, water, and climate for monitoring food safety and security (Jin et al., 2020). Hattenrath-Lehmann et al. (2018) used RS as an early warning system for shellfish safety, while the US Department of Agriculture used them to detect food contamination (https://cris.nifa.usda.gov/). However, using the RS approach in big data for other FS dimensions such as availability, access, and stability has received less attention thus far, necessitating a great deal of attention on a large scale from stakeholders.
IoT is the interconnection of devices, sensors, machines, and computing devices through internet mediums (e.g., Wi-Fi, Bluetooth and Radio Frequency Identification (RFID)). This technology has the potential to make the food chain more efficient, safer, and sustainable in the near future. Kaur (2021) modeled sustainable FS based on IoT technology and determined how to design a long-term FS system in India, where the government ensures FS for all through a public distribution system (PDS). The study also made a novel attempt to incorporate IoT into the design of the PDS to ensure FS, with IoT factors being modeled using Total Interpretive Structural Modeling (Fuzzy-TISM). FS can be ensured using IoT as it provides traceability, transparency and accountability, decreasing food waste and ensuring food quality from harvest to consumption (Nukala et al., 2015). For more accurate results, IoT could be combined with technological enablers such as artificial intelligence, robotics, blockchain, and RFID. The use of these technologies will help reduce food waste and enable better planning of distribution networks, lowering the overall supply chain carbon emissions (Irani & Sharif, 2016). Various studies have investigated the importance of IoT in FS, but they have focused only on some dimensions, implying that all FS dimensions require much attention. Ding et al. (2014) proposed a conceptual model to investigate the interdependencies between different functions and information shared in FSC. Fan et al. (2015) proposed a big data analytics-based algorithm to improve crop yield prediction accuracy. Masiero (2015) emphasized the importance of digitizing the food distribution function and the role of e-governance in preventing food fraud in Kerala. The role of information sharing in fresh FSC was examined by Nakandala et al. (2017), who identified the information needs of various supply chain entities. Table 1 summarizes various enabling factors for an IOT-driven sustainable FS system culled from the literature. However, understanding the relationships and their effect among different technologies is critical for designing an IoT-driven FS system that is also sustainable. The FS system is a multi-level, multi-stakeholder problem. The integration of various enabling factors is difficult to establish. Furthermore, the impact of one relationship may differ from that of another. As a result, the magnitude of the impact is also a crucial factor to consider. From the standpoint of policymaking and implementation, such factors must be structured in a conceptual model to ensure long-term FS. Moreover, such technologies with attributed enablers should cover all dimensions of FS, rather than just one; thus, a gap needs to be filled from both the global and stakeholder perspectives.
Data extraction tools (The UAE context)
Despite the lack of data for measuring FS at the household or individual levels, several data sources are available to analyze and monitor the UAE's FS progress at the national level. For example, data on most of the Suite of FS Index indicators for the UAE are available at FAOSTAT (http://www.fao.org/faostat/en/#data/FS). These statistics are mostly available as three-year averages. The Economist Intelligence Unit (EIU) publishes the Global Food Security Index (GFSI) score for the UAE every year as part of its multi-country FS monitoring. The available data from EIU and FAO on the FS dimension indicators used to construct the GFSI and Suite of FS Index can be used as inputs for deriving other simple and multi-dimensional measures of FS.
Data on the UAE's Food Balance Sheet can also be obtained from the FAOSTAT (http://www.fao.org/faostat/en/#data/FBS). The dataset contains information, among others, on food supply (kcal/capita/day), protein supply quantity (g/capita/day), domestic production, import and export, feed and other non-food uses, food stock variation, tourist consumption, and food losses. Furthermore, data on food price indices and food inflation can also be retrieved from the FAOSTAT (http://www.fao.org/faostat/en/#data/CP).
Similar datasets at disaggregated levels can also be obtained from national offices (e.g., for Dubai, it can be obtained from the Dubai Municipality). The various organizations within the UAE, such as the Ministry of Food Security, Ministry of Health and Prevention, Dubai Municipality, Abu Dhabi Agriculture and Food Safety Authority, and the Federal Competitiveness and Statistics Centre, could be consulted for obtaining a variety of data that can be used as inputs for estimating some of the FS indicators.
Delphi survey rounds could be considered an effective data source (Allen et al., Markou et al., 2020) for collecting the required data to validate the analytical framework and quantify sustainable FS in the UAE.
Parameter Reference Role in sustainable food security
Yield prediction based big data Engen et al. (2021), Fan et al. (2015 and Kamath et al. (2021) Assist in the procurement process and the distribution of food resources across different regions. Ensure better quality control, and higher yield Sharing information-based channels Singla, Nishu & Deepika (2020) and Wolfert et al. (2017) Better supply chain coordination is aided by information sharing. It also helps supply chain partners build trust. Talavera et al. (2017) The temperature can be adjusted depending on the type and quantity of stock in the refrigerator.
DATA FORMAT AND INFRASTRUCTURE
FS data can be unstructured or structured and stored in various formats, including TXT, JSON, and CSV. For instance, Singh, Shukla & Mishra (2018) collected SM data from Twitter in JSON and TXT formats, and then implemented the parsing method to covert JASON data to CSV data. There are also various formats for big data, such as raster and vector formats (SHP, TIF, CN, and NetcDF) (Ghiringhelli et al., 2017;Limbachiya & Gupta, 2015). On the other hand, Song et al. (2020) stored the data in relational databases with different attributes as a list of rows. Alfian, Syafrudin & Rhee (2017) used NoSQL and SQL databases to store IoT-generated sensor data with a large unstructured format and continuous data-generation characteristics. They also developed a real-time food quality monitoring system that employs sensor data from a smartphone and stores it in the MongoDB database. Supercomputing and cloud computing are two major components of data infrastructure (Yang et al., 2017b). To address the challenges associated with big data, supercomputing must be considered. The United States has long been committed to supercomputing to facilitate knowledge exchange between the Exascale Computing Project and the industrial user community (Witze, 2014). The development of supercomputing infrastructures is also a priority for the EU. So far, the EU has built eight supercomputing centers to enhance bioengineering applications (Butler, 1999). Tianjin, Jinan, Changsha, Shenzhen, Guangzhou, Wuxi, and Zhengzhou are China's seven national supercomputing centers. The Chinese government has created a food safety traceability platform (Berti & Semprebon, 2018) that collects 31 provincial food traceability data and connects national supercomputing centers. Its goal is to achieve food traceability from farm to fork while also providing services to food producers, such as food traceability, security, and oversight.
To enable big data research, several cloud computing infrastructures need to be developed (Yang et al., 2017a). In 2019, the EU Food Nutrition Security Cloud project (https://cordis.europa.eu/project/id/863059) aimed to integrate European research infrastructure by bringing together FNS data to address diet, health, and consumer behavior, as well as sustainable agriculture and bioeconomy. The Guizhou Food and Drug Administration in China released the food safety cloud system in 2014. It has now been transformed into an intelligent food safety supervision system, an internet plus inspection system, a traceability certification system, and a big data platform for government enterprises, testing institutions, and other social age organizations (Tao et al., 2018). Despite the importance of supercomputing and cloud computing infrastructures as distinct big data environments, the Middle East and North Africa (MENA) and GCC regions remain uninterested, necessitating significant attention to assist them in addressing food insecurity.
AGRICULTURAL BIG DATA (POTENTIAL, CURRENT STATUS, AND LIMITATIONS)
To meet the demands of the rapidly growing population, which is expected to reach nine billion people by 2050 (World Population Prospects, 2015), agricultural production and FSCs must be optimized by producing and delivering efficient food, feed, fiber, and fuel (Abe, 2017;Asseng et al., 2018;Asseng et al., 2019). This goal has become more difficult to achieve due to urbanization, climate change (Ali et al., 2022;Ali et al., 2020;Ding et al., 2021;Kheir et al., 2019), and water scarcity (Kheir et al., 2021b). AgBD will be a key component of the second green revolution, which will be required to meet the demands of the growing population. Furthermore, the crop growth simulation modeling approach has been proven to be a useful tool for determining the impact of climate uncertainty on crop yields (Asseng et al., 2013;Ejaz et al., 2022;Shoaib et al., 2021). Many countries and commodity markets are already using AgBD to detect supply chain disruptions in commodity crops like wheat, rice, corn, and soybean (Bock & Kirkendall, 2017;George Hanuschak, 1993;Rosenzweig et al., 2013). Precision agriculture has progressed as a result of advancements in RS data collection, such as improved spatial and temporal resolution, spectral resolution, and a variety of sensor platforms (e.g., satellite, aerial, and ground-based) (Mulla, 2013). Precision agriculture recently demonstrated a significant increase in crop yield production (Loures , 2020;Singh et al., 2020). Spatial data mining techniques (e.g., hotspot detection) can be used with AgBD to identify crops produced in small geographic areas or a set of regions that are vulnerable to climate change and natural disasters (Jiang & Shekhar, 2017;Shekhar, Feiner & Aref, 2015;Vatsavai et al., 2012;Xie et al., 2017). Furthermore, consumer datasets and market manner can be used to improve food access and nutritional outcomes, and geo-social media can be used to detect and control food-borne illness outbreaks in real-time. AgBD could help agricultural decision-makers in four ways: descriptive, prescriptive, predictive, and proactive (Shekhar et al., 2017). The goal of the descriptive axis is to use AgBD data to characterize spatial and temporal variability in soil, land cover, crop, and weather characteristics, as well as to identify stressors, traits, and infectious disease risk factors that need to be better managed. The prescriptive way is to look for the required innovations for farm management. The predictive axis is a predictive analysis that uses historical datasets and integrated soil, crop, weather, and market models to forecast outcomes like crop yields and food insecurity. Predictive analytics can also be used to improve decision-making to forecast the spread of infectious agents and limit their impact on crops and livestock. Finally, the proactive axis includes crop development and stress observations from multiple farms across large regions and time scales. The current state of AgBD can be divided into two categories: public and private data. Some examples of public AgBD are summarized in Table 2. Big data differs from one region to another based on various factors, including but not limited to the data availability, capacity building, and target of the study. Exploring sustainable FS necessitates detailed data covering all aspects of FS, putting pressure on decision-makers and scientists to initiate and prepare the necessary big data. To assist in filling this void, the big data paradigm should employ techniques, paradigms, and decision-making technologies, as illustrated in Fig. 2.
METHODS AND MODELS USED FOR FS ANALYSIS
FS has four dimensions: access, availability, stability, and utilization; thus, incorporating such dimensions into agricultural models necessitates careful consideration and deep efforts. There have been few studies that have linked agricultural models with FS dimensions and indicators in order to understand evolving intertemporal dynamics and assess the effects of agricultural system intensification (Laborte et al., 2009;Laborte, Van Ittersum & Van den Berg, 2007;Marín-González et al., 2018;Morris, 2003;Nicholson et al., 2021a). However, such studies have focused only on a few FS indicators, such as household outcome, and ignored other dimensions and indicators. Therefore, we reviewed more than 1,200 related articles on FS modeling at the household and regional levels to assess the frequency of use of various FS indicators and make future recommendations to close this gap. Optimization models were used in FS but only on a few indicators of food availability (Amede & Delve, 2008). Crop simulation models are used to predict crop yield as an indicator of food availability, either as regression models (Beyene & Engida, 2016;Bharwani et al., 2005) or as complex biophysical models (Lázár et al., 2015). For food consumption expenditures, other sophisticated models, such as Holden & Shiferaw (2004) and Louhichi & Gomezy Paloma (2014) were used. Scopus database was screened for approximately 1,250 articles related to household FS models, and 130 articles were reviewed for which FS indicators were summarized (Table S1). We also looked for studies that looked at the determinants of dietary diversity at the individual level (most commonly, among young children or women) or at the household level. Dietary diversity, or the number of different foods or food groups in one's diet, has been linked to several measures of household socioeconomic status that are frequently used as indicators of food insecurity (Jones et al., 2013). As a result, dietary diversity is frequently used as a proxy and stand-alone indicator of household food insecurity. We searched Google Scholar for relevant studies that provide empirical evidence about the determinants of the Household Dietary Diversity Score (HDDS), Household Food Insecurity Access Scale (HFIAS), and Food Insecurity Experience Scale (FIES). For this purpose, we used the following items in search: diet diversity determinants (130 articles), household FS determinants (870 articles), and experience scale of food insecurity (250 articles). From the 1,250 articles, 130 articles were reviewed. Figure 3 shows the network and associations of different models used for FS. It was found that statistical models were the most prevalent, and all models covered only two FS dimensions: availability and access. The detailed descriptions of these models, including type, classifications, the related references, and calculations of FS dimensions are presented in Table S1. Even though, many models and methods are used to investigate global FS, there are insufficient related studies in the GCC, particularly in the UAE, requiring much attention using the best tools to achieve most FS dimensions.
CONCLUSIONS
Despite the importance of big data tools in FS, some challenges must be addressed first (Wang et al., 2016). The most common challenges for FSC-related data, according to most experts, are data quality, accessibility, findability, reusability, interoperability, and a lack of standardization. Farmers, for example, use a variety of farm management systems, making standardization of farm management data (such as variable names) a challenge. Because of the lack of standardized communication protocols, the data produced by IoT devices today can be difficult to interpret, communicate, and share, which may be one of the reasons for the limited adoption of IoT technology in food safety (Bouzembrak et al., 2019). Handling big data issues is difficult and time-consuming, requiring a large computational infrastructure to ensure timely data processing and analysis. Even though many organizations have adopted cloud computing as a solution, research on big data in FS using cloud computing technology is still in its infancy. Scalability, availability, data integrity, security, privacy, and legal issues are just a few of the research challenges that have yet to be fully addressed globally and in the UAE.
Statistical, optimization, CGE, simulation integrated, and simulation biophysical models and methods were used globally. However, it was discovered that such models only covered a subset of FS dimensions, namely availability and access, while recording limitations with other dimensions. In the future, this will necessitate the use of a multi-model approach to investigate FS because it will cover most FS dimensions while achieving higher accuracy. Thus, currently, there are some FS gaps in the global and UAE contexts that require significant attention from scientists and decision-makers. The global gaps could be summarized as follows: limited global dissemination of big data digital sources, lack of political visibility and prioritization, lack of long-term investment in data and statistics, lack of coordination and political economy challenges, limited access to new data sources, utilization and stability dimensions were not covered well, model complexity and uncertainty of multi-dimensional FS, limited studies on multi-model approach, and deep learning approach not being used. In the UAE context, in addition to the gaps mentioned in the global context, limited big data sources, lack of long-term investment in data and statistics, insufficient investment in agricultural research, insufficient studies in STEM, lack of modeling studies, and lack of ML and deep learning (DL) in data collection and
Author Contributions
• Khalil A. Ammar conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the article, and approved the final draft.
• Ahmed M.S. Kheir conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the article, and approved the final draft.
• Ioannis Manikas analyzed the data, authored or reviewed drafts of the article, and approved the final draft.
Data Availability
The following information was supplied regarding data availability: The raw data is available in the Supplemental File.
Supplemental Information
Supplemental information for this article can be found online at http://dx.doi.org/10.7717/ peerj.13674#supplemental-information.
|
2022-07-04T05:06:58.580Z
|
2022-06-29T00:00:00.000
|
{
"year": 2022,
"sha1": "e462e70e03f7e56dcc2226886efa3d9da6b57914",
"oa_license": "CCBY",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "e462e70e03f7e56dcc2226886efa3d9da6b57914",
"s2fieldsofstudy": [
"Agricultural and Food Sciences",
"Computer Science"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
260286962
|
pes2o/s2orc
|
v3-fos-license
|
Retrospective evaluation of percutaneous 3D-navigated screw fixation for fragility fractures of the sacrum: technical notes and four-year experience
The incidence of fragility fractures of the sacrum is increasing due to demographic changes. In this study, we introduce the 3D-navigated monoportal percutaneous sacroiliac screw fixation (PSS) as a technical advancement for treating fragility fractures of the sacrum. We included all patients who underwent the 3D-navigated monoportal PSS for fragility fractures of the sacrum. The fractures were classified using the Fragility Fractures of the Pelvis score (FFP). We provide a step-by-step illustration of the surgical technique. The objective of this study was to assess the feasibility and safety of the investigated technique. Forty-six patients (36 female, 10 male) with a median age of 81.5 years were included in the study. The fracture classification revealed 23 FFP2 (50%), 5 FFP3 (11%), and 18 FFP4 (39%) fractures. In 35 cases (76%), only transsacral screws were implanted in S1 and S2, with an average incision-to-suture time of 52.6 min. The remaining eleven patients underwent additional anterior pelvic ring fixation, lumbar instrumentation, or kyphoplasty. There were no instances of nerve root, vascular, or pelvic organ injuries. The median postoperative in-hospital stay was six days. Out of the 36 patients who were followed up, four patients required revision surgery due to screw loosening. No significant risk factor for screw loosening was identified in the multiple regression analysis. The presented monoportal PSS technique for fragility fractures of the sacrum is a promising minimally invasive approach with a low complication rate and excellent short-term outcomes.
The incidence of age-related diseases, including fragility fractures of the pelvis, is on the rise due to demographic changes and an aging population 1,2 . Unlike high-energy pelvic fractures in younger individuals that can lead to severe blood loss and pelvic organ injuries, geriatric population may sustain pelvic ring fractures with minor trauma [3][4][5][6] . Concomitant hemorrhage or pelvic organ injuries are rare; however, these injuries often result in debilitating back pain. Therefore, the primary treatment goal for this patient group is to restore functionality and prevent immobility, which can lead to complications such as ulcers, skeletal demineralization, and muscle atrophy. Various osteosynthesis procedures are available depending on fracture morphology, instability degree, and pelvic anatomy. Given the age and comorbidities commonly found in the geriatric population, minimally invasive techniques with shorter incision-to-suture time and reduced soft tissue trauma are preferred over conventional techniques 7 . Additionally, the choice of osteosynthesis technique and materials is crucial due to the frequently encountered osteoporotic and poor-quality bone in this population 8,9 . The use of 3D navigation has shown improved accuracy in screw placement compared to 2D navigation or fluoroscopic control 10 . Any potential drawbacks, such as increased surgical time due to intraoperative imaging and computer navigation, can be mitigated by routine utilization of these technical devices with an optimized workflow.
Fragility fractures of the sacrum commonly involve both lateral masses of the sacrum and often coincide with fractures of the anterior pelvic ring. Therefore, a bilateral osteosynthesis is necessary for operative treatment of these fractures. Traditionally, a biportal percutaneous sacroiliac screw (PSS) fixation technique at the S1 and/ or S2 level is chosen as minimal invasive approach. However, biomechanical finite element model studies have demonstrated that transsacral screws provide superior stability compared to bidirectional sacroiliac screws for both unilateral and bilateral sacral fractures 11,12 . Furthermore, 3D statistical models have identified low bone mass in the sacral vertebral bodies of patients with fragility fractures of the sacrum, suggesting that transsacral osteosynthesis procedures with improved screw-bone contact may be advantageous 8,13 . Clinical data evaluating the feasibility and perioperative complication rate of 3D-navigated monoportally inserted transsacral screws is limited, although several single-center studies have reported promising results [14][15][16] .
The preliminary hypothesis of this study was that the 3D-navigated monoportal percutaneous sacroiliac screw fixation (PSS) technique is a safe and effective minimally invasive approach for treating fragility fractures of the sacrum, with low complication rates and favorable short-term outcomes. To investigate this, we conducted a retrospective study of our four-year experience with this technique and provided a step-by-step description of the procedure. Relevant surgical parameters, including screw positioning accuracy and perioperative complication rates, are evaluated.
Material and methods
Ethical approval. The ethical committee of the Medical Association of Hessen, under research no. 2021-2558-evBO, approved this study. All surgical procedures were conducted in accordance with Good Clinical Practice Guidelines. Informed consent was obtained from all subjects and/or their legal guardian(s).
Patient selection and data collection. In this retrospective cohort study, we examined the data of all patients who underwent 3D-navigated monoportal osteosynthesis as a treatment for low-energy fractures of the sacrum between 2018 and 2021. Overall, 124 patients underwent operative treatment for fractures of the posterior and/or anterior pelvic ring. Among them, 46 patients received treatment with extended sacro-iliac screws and were therefore included in this study. Patients who underwent additional screw fixation for a fracture in the anterior pelvic ring were also considered. Table 1 provides an overview of the patients' demographics and fracture configuration. Figure 1 illustrates the possible screw positions based on fracture configuration and individual sacral anatomy. The indications for surgical treatment were immobility and a pain level ≥ 5 despite adequate analgesics, as measured on the visual analogue scale. Exclusion criteria for 77 patients included the use of other osteosynthesis techniques for low-energy fractures, a history of a high-energy trauma, tumorous pelvic lesions, and combination with complex surgical stabilization procedures. Fractures were graded according the Fragility Fractures of the Pelvic Ring (FFP) classification, based on preoperative routine MRI and CT imaging 17 .
Analyzing the records of all patients, the pre-and postoperative clinical status was evaluated based on mobility (immobile, mobile to sitting, mobile with crutches, mobile with walker). The incision-to-suture time was measured, although due to the minimally invasive nature of the technique and negligible blood loss, blood loss could not be quantified. Surgical reports, documentation of the postoperative course, discharge letters, and Surgical procedure. The patient is placed in the prone position on the positioning block under general anesthesia. The arms are positioned with 90° abduction of the shoulder joints and 90° flexion of the elbows cranially. Disinfection is carried out over the entire dorsal and lateral pelvis. Circumferential sterile draping ensures 360° sterility up to the operating table column (Fig. 2). After a team time-out, the posterior superior iliac spine and the iliac crest are palpated unilaterally. Two fixation pins are then drilled into the iliac crest via stab incisions, to which the navigation tracking device (Brainlab AG, Munich, Germany) is attached.
To generate the 3D data set, the AIRO ® iCT (Stryker, Kalamazoo, Michigan, USA) is used. CT parameters and patient data are set in preparation. Laser guidance is employed to determine the extent of the surgical field. After the staff leaves the operating room, the in-house radiological-technical assistant obtains a CT AP scout of the pelvis, covering both ossa ilia and the sacrum. The iCT is performed under temporary apnea after preoxygenation of the patient, to avoid motion artefacts. The acquired data is verified and transferred to the institutional DICOM server and the navigation system (Brainlab Spinal Navigation Software Version 3.0, Brainlab Curve™, Brainlab AG, Munich, Germany). The OR table is then rotated 90° clockwise back to the working position. Following a fresh sterile covering of the floor-facing operating table and a glove change, the accuracy of the virtual reality is checked by aligning the navigation pointer and the fixation pins. Two transsacral fully threaded screws at S1 level; (B) Two transsacral fully threaded screws at S1 and S2 level; (C) Two transsacral fully threaded screws at S1 and S2 level plus one partially threaded screw through the superior pubic ramus; (D) Two transsacral fully threaded screws at S1 level plus one partially threaded screw through the superior pubic ramus; (E) One transsacral fully threaded screw at S1 level plus one partially threaded screw through the superior pubic ramus. www.nature.com/scientificreports/ To plan the incisions, the trajectories of the screws to be implanted are determined at the skin level using the offset setting (Fig. 3A).
After a skin incision of approximately 3 cm, blunt dissection is performed onto the cortex of the os ilium. The navigation instruments are registered, and a drill sleeve is inserted at the screw entry point on the ilium. Figure 3B illustrates the planning of a screw starting within the ilium and running through the sacrum and the corridor above the S1 foramina to the opposite side, bridging the contralateral sacroiliac joint. A navigated 2.6 mm drill wire is then inserted through the navigated sleeve and drilled according to the plan. A large-fragment fully threaded screw (TIS™ Königsee, Weye, Germany) of appropriate length, 7.5 mm in diameter and with a washer, is implanted over this wire. Depending on the fracture configuration and the width of the S1 corridor, a second screw is implanted in a similar manner, either caudal to the first screw at the level of S1 or through the corridor between S1 and S2 foramina in S2. The correct positioning of the screws is then documented fluoroscopy in the AP and lateral as well as in the inlet and outlet projections (Fig. 4).
In case of an additional fracture of the anterior pelvic ring requiring treatment, another navigated 2.6 mm drill wire can be placed through the superior pubic ramus into the pubic bone, using the same skin incision.
To verify the positioning of the navigated drill wire, the iCT is repeated and, once the correct wire position is documented, a partially threaded screw (TIS™ Königsee, Weye, Germany) is implanted according to the planned length. Following the flushing of the surgical field with sterile saline solution, the wound is closed using subcutaneous and skin sutures.
Feasibility. Primary objective of this study was to assess the feasibility of the presented technique. Therefore, the aforementioned parameters were collected to analyze the peri-and postoperative complication profile and the clinical status of the patients. www.nature.com/scientificreports/ Statistical analysis was conducted using IBM SPSS (IBM, USA, Version 27). Statistical significance was set at an error probability of p ≤ 0.05. Multiple regression analysis was performed using ANOVA to examine risk factors related to screw loosening.
Results
Of the 46 patients included in this study, ten were male (22%) and 36 (78%) were female. Median age was 81.5 years, ranging from 60 to 91 years.
In 32 cases (68%), a low-energy trauma caused the fracture, while in 14 cases (32%), no trauma was reported in the medical history. All patients experienced immobilizing pain (visual analogue scale ≥ 5) despite receiving adequate analgesics. No case presented with newly diagnosed neurological deficits. The preoperative grading of fractures using the FFP classification revealed a heterogeneous distribution of fracture types in our collective, as shown in Fig. 5. Table 1 provides an overview of the patients' demographics and fracture characteristics.
The technique of monoportal osteosynthesis was successfully applied in all cases as planned preoperatively. Regardless of the patients' body size or BMI, a single iCT scan was sufficient to visualize the entire pelvis and enable the planning of screw trajectories required for osteosynthesis of the posterior and/or anterior pelvic ring in every case. www.nature.com/scientificreports/ In 35 cases (77%), two transsacral screws were planned and implanted. In six cases (13%), an additional partially threaded screw was used to treat a coexisting fracture of the anterior pelvic ring, and in five other cases, additional surgical procedures (kyphoplasty, lumbar instrumentation) were necessary.
As mentioned earlier, the quantification of intraoperative blood loss was not possible due to the minimal amount of bleeding, which did not require the use of a suction device.
The relevant surgical and anesthesiologic reports, as well as the documentation of the postoperative course and outpatient follow-up examinations, were reviewed for intraoperative and perioperative complications. No perioperative nerve root injury or postoperative neurological deficit were observed. Additionally, there were no incidents of vasculature or pelvic organ injury in the investigated cohort. Postoperative hematoma or wound www.nature.com/scientificreports/ healing disorders requiring revision surgery were also not encountered. In summary, no complications were attributed to the surgical technique used. The mean incision-to-suture time required for implanting two screws in S1 or two screws in S1 and S2, respectively, was 52.6 min (SD 13 min). In cases where an additional screw was planned and implanted in the pubic bone, the mean incision-to-suture time was 71 min (SD 13.4 min).
Whether both screws were implanted in S1 or if a second screw at the level of S2 was necessary depended on individual anatomical factors and the intraoperative situation after the first screw was implanted.
Postoperative control of screw placement through ap, inlet, outlet, and lateral X-rays of the pelvis revealed correct positioning of all implanted screws, eliminating the need for any corrective measures.
On the first postoperative day, all patients, except one who remained bedridden, were mobilized under physiotherapeutic supervision. To prevent the risk of stumbling and sudden weight-bearing, initial mobilization was performed using a walker or crutches, with full body weight being gradually induced. At discharge, 14 patients (30%) were mobile with crutches, 30 patients (65%) used a walker, one patient (2%) was able to sit, and one patient (2%) remained bedridden. This resulted in a 98% improvement in mobility compared to the preoperative status.
The median length of postoperative hospital stay was 6 days, ranging of 3-30 days (excluding one outlier who required inpatient treatment for Covid-19 infection under isolation conditions). Patients were routinely discharged to inpatient geriatric follow-up treatment for further clinical and functional consolidation.
Out of the initial 46 patients, 36 attended the scheduled follow-up examinations three and twelve months postoperatively, resulting in a follow up rate of 78%. X-ray examinations of these patients revealed radiological signs of screw loosening, which was confirmed in subsequent CT scans of the pelvis in seven cases (19%).Among these cases, four patients required reoperation, resulting in a reoperation rate of 11% in relation to the followed cohort. In two cases, the removal of the loosened screw was sufficient as fracture healing was complete. In one case, the loosened screw was replaced with a thicker hydroxyapatite coated screw (SI-LOK ® , Globus Medical, Audubon, Pennsylvania, USA), and in another case, lumbopelvic stabilization was performed. In the remaining cases where screw loosening was asymptomatic no specific treatment was necessary. Routine microbiological analyses of the explanted screws yielded negative results, indicating that mechanical factors were the likely cause of the loosening.
Multiple regression analyses were conducted to assess potential risk factors for screw loosening, but none of the following parameters reached statistical significance: overlapping screw wavers, FFP type of fracture, presence of a H-shaped fracture, presence of a comminuted fracture, and screw localization.
Discussion
In this study, we present our results from four years of experience using the 3D-navigated monoportal PSS for fragility fractures of the posterior and/or anterior pelvic ring. This technique, utilizing iCT and a computer navigation system, represents an evolution and refinement in the field of minimally invasive osteosynthesis techniques. The advantages of this technique include the ability to safely place screws in the S1 level, as well as within the narrow pathways of the S2 level of the sacrum and the ramus of the pubic bone, without causing harm to adjacent sensitive tissue. All screws could be reliably implanted along the preoperatively intended trajectory, while still allowing intraoperative flexibility to accommodate changes in strategy and the need for additional screws. A considerable number of the treated fractures exhibited signs of dislocation (FFP3&4), but low-energy fractures did not require repositioning maneuvers. Preoperative secondary dislocation of fractures did not occur. However, by generating the 3D dataset immediately prior to trajectory planning and screw implantation without further patient mobilization, it would be possible to adapt to changing anatomical situations. The overall complication profile was very low with no instances of nerve root or vessel injury, and excellent results in terms of incision-to-suture time and blood loss, both of which are crucial factors for the clinical outcome of geriatric patients. However, in the long-term follow-up, we observed a considerable rate of screw loosening, which necessitated revision surgery in a several cases. Secondary escalation of the surgical strategy in terms of spinopelvic fixation due to insufficient healing of the fragility fracture was required in one case only.
In view of the experienced hardware failure, cement augmentation must be discussed as an alternative minimally invasive treatment for fragility fractures. Various procedures, such as balloon sacroplasty, radiofrequency sacroplasty, and cement sacroplasty, are available and have shown to be feasible treatment options [18][19][20][21] . However, it is important to note that vertebro-sacroplasty, in particular, carries a significant risk of cement leakage. Varying data in the literature, reporting cement leakage rates ranging from 0.4 to 24%, indicate a highly user-specific complication profile 22,23 . Moreover, cement augmentation is not recommended for dislocated sacral fractures.
Osteosynthesis of sacral fractures using sacroiliac screws under guidance of 3D computer navigation has shown potential as a superior alternative to conventional 2D-guided minimally invasive techniques 10,24 . Given the inter-individual variations in the spatial anatomical features of the sacrum, which can result in potentially narrow trajectories for screw implantation, 3D navigation provides an additional safety factor. This is especially relevant for transsacral screws, which must navigate the sacral neuroforamina on both sides within the same trajectory. As mentioned earlier, in addition to the advantages of 3D navigation, transsacral screws have demonstrated superior stability values in a finite element model of sacral fractures when compared to regular sacroiliac screws. It is therefore advisable to consider fixation with a single sacroiliac screw in both the S1 and S2 segments. Regardless of the type of sacroiliac screw used, if only one screw can be implanted, fixation in the S2 segment appears to be preferable over fixation in the S1 segment 11,12,25 . Based on these findings, we opted to implant S2 screws whenever the narrow anatomy of the trans-sacral corridor allowed for it.
In summary, minimally invasive 3D-navigated osteosynthesis of fragility fractures of the pelvis appears to be a promising approach, particularly in geriatric patients. However, the published data of its clinical application www.nature.com/scientificreports/ is limited, and the optimal fragility fractures of the pelvis is yet to be determined 7 . Therefore, our intention was to provide an illustrated description of this promising surgical technique and report our four years of clinical experience using it. The technical feasibility of our setup, which involved intraoperatively acquired CT scans and a well-established 3D navigation system, was confirmed by the safe and precise implantation of screws, as well as the relatively short time from skin incision to wound closure and minimal blood loss. Another research group investigating this technique using alternative imaging and navigation devices also achieved consistently accurate screw positioning and a low perioperative complication profile 26 .
The arrangement of the technical devices in the operating theatre and the crucial surgical steps were integrated seamlessly into the routine workflow of our department. However, we acknowledge several potential disadvantages of the presented technique and the devices used. These include the cost of both implementing an intraoperative CT and 3D navigation system, which may limit its availability in certain hospitals or healthcare settings. Additionally, the higher radiation dosage associated with intraoperative CT scans compared to conventional procedures must be carefully considered. Proper radiation safety measures should be implemented to minimize potential risks associated with radiation exposure for both patients and surgical staff.
In our operative setting, all staff leave the operating theatre during the brief phase of acquiring the CT data set, ensuring no radiation exposure for the staff. Similar considerations apply for modern 3D fluoroscopy scanners, which use comparable or even higher radiation dosages 27 . Compared to these, the advantage of CT with its larger field of view becomes particularly apparent in 3D-navigated surgery of the spinopelvic region, especially in obese patients.
It is also important to note that the time consumption associated with intraoperative CT should be evaluated in the context of the specific procedure. Factors such as set-up time, image calculation time, and the need to briefly leave the operating room during acquisition can contribute to the overall time consumption. On the other hand, technically evolved intraoperative CT-based navigation systems offer precise real-time guidance and visualization, which can enhance surgical accuracy and potentially reduce the time required for procedures that particularly precise implant positioning 28 .
The peri-and postoperative complication rates were found to be excellent, and the intended early postoperative mobilization of patients was achieved in nearly all cases. Acknowledging the limitations of a small study cohort, it is important to note that general complications, including postoperative hematomas and wound healing disorders, are ultimately unavoidable. Nonetheless, the demonstrated minimally invasive approach shows its merits in terms of a low complication rate. One limitation of the minimally invasive approach could be primary or secondary dislocation of the pelvic ring fracture, which is more common in high impact trauma. However, due to the low-energy nature of fragility fractures of the sacrum and the typically intact stabilizing ligamentous apparatus, severe dislocations of these types of fractures are uncommon 29,30 . In case where the fracture is close to the symphysis or involves a dislocation of the symphysis gap, the safe implantation of a superior ramus screw into the pubic bone may not be feasible, and an additional open procedure may be necessary.
Another limitation of the presented study is the absence of a control group using alternative surgical techniques. Screw loosening occurred some cases with the studied technique, but the need for reoperation was low and consistent with previously published data. In a comparative study of different osteosynthesis techniques, Eckardt et al. found a good functional outcome, similar to our results, and comparable screw loosening rates, resulting in a reoperation rate of 9%. This showed a marked, yet non-significant superiority over a reoperation rate of 26% in the group of patients treated with regular sacroiliac screws 14 . Additional larger case series are needed for a more precise estimation of loosening rates. Regression analyses of relevant parameters in our patient cohort failed to identify significant risk factors. Given this context, we must assume that our sample size is too small to accurately assess causative risk factors. One possible cause of screw loosening to consider, is the particularly high biomechanical stress on a screw osteosynthesis bridging both sacroiliac joints, exposing it to opposing nutation forces. However, the fixation of transsacral screws in a total of four layers of cortical bone is considered protective against screw pullout or loosening. In individuals with osteoporosis and compromised bone quality in the sacral body, the compression applied to the sacroiliac joint and the sacral fracture may be reduced 31 . This reduced compression decreases the resistance against the vertical force that occurs during mobilization and may pose a vulnerability for implant loosening. In a recent biomechanical study of osteosynthesis of fragility fractures, Zderic et al. presented a prototype of screw-in-screw fixation in regular S1 screws and found lower implant movement compared to conventional S1 screws and comparable implant movement to transsacral screws 32 . Furthermore, Wagner et al. reported a screw loosening rate of only 4.7% with combined osteosynthesis using a transsacral bar and unilateral or bilateral sacroiliac screws 33 . Further studies on these techniques are of great interest, aiming to improve patients' outcomes by minimizing implant loosening rates to a minimum.
Conclusion
The minimally invasive monoportal PSS for fragility fractures of the sacrum, as presented here, is a feasible and promising technological advancement in surgical treatment strategies for such injuries. Further refinement of the technique should focus on achieving reduced implant loosening and subsequent reoperation rates. This necessitates additional studies on biomechanical properties, comparative analyzes to other techniques, and the optimization of patient and implant selection.
Data availability
All data generated or analysed during this study are included in this published article [and its Supplementary Information files]. The data of this work are part of the dissertation of Mr. Martin Naisan.
Author contributions
A.K. contributed with the following: Conception, design of the study. Acquisition, statistical analysis and interpretation of the data. Creating and drafting the manuscript. M.N. contributed with the following: Conception, design of the study. Acquisition, statistical analysis and interpretation of the data. S.K. contributed the following: Creation of the drawings shown in Fig. 1. M.R., F.R. and P.H. contributed with the following: Critical revision of the manuscript for important clinical and technical content as well as supervision of the project. Consent to publish has been received from all participants.
Funding
Open Access funding enabled and organized by Projekt DEAL.
|
2023-07-30T06:17:09.565Z
|
2023-07-28T00:00:00.000
|
{
"year": 2023,
"sha1": "a403a2af5d9844404aa877d2ae037b53eafa25bf",
"oa_license": "CCBY",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "9264f8380d44a16d1fdcc3bc277e6ca9820c97ed",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
250999858
|
pes2o/s2orc
|
v3-fos-license
|
Numerical Investigations on Stresses and Temperature Development of Tool Dies during Hot Forging
Hot-forming tools are subjected to high thermal and mechanical stresses during their application. Therefore, a suitable design of the tool die is important to ensure a long tool life. For this purpose, numerical simulations can be used to calculate the occurring stresses and the temperature development in the tools during the course of a stroke or over several forging cycles. The aim of this research is to investigate the effect of different radii on the resulting stresses in the lower die of the forming tools. Furthermore, the temperature evolution over several cycles is analysed to determine their effect on the temperature. When investigating the stress, it was found that a larger radius leads to a reduction in stresses. In addition, it could be numerically proven that the base temperature of the die levels off after a certain number of cycles. These findings will be used in further research dealing with the service life calculation of dies subjected to thermo-mechanical alternating stresses.
Introduction
For hot-forming tools, martensitic tool steels are often used, which offer the highest possible strength. During hot forming, the tool materials are subjected to high mechanical, thermal, tribological and chemical stresses resulting in a complex load spectrum [1]. By exceeding the local yield stress in the highly loaded areas, plastic deformations occurs, which leads to material fatigue and finally to tool failure due to cracks [2]. For this reason, the finite element method (FEM) is increasingly used to determine the local stress values for the design of tools [3]. Since the tools are subjected to thermo-mechanical alternating stress during their service life, which leads to elasticplastic failure, models for determining the tool life become more important [4].
Hot-forming tools are subject to high thermo-mechanical stress. This stress occurs cyclically due to the type of use. This cyclic alternating stress leads to failure in the area of the highest stress, which often manifests itself in the form of cyclic cracking [5]. Cyclic cracking results from cyclic thermal hardening and softening, which must be taken into account when planning the tools [6]. Accordingly, models to describe the material behaviour are needed that also represent cyclic cracking. A first approach is in the work of [7], which evaluates failure based on strain and kinematic factors for hot stamping. Another study dealt with crack initiation because of plastic deformation during cold forging [8]. This work provides basic research for a new plasticity model taking into account the cyclic softening of the material during the hot-forging process in order to further improve the numerical model. The final overarching goal for the future is to calculate the service life of the tool.
Optimised numerical tool design can take into account the process variables that occur in the manufacturing process [9] and enable a more efficient design of the tools and step sequences. As a result, an optimisation of the tool configuration is already possible in the construction phase of the forging process through stress-adapted dimensioning with little experimental effort for process design [10]. For this purpose, it is important to determine the occurring temperatures and stresses as well as the effects of design changes [11].
During hot forging, the near-surface areas of the forging dies heat up quickly to peak temperatures of up to 630 °C. In combination with cooling lubricants the die surface can be cooled down to temperatures between 100 °C and 200 °C, resulting in a temperature difference of approximately 400 °C to 500 °C. Most of the thermal energy is introduced into the surface layer of the die by the heated semi-finished product, but the preheating of the forging dies and the friction that occurs during the process also cause a rise in temperature. Typically, the semi-finished product temperatures are up to 1,250 °C in hot forging [12]. A high temperature of the semi-finished product reduces the flow resistance of the material [13]. With a pressure contact time of 50 ms, the heating rate of the die is up to 10,000 K/s, but the temperatures at the surface depend on the process and the geometry. According to [14], temperatures of 400 °C up to 450 °C can be reached during hot die forging, depending on the measuring point in the die. The temperature in the surface increases for a short time (approximately 0.15 s).
Using an FE-based process calculation, the heat transfer from the semi-finished product to the environment and to the tools during the forming process can be calculated locally. Due to high mechanical stresses, it is not possible to measure the temperature in the tool edge layer with the measuring equipment known today. The FEM offers great advantages for the calculated temperature determination of these areas. Within the framework of a research project, various points near the tool surface were measured in experimental tests with the help of thermocouples. The temperatures measured there are used as a comparative value for numerical simulation. If a good agreement between the simulatively and experimentally determined values is achieved at these points, the boundary-layer temperature of the tool can be concluded with the help of the FE model. In addition, the FEM can then also be used to calculate the temperatures or stresses occurring over several cycles [15].
In this research, the process under investigation exhibits high cyclic thermo-mechanical loads, which lead to the formation of cracks in the lower die in the most heavily loaded area. This lies in the lower radius of the inner area of the die. Accordingly, a numerical process investigation is essential to prolong the tool life and to calculate the load progression. The simulation takes into account the thermo-mechanical interactions between the semi-finished product and the dies that are used in hot forming processes. In general, there is the possibility to consider the tool stress within the framework of a thermo-mechanically coupled simulation or in one without coupling [15]. In the thermomechanically coupled calculation approach, the tools are considered as deformable components within a process simulation. In this case, the thermal-mechanical tool-material behaviour under normal operating load is already taken into account during the process simulation. In the simulation without coupling, the forming process itself is calculated first and the tools are assumed to be rigid and not thermally conductive. The load on individual tool components is only calculated in a subsequent simulation, but thermal effects are not taken into account. If time and costs have to be saved, the thermo-mechanical interactions are neglected in the simulation, which is why the simulation is carried out without coupling. If an exact representation of the stresses with thermal interaction is to be investigated, the thermo-mechanical coupled simulation should be selected. Since the thermo-mechanical interactions are important in the context of this research, this approach is chosen.
The aim of this work is to identify the influence of different radii on the lower die area by numerical calculations. Such investigations should make it possible to identify the most suitable radius for subsequent application in hot forging for the overall research project. In this way, it is ensured that a compromise between tool life and crack initiation is achieved and that there is no early failure due to crack formation. Another aim of this work is to numerically identify the number of forging cycles until the temperature in steady state is reached in the die. The knowledge gained here with regard to the more precise definition of the boundary conditions can be used for further numerical calculations in order to realise a more precise process simulation with regard to the calculation of tool life.
Numerical Model
Within both investigations (stresses at different radii and temperature evolution over several cycles), the semi-finished product of material 1.7225 according to ISO 683-1 (also referred to as AISI 4140 or 42CrMo4) is used, which has already been characterised and parameterised for simulations within the scope of a previous work [3]. The tool steel 1.2367 according to ISO 4957-2 (also referred to as X38CrMoV5-3) is selected as a typical die material. Flow curves of this material were determined from stress-strain diagrams in previous research such as [4], and in both prepared for usage in the simulations. Complementary material data like yield strength and tensile strength were supplemented by calculations with JMatPro so that a classification of the calculated stresses to the bearable loads can be made. The numerical models were built up within the FE-Software Simufact Forming 16.0 and consist of a lower die with ejector, an upper die and a suitable semi-finished product (Fig. 1). This geometry was chosen here because failure can be induced specifically in the lower radius and is thus suitable for investigating crack initiation by thermo-cyclic loading. The upper die and the ejector are considered as rigid and non-heat-conducting during the simulation, as they are not in the focus of this research. In addition, this simplification has no influence on the temperature and thus the stress development in the lower radius of the lower die. This was checked in advance by simulation and it was found that it has no influence on the lower die considered here. It only has a small influence on the semi-finished product in the upper surface area. This can save resources and the influence can be neglected in this framework. The lower die has been set as deformable with heat conduction. Therefore, the temperature development and the resulting stresses can be calculated within the thermo-mechanical coupled simulation. In addition, this coupled simulation can generate more accurate results regarding stresses arising in the lower die, as the temperature influence is also taken into account. This model of a hot-forming process was used for investigations on the different radii and for the temperature development. For the forming process, a stroke of 29.34 mm is considered which results in a flash thickness of 0.9 mm. A press with a constant speed of 290.57 mm/s was set, resulting in a process time of 0.14 s. This corresponds to the abstracted construction of the press planned for later experimental trials. The friction was approximated with the combined friction model using a friction coefficient µ of 0.15, and a friction factor m of 0.3 based on the information of [16], and the experimental values obtained from previous tests [3]. The initial temperature distribution of the semi-finished product was determined by an additional cooling simulation, which depicts the transfer time of 5 seconds from the heating unit into the press. In the experimental process, the transfer is realised with a robot. The initial temperature after heating is set to 1,250 °C. The basic temperature of the die is 180 °C. A mesh study was carried out to identify the appropriate discretisation of the billet and the lower die for this forming process and the mesh properties determined were used in this research.
Procedure of load analysis with respect to different radii. In this section, the effect of different radii in the bottom die on resulting stresses is investigated. For this purpose, six different variants with a radius of 0.30 mm, 0.50 mm, 0.75 mm, 1.00 mm, 1.50 mm and 2.00 mm were designed (Fig. 1). These studies are used for a more specific design of the die for the selected hot forming process. The loads that occur make it possible to estimate which variant is likely to have a longer tool life compared to the others. Since the overall aim of the research project is not to achieve the longest service life, but rather a suitable one for investigating cyclic crack formation, smaller radii are also taken into account in the comparison. The simulations were evaluated in relation to the equivalent stress and the maximum principal stress. To calculate the loads, a forming process representing the first forging cycle was simulated in each case. For the evaluation of the occurring stresses, the temperature development in the vicinity of the lower radius is also taken into account. In the case of metals, there is a strong temperature dependence of the material parameters, which is why an influence on the occurring stresses can be assumed. Increasing the temperature reduces the mechanical properties and the tools would fail faster under the same load. Accordingly, when comparing the material parameters with the occurring stresses, the consideration of the existing temperatures is of particular importance. By comparing the data, a statement can be made about the probability of crack formation and plastic deformation.
Procedure of the temperature investigation. In the second section, the procedure for investigating the temperature development in the lower die over 30 cycles is presented. A single geometry, the one with the highest stresses in the lower die, was selected and examined as an example.
When setting the boundary conditions for the simulation, the same parameters are selected for the start in the first cycle as were chosen in the previous investigations on the effects of the radius. The semi-finished product has the same initial temperature (1,250 °C) in every calculated cycle. After each calculated cycle, both the thermal and the mechanical load on the bottom die are transferred to the simulation of the next cycle, so that the temperature evolution on the die can be tracked.
When investigating the temperature development, the cooling effect of the cooling lubricant was abstracted by two different variants, without implementing an active cooling process in the abstraction. This means that the otherwise active cooling by means of cooling lubricant is transferred within the simulation into a cooling process without actively influencing the tool cooling. On the one hand, cooling with a small abstracted cooling effect and, on the other hand, with a high abstracted cooling effect between the strokes are simulated. From this, a conclusion can be drawn about the influence of the cooling effect on the temperature development. To compare the results, four measuring points (MP1 to MP4) were defined based on the first cycle to determine the resulting temperature in the lower die (Fig. 2). The measuring points are oriented in such a way that they represent both the temperature in the lower radius (MP1 to MP3) and the one further inside the die (MP4). The distance of MP4 from the surface is 5.00 mm. A total of 30 cycles with the abstracted cooling were examined and the temperatures before and after cooling were analysed. 30 cycles were chosen because a constant temperature range has already been established at this point. The measurement time before cooling means that the temperature was determined at the end of the
Achievements and Trends in Material Forming
forming process. Afterwards, the temperature after the abstracted cooling was evaluated so that the influence of the cooling could be analysed.
Key Findings
In order to be able to make a statement about the influence of the radius on the stresses, the data from the stress calculations are dealt with first. For this reason, the results of 100 % stroke are shown. Moreover, the most advantageous variant is selected and then the temperature evolution in this lower die over the 30 cycles is presented. Development of stresses in the lower die. The lower die experiences a thermo-mechanical load, which is why the choice of the correct radius is of particular importance. Since a suitable service life of the lower die is desired for the investigation of cyclic crack detection, the stresses should not be too low. In order to be able to put the results of the stresses into the right context, the temperature development in the lower die must first be considered (Fig. 3). Here, the area of the radius is considered in particular. The results show that in the lower area there is no significant temperature difference between the radii. For all radii, the temperature is about 350 °C with a radius of 0.3 mm and 400 °C with a radius of 2.0 mm. Due to an increase in temperature with increasing diameter, it can be assumed that there is a higher heat input due to a larger contact area in the radius. Fig. 4, it can be seen that the maximum values are limited to the area of the radius. Thus, the maximum is between 560 MPa to 650 MPa depending on the radius size. Furthermore, it can be seen that the smallest radius selected, 0.3 mm, has the highest value regarding the equivalent stress of 650 MPa. As the radius increases, the equivalent stress decreases to 560 MPa at a radius of 2.00 mm. It can also be seen that the area of the maximum value is smaller with a smaller radius. As the radius increases, this area becomes larger because the load is also applied over a larger area. In addition, the notch effect also decreases as the radius increases.
At 400 °C, the yield strength for tool steel 1.2367 is 1,100 MPa and the tensile strength is 1,300 MPa (calculated with JMatPro). If lower temperatures occur in the region under consideration, this has no negative influence on the service life of the die in the radius. The yield strength at 400 °C can be compared with the equivalent stress in the range at the different radii. It is noticeable that the equivalent stress is below the yield strength. Accordingly, it can be assumed that no plastic deformation will occur in the area under consideration within the first cycle. Comparing the positive maximum principal stress at the different radii, a decrease of the maximum value from 480 MPa at a radius of 0.30 mm to 350 MPa at a radius of 2.00 mm can be seen (Fig. 5). All dies have the same orientation in the stress profile. In addition, a downward and slightly lateral orientation can be seen in all of them. The area of the maximum values increases as the radius increases. In addition, the positive maximum value for all variants is in the lower radius, which was to be expected due to the shape of the die. The focus is mainly on the positive maximum principal stresses, as these represent the tensile load in the die. If the tensile load is too high, cracks can develop, which then lead to failure. In the area of the lower radius, the strongest force acts outwards, creating the tensile load. The selected tool material 1.2367 has a tensile strength of 1,300 MPa at 400 °C (calculated with JMatPro). In comparison with the occurring positive maximum principal stress, it can be seen that the tensile strength is above the calculated stress. This means that there will be no failure due to cracking within the cycle under consideration. If all results are summarised, the expected tendency in the stresses can be clearly seen. Despite the stress peak in the area of the lower radius, the material characteristic values of yield strength (1,100 MPa) and tensile strength (1,300 MPa) are not exceeded in comparison with the equivalent stress (max. 650 MPa) and the maximum principal stress (max. 480 MPa). This means that no plastic deformation and no crack will occur in the considered cycle based on the given material properties.
It is possible to use the distribution of the stress type (tension-compression) to recognise how the load acts on the radius. By compressing the semi-finished product, a force acts radially outwards on the upward-extending wall of the bottom die. This creates an expanding force on the lower radius and a tensile force on the radius. This mode of action does not depend on the selected radius and leads to a cyclical tensile alternating stress. This type of load creates cyclic cracks in the die, which are the target of this research.
Temperature development over the number of cycles. The temperature development in the lower die was simulated with two different abstracted cooling scenarios for the lower die with a radius of 0.3 mm. When comparing the results, it is noticeable that the tendency is the same for both variants. The temperature at four measuring points at the end of the forming path and at the end of the cooling is considered. The tool temperature after cooling with a short cooling effect reaches a constant temperature range after approximately 16 cycles (Fig. 6, top). Over the cycles there is first a high rise up to 250 °C before it settles down to a base temperature of 230 °C after cooling. The temperature at 100 % stroke behaves similarly, even though greater fluctuations in the results can be seen. The only difference lies in the height of the temperature (about 350 °C). In the simulations for the higher cooling effect (Fig. 6, bottom), the temperature in the die could be reduced further to about 210 °C. The temperature settles at an earlier point, namely after about five to eight cycles. As with the other cooling effect, the temperature inside the lower die (MP4) hardly changes before and after cooling.
The comparison of the temperature development shows a high dependency between base temperature and cooling effect. Accordingly, it is important in the further course of the project which intensity of cooling lubrication is chosen. If a cooling lubrication is chosen that lowers the temperature too much, the thermal load is also reduced. This is disadvantageous for a service life model that is based on cyclical thermomechanical stresses, as is planned in the rest of the research project. Furthermore, it can be seen that the highest temperature differences are in the surface area of the lower die. The difference between base and peak temperatures is approximately the same for both cooling effects considered. Accordingly, a cooling lubrication should be used that does not strongly reduce the base temperature. This allows higher peak temperatures to be reached within the process, which leads to a higher thermal load in the die.
Conclusion and Outlook
Based on the investigations of the stresses and the temperatures in the lower die with different radii, the following can be stated. It is important to consider the selected geometric shape in relation to the influence of temperature development. In addition, an influence can be seen in the stresses that arise at different radii. It has been shown that a larger radius leads to lower stresses in the lower die, which is also already known from the literature.
The geometry chosen here is suitable for investigating crack initiation due to thermo-cyclic loading, since failure can be specifically induced in the lower radius. By targeting the crack initiation, subsequent investigation and installation of measuring equipment during the experimental test, for
566
Achievements and Trends in Material Forming example, to record the temperature development in the lower die, is facilitated. The chosen simplifications related to the reduction of the simulation time by using non-heat conducting components saved resources. It was possible to prove in advance that this simplification has only a minimal influence on the semi-finished product and not on the lower die investigated here. Overall, it can be concluded that the starting temperature or the selected radius has a greater influence on the resulting stresses. In addition, this investigation has identified a highly stressed die that will withstand the first cycle. This will be validated in further researches. The results from the stress calculations and the temperature development are used for the design of a suitable lower die for the experimental forging tests to follow. Although failure due to crack initiation of the lower die is aimed for, this is not to take place in the first cycle. Instead, a service life of 500 to 1,000 parts should be available, which allows a good analysis of the crack initiation. The mechanism of time-and temperature-dependent crack growth is to be used as a basis for the accurate prediction of the service life. A further basis for this is to be an extended time-and temperaturedependent plasticity model for the tool steel used here, with an overall mechanism-based approach describing the softening.
By simulating the temperature development via the number of cycles, it can be stated that this process does not need many cycles for the temperature to run in a steady state. For later simulations or experiments, this means the following: In the experiments, the temperature in the die can be recorded within the first 30 to 50 cycles in order to determine the basic temperature in the die. This depends on the starting temperature of the die and the cooling strategy used. Furthermore, process parameters such as semi-finished product temperature, process time and geometry are important. By determining the temperature in the first 30 to 50 cycles, the simulation can be more accurately specified based on measurements from experiments. Further calculations can then be carried out with the basic temperature that is then set, so that the calculated loads correspond to the steady state in serial forging tests. Furthermore, the base temperature in the die can be further reduced by a greater cooling effect during cooling lubrication. This should be avoided in the context of this research question, as a too strong reduction of the peak temperature reduces the thermal load. Since the thermal load on the die is important for the upcoming experimental tests and for setting up the targeted service life model, this must be taken into account.
Based on these results, the variant with the smallest radius can be recommended as the preferred die. This ensures that failure does not occur within the first load cycle. Nevertheless, the thermal and mechanical load is in a range that suggests a failure due to cyclic loading. Accordingly, the next step is to fabricate the dies and validate the simulations through experimental forging tests. At the same time, a service life model is built together with a plasticity model that describes the cyclic softening of the tool material and, based on this, calculates the crack initiation.
|
2022-07-24T15:07:09.853Z
|
2022-07-22T00:00:00.000
|
{
"year": 2022,
"sha1": "f63346e00cc7a3fea4228391e4613eb418b64747",
"oa_license": "CCBY",
"oa_url": "https://www.scientific.net/KEM.926.559.pdf",
"oa_status": "HYBRID",
"pdf_src": "ScientificNet",
"pdf_hash": "ccaee179b1ca4c17302f86072d1297ee95c23e95",
"s2fieldsofstudy": [
"Engineering",
"Materials Science"
],
"extfieldsofstudy": []
}
|
62886832
|
pes2o/s2orc
|
v3-fos-license
|
Network Modelling of Functioning System of the Process Module of Oil-Contaminated Wastewater Treatment
The article discusses network modeling of oil-contaminated waste water treatment at the stage of the process module functioning of water jet cleaning of waste water in the oil fields and petrochemical industries. Based on the review of the main modeling methods of discrete-continuous chemical processes, expediency of using the theory of Petri nets (PN) for modeling the process of wastewater treatment in the oil fields and petrochemical industries is substantiated. It is proposed to use a modification of Petri nets which is focused on modeling and analysis of discrete-continuous chemical processes by prioritizing transitions, timing marks in positions and transitions. A model in the form of modified Petri nets (MPN) is designed. A software package to control the process for wastewater treatment is designed by means of SCADA TRACE MODE.
Introduction
A serious problem for human existence is to protect water recourses from the increasing polluted wastewater discharged by industrial enterprises.
Cleaning and deep cleaning of process effluents is a prerequisite for the maintenance of ecological balance of the environment.However, existing wastewater treatment processes are not perfect as they do not provide an adequate level of treatment and require further research.
Modern process systems of wastewater treatment have a complex multi-layered structure, therefore, they can be considered as complex cybernetic systems.When studying them, the strategy of system analysis is used.Given the task complexity of modeling and analysis of such systems, it is necessary to apply modern methods of mathematical and computer modeling.
Materials and Methods
In solving the problems set up in the study, the methods of systems analysis, computer modeling, Petri nets theory, graph theory were used.
Theory
The continuous growth of oil production and consumption of petroleum products involves a significant increase in oil-contaminated wastewater whose effective cleaning is a prerequisite for preserving the environment.
Modern treatment facilities in oil fields and large petrochemical plants are structurally complex systems.Therefore, of considerable interest are the conditions for their emergency operation in which waste water has dynamically varying parameters both by composition and by flow rate up to the volley of sewage indicators (Fesina, & Savdur, 2014).The efficiency of such systems can be achieved by using modern methods of information processing, using the methods of complex objects system analysis based on the mathematical description of the process (Hunt, Timoshkina, Baudains, & Bishop, 2012).
In accordance with the principles of system analysis, industrial wastewater treatment plant is a chemical and engineering system, which includes a set of interrelated material, thermal and information flow units, each of which has a hierarchical structure (Motameni, Movaghar, Shirazi, Aminzadeh, & Samadi, 2008).Waste water treatment can be divided into interconnected subsystems characterized by a hierarchical structure.Management tasks at each level of the production hierarchy are different, but the general objective is wastewater treatment to standard indicators or to provide recycling water supply level.
A main area of studying complex systems, which wastewater treatment represents, is informational approach that is based on mathematical modeling of the object (Huilinir, Aspe, & Roeckel, 2011).Modeling and computer experiments with model-replacement of an object are an effective means to create management systems, to consider the object's behavior in emergency situations, to evaluate its structure and control rules, as well as to take into account the stochastic nature of disturbances (Haroonabadi, Teshnehlab, & Movaghar, 2008;Ruiz, Sin, Berjaga, Colprim, Puig, & Colomer, 2011).There are two approaches to the modeling of real objects.In the first approach, the object is represented as a dynamic system with a continuous variable.This approach is widely used in modeling chemical and engineering systems with continuous organization of processes (Peter, 1976;Buswell & Mueller, 1952) provided its stationarity and the invariableness of physical and chemical parameters.In the second approach, the object is represented as a dynamic system with discrete events.These include manufacturing systems, assembly lines, computer networks.
Dynamic system with discrete events class also includes discrete-continuous chemical and engineering systems.Solving the problem of managing such discrete dynamical systems requires the use of special mathematical methods.Traditionally, for this purpose the state machine approach, logical-linguistic and simulation modeling are used, along with the theory of graphs and networks, PN (Zhou & Li, 2010).Comparative analysis as the primary unit of mathematical modeling enables to select the PN theory (Zhou & Li, 2010).PN enables to simulate discrete parallel asynchronous processes (Zhou & Li, 2010), to get a graphical representation of the network, to describe the system at different abstraction levels, to present the system hierarchy (Barzegar & Motameni, 2011), to analyze models using modern software packages.
The process of oil-contaminated wastewater treatment is realized in the device using the swirling flows with consistent and effective implementation of all stages of the damage mechanism of emulsion structure of heterogeneous stream of oil-contaminated wastewater according to scheme: hydroclone -cylindrical chamber of top and bottom drain -settler.
To describe the system, we propose to use N-schemes, based on the mathematical apparatus of Petri nets, whose advantage is possible representation of the network model both in analytical form, automating the process of analysis, and in graphical form providing visualization of the model developed.
When analyzing chemical and engineering flow diagrams one should consider the main limitation of the N-scheme formalism, which consists in the fact that they do not account for the time characteristics of the simulated systems, since the enabling time of the transition is considered to be zero.Given these conditions, we have proposed the modified Petri net.MPN is Petri net in the form of С=<P,T,I,O,M,L,τ 1 ,τ 2 > where T = {t j } -finite non-empty set of symbols called transitions are measured depending on the number of conventional product portions with a continuous feeding to apparatus in the process flow.
P={p i }-finite non-empty set of symbols called positions.In our case it is a set of process flow devices; I: PxT → {0, 1} -input function, which for each t i transition gives the set its position p i ∈ I (t j ).O: PxT → {0, 1} -the output function, which reflects a transition to the set of output positions p i ∈ O (t j ).Thus, for each transition it is possible to determine the set of input position I (tj) and the output position O (t j ) as: I (t j ) = {p i ∈ P / I (p i , t j ) = 1}; O (t j ) = {pi ∈ P / O (p i , t j ) = 1} (1) M: P → {1, 2, 3, ...} -marking of net which assigns a non-negative integer to each position which is equal to the number of marking in a given position, which varies during the operation of the net.
Enabling the transition changes the marking instantaneously (p)=(M (p 1 ), M (p 2 ), M (p 3 )…M (p n )) for marking M / (p) by the following rule: Equation ( 2) means that the transition t j subducts one marking from the position of each of its input and adds one marking to each of the outputs.( ) p1 -position which informs about current volume of preproduct portions in the apparatus; p2 -position which informs about current volume of the portion processed in the apparatus; p3 -position which informs about space in the apparatus; t1 -transition modelling preproduct portion charge in the apparatus; t2 -transition which models processing of the portion charged; t3 -transition which models discharge of the portion processed.
Interpretation of chemical and engineering process in terms of the transition by the value of the input and output streams determines the condition of filling or emptying of the container unit (position).We accept the following evaluation parameters of the process: where V (τ) I -the current volume of i-th device of the process; V 0i -full volume of i-th device of the process; j = 1,2…k (k → ∞) -dose flow index in the interval Δτ.Using the PN-model, a software of wastewater treatment process module which simulate the operation of treatment in virtual time was developed.Software package for wastewater treatment process control system was developed with means of SCADA TRACE MODE (Nasby & Phillips, 2011).The process control system allows supervisory control of the main elements of the management system, to stop wastewater treatment system and analyze its state as a whole, and to predict the development of emergency situations (Huilinir, Aspe, & Roeckel, 2011).
Conclusions
The paper presents the network modeling approach based on the theory of modified Petri nets that determines the dynamics of the operation of the hydrodynamic treatment module of oil-contaminated wastewater.The possibility to develop systems of information management to solve the problem of reaching target values for wastewater treatment is shown.Constructing mathematical models of systems functioning of effluent treatment of petrochemical plants in the form of modified Petri nets enables to study the system communications and the rules for entire system functioning.The developed software of wastewater treatment systems enables to analyze the state of the treatment system as a whole and to predict the development of emergency situations.
Figure 1
Figure 1 presents a process diagram of a water jet cleaning unit of oil-contaminated wastewater.
Figure 1 .
Figure 1.A process diagram of a water jet cleaning unit of oil-contaminated waste water
τ 1 :
T→N и τ 2 : P→N functions which determine the delay time when enabling transition and the delay time in the position.The dynamics of MPN is determined by marking movement which simulates discrete flow balance of preproduct in the defined limits by the volume of wastewater treatment plants.Table 1.The state of individual apparatus (positions) for the chemical and engineering production in analytical and graphical form Process scheme of apparatus Apparatus model in the form of Petri net where 1 − i v , i v -volume flow rate at entrance and exit of i-th apparatus (m 3 /sec);V ( ) i τ , i V 0 -full and current volume of i -th apparatus (m 3 ).
Figure 2 s layer of oi
|
2018-12-21T19:10:08.588Z
|
2015-04-30T00:00:00.000
|
{
"year": 2015,
"sha1": "dba1748858b08b04dad737f4ad2839d60c37b148",
"oa_license": "CCBY",
"oa_url": "https://ccsenet.org/journal/index.php/ass/article/download/48362/26074",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "dba1748858b08b04dad737f4ad2839d60c37b148",
"s2fieldsofstudy": [
"Engineering"
],
"extfieldsofstudy": [
"Environmental Science"
]
}
|
16485534
|
pes2o/s2orc
|
v3-fos-license
|
The cost of diabetes chronic complications among Iranian people with type 2 diabetes mellitus
Background To evaluate the cost of diabetes related micro- and macrovascular complications in Iranian people with type 2 diabetes mellitus. Methods In routine clinical practice, people with type 2 diabetes mellitus were assessed for 10 years at a diabetes care center. The type of medications and clinical data were extracted from patients’ documents. Mortality rate and the incidence of micro- and macrovascular complications recorded in patients’ documents were analyzed. Cost analysis was comprised of 1) para clinic costs as well as laboratory, medications, clinical visits and nonmedical costs 2) inpatient costs as well as hospital admission costs, disability, and mortality costs. Results From 1562 people with type 2 diabetes mellitus, a total of 1000 patients with mean duration disease of 11.2 years, who had completed information in their documents, were studied. All people were free from complications at baseline. Mean cumulative incidence of diabetes-related complications over 10 years were 10.9 ± 3.5%, 8.0 ± 3.1%, 4.6 ± 1.7%, 9.1 ± 3.6% and 2.3 ± 0.9% for peripheral neuropathy and diabetic foot ulcer, nephropathy, ophthalmic complications, cardiovascular disease and death, respectively. People with better glycemic control had less complication and also related expenditures. Average para clinic cost per patient was 393.6 ± 47.8 and average inpatient cost per patient was 1520.7 ± 104.5 USD. Conclusions Our findings demonstrate considerable incidence of diabetes chronic complications and also high health care expenditure for related complications among our patients. As the number of people with diabetes continues to rise, early detection of the disease and implementation of timely and appropriate therapeutic strategies could decrease the burden of diabetes chronic complications and also huge related expenditures.
Background
The prevalence of type 2 diabetes mellitus is rising dramatically worldwide, mostly in developing countries. It is associated with highly cost-demanding complications [1][2][3]. In 2011, the number of people with diabetes mellitus was estimated to be 366 million. Present continuous trend would lead to 10% of the world population with diabetes by 2030 [4]. The data represented by the National Survey of Risk Factors for Non-Communicable Diseases of Iran indicate that 7.7% of adults younger than 65 years had type 2 diabetes in 2008 [1]. Glycemic control is the main step of treatment in people with type 2 diabetes and could be reached via nutritional therapy [5,6], physical activity, oral glucose lowering drugs (OGLDs), and insulin therapy [7,8]. Poor glycemic control results in chronic complications including microvascular (peripheral neuropathy and foot ulcer, nephropathy and retinopathy) and macrovascular diseases (cardiovascular, cerebral and peripheral vascular disease) [9,10] with profound impact on the quality of life of millions of patients and their families [11]. In addition, diabetes as one of the most costdemanding health conditions imposes a large economic burden to the healthcare system, most of which is the monetary value associated with disability and loss of life as a result of the disease itself and its related complications [12]. Oglesby et al. reported that diabetes related costs were 16% and 20% lower in people with good glycemic control compared with those with fair glycemic control and poor glycemic control [13]. Menzin et al. found that people with a mean A1C ≥ 10% had higher diabetes related hospital costs than those with a mean A1C <7% [14]. It has been estimated that global health expenditures of diabetes and its complications were about 376 billion US dollar (USD) in 2010, while a rise to 490 billion USD in 2030 is anticipated [15]. Although the results from previous studies have assumed costs, disability and loss of life associated with the disease in people with diabetes [16,17], to the best of our knowledge, there is not any report on incidence of diabetes chronic complications and concomitant costs in the routine clinic practice in Iran. So, the incidence and cost of diabetes chronic complications in Iran are still under debate.
The aim of this study was to evaluate the incidence of micro-and macrovascular complications as well as to analyze the costs of complications in people with type 2 diabetes in the routine clinical practice.
Methods
In a routine clinical practice people with type 2 diabetes (ICD-9/diagnostic code 250.x) were assessed retrospectively for 10 years from 1 December 2002 to 1 December 2012 at a public diabetes care center using paper based medical records. This center cares for over 1000 diabetic patients per month. People who had at least one followup visit per year were included. Demographic characteristics were recorded at the beginning of study. The type of the antidiabetic medications administered and clinical data, including hemoglobin A1C (HbA1c), fasting blood glucose (FBG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglyceride (TG), body mass index (BMI) and urine micro-albumin were extracted from patients' documents yearly. Mortality rate and incidence of microand macrovascular complications recorded in patients' documents were also reported. Diabetes chronic complications were defined as peripheral neuropathy and diabetic foot ulcer (ICD-9/diagnostic codes 250.6, 707, 736), nephropathy (ICD-9/diagnostic codes 580-586), ophthalmic complications (ICD-9/diagnostic codes 362.01, 365, 366) and cardiovascular disease (CVD) (ICD-9/diagnostic codes 401-405,410-413,414.0-414.8) (Table 1).
Patients were screened annually for peripheral neuropathy using history of pinprick sensation and tests such as vibration perception (using a 128-Hz tuning fork) and monofilament pressure sensation. Foot ulcer was also screened by medical examination annually. Urine albumin excretion test was assessed annually for diabetes nephropathy. The serum creatinine was used to estimate glomerular filtration rate (GFR) and to assess the stage of chronic kidney disease (CKD). Additional assessment was considered by a nephrologist if needed. Dilated and comprehensive eye examination was done by an ophthalmologist annually. Typical or atypical cardiac symptoms were asked and resting electrocardiography (ECG) was recorded annually. If there was any abnormality in history and resting ECG, people were referred for further assessments by a cardiologist. All additional notes regarding diabetes chronic complications in patients' documents were also analyzed and the first unmoral findings in mentioned tests were considered as the incidence of complication. Ethical approval was granted from the ethics board at Tehran University of Medical Sciences.
Laboratory measurements
After an overnight fasting of at least 12 hours, venous blood samples were drawn and were sent to hospital laboratory for analysis. All samples were analyzed by unique laboratory center. FBG concentrations were determined using the glucose oxidase method. Percentage of HbA1c was measured using the high performance liquid chromatography (HPLC) method [18]. Serum concentrations of HDL, LDL and TG were determined using enzymatic methods with available commercial kits (Pars Azmun, Karaj, Iran) in a Hitachi 704 automatic analyzer (Tokyo, Japan) [19].
Direct medical costs
Medical costs of each patient were obtained by a checklist from their documents, described in Table 2. Any pharmaceutical, laboratory/diagnostic and rehabilitative care as well as visiting specialists, general practitioners, nurses, opticians, podiatrists, and dieticians as well as any hospital admissions were recorded annually. All related costs were calculated using governmental prices and finally total costs were adjusted to 2012 USD.
Direct non-medical costs
A patient self-estimate questionnaire was used to assess non-medical expenditures due to services such as transportation for patients and their family to clinic and taking care of dependents.
Indirect costs
Costs of productivity loss due to diabetes -related health problems were determined annually by days absent from work, poor work performance, low earning's capacity from disabilities, and labor loss due to early mortality. Mortality costs were calculated as the lost earnings owing to premature mortality. We calculated the number of years of life lost and also the number of death in different age and gender group attributable to type 2 diabetes using our survival information. The number of days absent from work due to diabetes -related health care was recorded and the average of net daily wage was asked from each patient; then, the loss of value because of work absence was calculated using the formula below: Value loss due to absences from work (USD) = Number of days absent from work × Average of daily wage (USD).
To retrieve their indemnity claims during one year we used their insurance ID from the social security organization (SSO) database. At the end, all costs were categorized in two groups; para-clinic and inpatient costs. Costs from the health society perspective were converted from Iranian Rials (IRR) into USD at an official exchange rate of 12,260 IRR/USD 2012 in order to facilitate international comparison [20]. Average annual inflation rates of 21.5% in health care expenditures and 18.6% in gross incomes from 2002 to 2012 were applied to the costs [20].
Statistical analysis
Data are presented as Mean ± standard deviation (S.D.). Incidences were calculated using the whole number of people who experienced new complications yearly as the numerator and the total number of people alive during the same period as the denominator. Only the first event in a given year contributes to each rate and cumulative incidence of each complication was calculated. Considering age, gender and diabetic control status, stratification analysis was conducted. Subgroup analysis for status of glycemic control was defined in three categories: good control (A1C <7%), fair control (7% < A1C <8.5%) and poor control (A1C >8.5%). The assumptions of normal distribution as well as the equality of variance in subgroups were evaluated and if the assumptions failed, transformations were used to induce normality. Categorical and continuous variables were compared using Chi-square and Student's t test, respectively. Mixed between-within analysis of variance (ANOVA) and nonparametric tests were performed for between group comparisons with post-hoc followed test. P values < 0.05 were considered statistically significant.
Results
A total of 1562 people with type 2 diabetes from 1 December 2002 to 1 December 2012 were regularly referred to a clinical endocrinology and metabolism center in Tehran, Iran. Five hundred and sixty two people were excluded, and a total of 1000 people were assessed. The main exclusion criteria were incomplete documents, missed follow-up and gestational diabetes. Baseline characteristics and clinical parameters are shown in Table 3. There were no significant differences between men and women regarding baseline characteristics (P = 0.56). The mean age was 53.5 ± 9.4 years (M: 55.5 ± 8.1, F: 51.8 ± 7.6). Mean duration of diabetes was 11.2 ± 4.2 years in total population, whereas it was 4.7 ± 1.5, 9.2 ± 2.6 and 16.1 ± 5.5 in age category of <40, 40-60 and >60, respectively. After 10 years, clinical characteristics were improved (Table 3). Total diabetes related new events over 10 years were 289, 155, 103, 180 and 78 for peripheral neuropathy and diabetic foot ulcer, nephropathy, ophthalmic complications, CVD and death, respectively. So, yearly cumulative incidences of diabetes-related events over 10 years were 10.9 ± 3.5%, 8.0 ± 3.1%, 4.6 ± 1.7%, 9.1 ± 3.6% and 2.3 ± 0.9%, respectively (Table 4). All cumulative incidence rates were higher in people older than 60 years ( Figure 1). A significant gender difference was also noted (P = 0.023), with men showing a higher incidence rate in diabetes chronic complications (Table 4).
Stratification analysis according to the status of glycemic control is shown in Figure 2. Regarding the status of the glycemic control, 34.7% of the subjects achieved good glycemic control, while 35.0% were fairly controlled and the rest were still poorly controlled after 10 years. All people with better glycemic control had fewer incidences in diabetes chronic complications. Yearly death rate was: 0.2 ± 0.1 in people with good glycemic control; 1.4 ± 0.5 in those with fair glycemic control and 3.5 ± 1.8 in those with poor glycemic control (Table 4).
In people with poor glycemic control (A1C > 8.5%) significantly higher rates of complications were seen especially in the last years (P = 0.01) ( Table 4). In the last year 65%, 35%, 20% and 47% of people were suffered from peripheral neuropathy and diabetic foot ulcer, nephropathy, ophthalmic complications and CVD, respectively. The mean cumulative mortality rate was 2.3 ± 0.9% person per annum. Figure 3 shows the survival curve of people with type 2 diabetes after 10 years.
Medication
Drug treatment was shifted from oral glucose lowering drugs (OGLDs) to insulin and OGLDs plus insulin during the 10 years. The portion of people who were treated with insulin was 5.75% at the end of the first
Economic endpoints
Average direct medical cost for each chronic complication is presented with details in Table 2 inpatient costs were 1599.7 ± 59.8 for males. Our cost analysis indicated that CVD and nephropathy imposed the highest expenditure in the subjects. These complications and also related costs were remarkably higher in people older than 60 years. Disability and mortality expenditures were 1050.3 ± 78.4 and 872.5 ± 49.9 among men and women, respectively. Disability and mortality costs were higher in people with 40-60 years old (1220.4 ± 63.6). The highest total cost was attributed to the age category >60 rather than the other groups (P = 0.03). Our study showed that average annually cost of each patient older than 60 years was 2277.4 ± 58.9 (Table 5). Better glycemic control was associated with less expenditures and average total cost for people with good glycemic control (A1C <7%) was 409.0 ± 43.1 while it was 3788.1 ± 110.4 for people with poor glycemic control (A1C > 8.5%) ( Table 5).
Discussion
The current study evaluated the incidence of micro and macrovascular complications in people with type 2 diabetes. Results of the present study indicated that diabetes chronic complications, and total healthcare costs were affected by age, gender and glycemic control as higher incidence of complications and health care expenditure were observed in the aged males (>60 years), and people with poor glycemic control (A1C >8.5%). Consistent with the Morgan study [21], the incidence of microvascular complications, particularly retinopathy and nephropathy was correlated with duration of the disease in this study. Diabetes health problems affect the economically productive people in the age range of 40-60 in the Middle East [15]. We found that middle aged people (40-60 years old) had more disability costs due to diabetes health problems which emphasis the importance of attention to people with more productivity in the society. Costs and incidence rates in men were significantly higher than women which may be because men are unwilling to seek medical advice compared to women [22]. Previous reports indicated that tight glycemic control, as close as to non-diabetic glycemic range has been effective to reduce diabetes chronic complications and overall mortality [23]. Esposti [24]. Our study showed that better glycemic control could lead to less incidence of chronic diabetes complications and costs with the lowest diabetes related cost per patient in people with good control (A1C < 7%). The UK prospective diabetes study (UKPDS) reported that reduction in A1C level is likely to reduce the risk of diabetes chronic complications, with the lowest risk being in people with A1C levels in the normal range [8]. Considering the results of the current study, total per patient cost of illness for type 2 diabetes in Iran was 1914.3 ± 136.4 USD/year which is remarkably higher than the last report in 2009 (by 1707.4 USD/year) [17]. Health expenditure related to complications was 510 ± 65.3 USD, which is significantly greater than previous reports of 238.7 USD [25]. In a study conducted in Iran in 2009, it was revealed that 8.69% of total health expenditure was spent on diabetes; the estimated cost devoted was 3.78 billion USD comprising almost 2.04 billion direct (medical and non-medical) and 1.73 million indirect costs. In addition to the imposed costs of diabetes on the health system, it may reduce quality of life [17]. Therefore, a better perception of the burden may lead to the identification of more optimal treatment strategies, prevention at the individual level and prioritization of public health resources at the national level [26]. Therefore, achieving more precise data through switching to other interventions in future researches may help making more reliable decisions. Our findings demonstrated that the incidence of peripheral neuropathy and diabetic foot ulcer and CVD were higher than other diabetes-related complications.
Peripheral neuropathy and diabetic foot ulcer
It has been discussed that 50% of people with diabetes would suffer from neuropathy during 25 years after diagnosis [27,28]. In the current study, the yearly cumulative incidence rate of peripheral neuropathy and diabetic foot ulcer was 10.9 ± 3.5% that was the highest diabetes-related complication in total study population. This incidence seems remarkably high for people with type 2 diabetes from a developing country like Iran. Abbasian et al. reported that prevalence of neuropathy in Shahroud, Iran was 77.3% [29]. Mean duration of diabetes in our population was 11.2 years. One study of north-eastern Arizona found neuropathy in 21% of people with diabetes with more than 10-year duration of disease [30]. Consistent with the other studies our study confirms the association between the incidence rate of neuropathy, age and duration of diabetes [31]. Janghorbani et al. showed the prevalence of neuropathy was higher in elderly people rather the youngers [32]. In our study, drug treatment was shifted from OGLDs to OGLDs plus insulin or insulin alone during the 10 years. The prevalence and incidence of peripheral neuropathy seemed to be higher in patients treated with insulin rather than OGLDs-treated patients [32]. A higher incidence of peripheral neuropathy among insulin-treated people could be attributed to the longer duration of diabetes, delay in insulin treatment and possible insulin neuritis at the time of neurological examination. Some studies have presented evidence of the association between poor metabolic control and increased risk or incidence of neuropathy in type 2 diabetes [33]. In our study, mean cumulative incidence of peripheral neuropathy and diabetic foot ulcer per year was: 2.1 ± 1.1 in people with good control; 7.5 ± 2.8 in those with fair control and 21.1 7.4 in those with poor control. It is found that prevalence of peripheral neuropathy in diabetes was lower in people with good glycemic control (A1C < 7%) rather than those with poor glycemic control (A1C ≥ 9%) [32,34]. In a study over 3 years of observation, the attributable cost for a 40-65 year-old with foot ulcer was 27,987 USD for 2 years after diagnosis of diabetes [34]. The cost of peripheral neuropathy and diabetic foot ulcer was 77.9 ± 8.5 in diabetic people. We demonstrated that good glycemic control could be associated with less expenditure of peripheral neuropathy and diabetic foot ulcer care. So, good glycemic control with early and comprehensive neurological investigations for peripheral neuropathy in a patient with type 2 diabetes is necessary to have better management of disease and also decrease related costs.
Nephropathy
Nephropathy occurs in 20-40% of people 10 years after the onset of diabetes [35,36]. The incidence of nephropathy was found to be 8.0 ± 3.1% in our study, which leads to 143.9 ± 13.5 USD average cost of care in male and females. This is consistence with Amini's report that found the incidence rate of 8.2% for diabetic nephropathy [37]. Moradi et al. stated a prevalence of 25.6% for diabetic nephropathy [38]. Other studies among Iranian people have reported the diabetic nephropathy prevalences of 13.7 and 13.9% [29,39]. In our study, a positive association was found between the incidence of nephropathy and patients' age with the highest incidence in older people (>60 years). We found that better glycemic control (A1C < 7%) caused less incidence and also cost of nephropathy care. There are evidences that microalbuminuria is significantly associated with subsequent mortality from cardiovascular and coronary heart diseases [40]; thus, early screening for microalbuminuria and tight glycemic control, blood pressure and glomerular filtration rate in diabetic people is strongly recommended to improve the renal outcomes in order to decrease diabetic related renal diseases in the future.
Ophthalmic complications
In the current study, ophthalmic complications had the lowest incidence rate (4.6 ± 1.7%) rather than other diabetes chronic complications. Based on our results, 78.4 ± 8.9 USD of the health expenditure was devoted to ophthalmic complications in our population. This finding is in accordance with Javanbakht's report that ophthalmic complications had the lowest medical expenditure among type 2 diabetes chronic complications [17]. Additionally, diabetic retinopathy is a major cause of visual impairment and blindness among adults aged 20-74 years [41]. More than 60% of people with type 2 diabetes develop some degree of retinopathy after 20 years [42]. Due to changes in lifestyle and increase in ageing populations, the prevalence of ophthalmic complications is rising worldwide [43]. This is consistent with our study that people >60 years had higher incidence and cost of ophthalmic complications. This finding is supported by previous studies showing that prevalence of diabetic retinopathy is intensely related to aging and duration of the disease [21,23], though more than 90% of vision loss resulting from retinopathy can be prevented with appropriate glycemic control and ophthalmologic care [44]. Our study indicated with improvement in glycemic control (From A1C > 8.5% to A1C < 7%) a significant decrease was seen in incidence rate of retinopathy and also related health care costs in people with type 2 diabetes.
CVD
Mortality rate among diabetic populations is mostly attributed to CVD than microvascular complications [45]. CVD is the primary cause of morbidity and mortality in diabetic people [46]. Diabetic cardiovascular complications required the highest health resource services than any other chronic complication in 2012 in the U.S. (almost 20 million USD) [47]. In the present study, it was ranked as the second complication with a mean cumulative incidence rate of 9.1 ± 3.6%, imposing approximately 214.1 ± 12.8 USD to the patient yearly. Considering the health expenditure per capita in 2010 in Iran (317 USD) [48], CVD care expenditure in diabetic people seems to be very high. Khalili et al. reported that the incidence rate of CVD among Iranian people with type 2 diabetes was 1.2% [49]. The high incidence of CVD in our study might be attributed to the fact that this study was conducted in a referral care center. Previous studies have reported an association between the degree of hyperglycemia and increased risk of myocardial infarction [50], and stroke [51]. The UKPDS showed that a reduction in the risk of myocardial infarction was associated to lower value of A1C in people with type 2 diabetes while each 1% reduction in A1C level was associated with reductions in risk of 14% for myocardial infarction [8]. In consistent with UKPDS study, our finding showed lower incidence of CVD with good glycemic control. Considering the clinical and economic impact of cardio metabolic risk factors in diabetic people, it is clear that the strategy for controlling the costs should include modification of the cardio metabolic risks in people who have already developed CVD and diabetes.
Disability and mortality
During 10 years in the current study, the mean cumulative incidence rate of death was 2.3 ± 0.9%. Although the prevalence of diabetes is growing, data on mortality rates in affected populations are limited. A report conducted by Bertoni et al. revealed that diabetes was associated with excess mortality in elders and the rate increased with age [26]. As described in the literature, people with diabetes have lower rates of life expectancy and 85-yearold people spend only 32% of their remaining life active compared to 42% in non-diabetic people in the same age group [52]. In this study, the mortality rate was higher in men and death rates in both men and women increased significantly with age. UKPDS reported each 1% reduction in A1C was associated with 21% reductions in risk of diabetes related mortality [8]. Our results indicated that poor glycemic control resulted in higher rates of mortality (3.5 ± 1.8). However, it is clear that tight glycemic control is a fundamental factor to reduce mortality in diabetic people. The increased prevalence of diabetes among younger people suggests that it will become more common in the working-age population. It is estimated that diabetes causes a one-third reduction in earnings due to reduced productivity [53]. We showed that people in the workingage (40-60 years) have been affected more than both younger and older people; thus, the disability and mortality cost was significantly higher in this group. Tunceli showed that diabetes affected employers and society by reducing employment and also by contributing to work loss and related limitations for employed diabetic people [54]. We found that disability and mortality caused 1220.4 ± 63.6 USD costs yearly in 40-60-year age group with a significant higher expenditure in men while the lowest cost was seen in people with good glycemic control (A1C < 7%). Thus, good glycemic control as close as normal people leads to less disability and mortality in the society and have a profound impact on economic burden of diabetes.
Medication trend
In the current study, medication was shifted from OGLDs to OGLDs plus insulin or insulin alone during the 10 years. In the first year, 12.14% of people received OGLDs plus insulin and insulin solely while this figure reached to 47.88% in the 10th year. OGLDs are preferred as the initial choice of treatment by both patients and physicians [55]. However, insulin may be considered as the first-line treatment in in lean subjects or people with severe weight loss, acute illness or systemic underlying diseases such as renal or hepatic diseases [55]. Although patient's reluctance often delays the initiation of insulin therapy, the progressive nature of the disease and relative insulin deficiency over time will necessitate the initiation of insulin [55]. An analytical study on prescriptions in diabetic people in Iran revealed that insulin comprised 8%, 9%, 13% and 9% of all yearly prescriptions in 2006-2009, respectively. Whereas 59%, 66%, 71% and 72% of the prescription proportions were regarded to OGLDs in the same time period [56].
We found that glycemic control in older people is more difficult. On the other hand, the incidence rate of complication was higher in elderly people. Although guidelines recommend insulin therapy in uncontrolled glycemia [7], late insulin therapy causes poor glycemic control in elderly people. So, we recommended early insulinization in diabetic people to reduce diabetes-related complications. Considering increased prevalence of diabetes and the necessity of insulin therapy to reduce the risk of diabetes chronic complications it seems there is additional needs to insulin in Iran. To have a secure pharmaceutical market considering rate of disease, demands and resources [57] and also in order to prevent insulin shortage as a consequence of recent international sanctions, policy makers should pay more attention to local insulin production in Iran.
Limitations
Although this is the only study evaluating incidence of diabetes chronic complications and related costs in the routine clinical practice during 10 years in Iran, there are some limitations. As we did not have an integral data regarding the rate of hypoglycemia in the documents, the results of direct and indirect costs may be underestimated. In addition, as the data were collected from medical records, it is possible that missing data in medical records had led to an underestimation of the costs.
|
2017-04-26T03:04:35.065Z
|
2014-03-04T00:00:00.000
|
{
"year": 2014,
"sha1": "0a1eca7afc88cc61433a09a1a33bfdb16505f5e8",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1186/2251-6581-13-42",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "005b39445f63a7c735a36cab2a6cbb14a7d4a910",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
15571811
|
pes2o/s2orc
|
v3-fos-license
|
Relationship between gene expression and the accumulation of catechin during spring and autumn in tea plants (Camellia sinensis L.)
The tea plant (Camellia sinensis L.) is an important commercial crop with remarkably high catechin concentrations. Tea is popular worldwide given the plant's health benefits. Catechins are the main astringent substance in tea and are synthesized mainly via the phenylpropanoid pathway. In this study, eight cultivars of tea plants harvested both in spring and autumn were used to investigate differences in catechin concentrations by using high-performance liquid chromatography. The expression levels of genes associated with catechin biosynthesis were investigated using reverse transcription-quantitative polymerase chain reaction. The results indicated that the total catechin (TC) concentrations were significantly higher in tea plants harvested in autumn than in those harvested in spring, based on higher concentrations of epigallocatechin (EGC) in autumn tea (P<0.01). The expression of the genes phenylalanine ammonia-lyase (PAL), flavanone 3-hydroxylase (F3H), flavonoid 3′,5′-hydroxylase (F3′5′H), dihydroflavonol 4-reductase (DFR), and anthocyanidin synthase (ANS) is closely related to the TC content of tea plants in both spring and autumn. Positive correlations between PAL, cinnamate 4-hydroxylase (C4H), F3H, and DFR expression and EGC accumulation in autumn tea were identified, with correlation coefficients of 0.710, 0.763, 0.884, and 0.707, respectively. A negative correlation between ANS expression level and EGC concentrations in tea plants harvested in spring was noted (r=−0.732). Additionally, negative correlations between F3H and ANS expression levels and the catechin content were identified in spring tea, whereas the correlations were positive in autumn tea. Significant differences in the F3H and ANS expression levels between spring and autumn tea indicate that F3H and ANS are potentially key genes affecting catechin accumulation in tea plants.
INTRODUCTION
Tea is a globally economically important commodity and is enjoyed by people worldwide because of its health benefits. 1 Catechins are the primary astringent substances in tea. Catechins have multiple effects on human health and play important antibacterial, antiviral, anti-radiation, and anti-aging roles. In addition, these compounds are involved in the prevention of cardiovascular disease and cancer. [2][3][4][5][6] They are abundant in the young leaves and buds of tea plants and can be divided into non-esterified catechins (including catechin (C), epicatechin (EC), gallocatechin (GC), epigallocatechin (EGC)), and esterified catechins (including epicatechin gallate (ECG) and epigallocatechin gallate (EGCG)). 7 Catechins account for approximately 70% of all polyphenols in tea and are derived from multiple branches of the phenylpropanoid biosynthetic pathway, one of the most characterized secondary metabolic routes in plant systems. [8][9][10] Flavan-3-ols (also known as catechins) are mainly produced via the naringenin-chalcone R naringenin R dihydrokaempferol pathway (Ashihara et al., 2010). 11 As shown in Figure 1, the steps to produce dihydromyricetin are catalyzed by the following enzymes: phenylalanine ammonia-lyase (PAL), cinnamate 4-hydroxylase (C4H), chalcone synthase (CHS), chalcone isomerase (CHI), flavanone 3-hydroxylase (F3H), flavonoid 39-hydroxylase (F39H), and flavonoid 39,59-hydroxylase (F3959H). [12][13][14][15][16][17][18] Dihydroflavonol 4-reductase (DFR) catalyzes important steps in the control of metabolic fluxes, which feed into biosynthetic pathway branches, leading to the production of anthocyanins and proanthocyanidins. 19 Non-esterified catechins are produced in steps involving sequential reactions catalyzed by leucoanthocyanidin 4-reductase (LAR), anthocyanidin synthase (ANS), and anthocyanidin reductase (ANR). [20][21][22] EGC and EC are converted into esterified catechins (EGCG and ECG) via the sequential action of flavan-3-ol gallate synthase (FGS), 11 and various catechin monomers are synthesized from dihydrogen arbutus pigment. 10,11,23 To date, studies on catechin biosynthesis have mainly focused on the influence of light, 24,25 drought, 26 high ultraviolet (UV) radiation levels, 27 low temperature, 28 and pathogen infection, 29 as well as on the different maturing tissues, 30 the albino phenomenon, 31 and dark treatment. 32 In general, the quality of sweetness, flavor, texture, and taste in spring tea is better than that in autumn tea. 33 However, most previous studies have focused on the effects of environmental factors and alterations in amino acids, polyphenols, and caffeine content during seasonal changes in tea plants. 34,35 Chemical analysis of bud/leaves harvested in spring revealed a higher concentration of amino acids and aroma characteristics, but lower levels of total polyphenols and caffeine than those in tea plant leaves harvested in autumn. 34,35 Little information is available on the molecular biology of catechin accumulation and its biosynthesis-related gene expression in tea plants during the spring and autumn seasons.
In this study, we analyzed the catechin content and expression levels of genes involved in the biosynthesis of catechins in eight tea plant cultivars harvested both in spring and autumn (hereafter spring and autumn tea, respectively). These results provide insight into the possible mechanisms regulating catechin biosynthesis in tea plants and may lead to a better understanding of differences in catechin content between spring and autumn tea and the underlying genetic mechanisms.
Catechin concentrations
Catechin extraction was performed using 0.2 g of fresh tea bud leaves from each sample. 37 Catechins were analyzed according to Zhang et al. 38 with some modifications. Bud leaf tissue (0.2 g FW) was homogenized with a mortar and pestle and placed in a 10-ml centrifuge tube containing 6 ml of 70% (v/v) methanol. The tube was heated at 80 6 C for 20 min in a shaking thermal water bath and then centrifuged (Centrifuge 5810 R; Eppendorf, Hamburg, Germany) at 4 6 C and 10 000 rpm for 15 min. The supernatant was transferred into a new 10-ml centrifuge tube and filtered through a 0.45mm organic membrane; 70% (v/v) methanol was added to dilute the supernatant to 100 ml. A standard curve was prepared by weighing standard solutions of C, GC, EGC, EC, EGCG, and ECG purchased from Sigma-Aldrich (Sigma, Milwaukee, WI, USA) in 70% (v/v) methanol. 37 As a result, stock solutions containing all six catechins at 100 mg ml 21 , 50 mg ml 21 , 25 mg ml 21 , 12.5 mg ml 21 , and 6.25 mg ml 21 were prepared. Different concentrations of standard solution were prepared via dilution of the stock using the same solvent, 70% (v/v) methanol. Extracted samples and standards were stored at 4 6 C and protected from light before measurement.
High-performance liquid chromatography (HPLC) analysis was performed in an L-2000 HPLC-UV detector (Hitachi, Tokyo, Japan) with an injection volume of 10 ml and a C 18 ODS column (250 mm34.6 mm, 5 mm; Phenomenex, Tokyo, Japan) plus a C 18 guard column (10 mm33.0 mm) at an oven temperature of 30 6 C. Mobile phase A consisted of a 5:95 (v/v) mix of acetonitrile:double distilled water containing 0.05% (v/v) orthophosphoric acid; mobile phase B consisted of a 50:50 (v/v) mix of acetonitrile:double distilled water containing 0.05% (v/v) orthophosphoric acid. The flow rate was 0.5 ml min 21 , and the detecting wavelength was 231 nm. The gradient elution procedure is presented in Table 1. 38 Peaks were identified by comparing sample retention times to those of authentic standards.
RNA extraction and quantitative RT-PCR
Total RNA from bud leaves was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, USA, http://www.invitrogen.com) according to the manufacturer's instructions. First-strand cDNAs were synthesized using the FastQuant RT kit (with gDNase) (TIANGEN, Beijing, China, http://www.tiangen.com). According to the TIANGEN SuperReal PreMix Plus (SYBR Green) kits (TIANGEN, Beijing, China, http://www.tiangen.com), quantitative real-time (qRT) PCR was performed on a MiniOpticon Real-Time PCR system (Bio-Rad Laboratories Inc., USA). All primers used for qRT-PCR are listed in Table 2. After completion of the reactions, the threshold cycle (C T ) value for each reaction was recorded, and the fold difference in transcript level between greenhouse and open field samples was calculated. The mean values of three replicates and relative to a b-actin standard are given.
Statistical analysis of the data
The data are expressed as the mean 6 standard deviation (SD) for three replicates. Relative quantification values were calculated using the 2 2DDC T method. 39 The data were subjected to analysis of variance (ANOVA), and the statistical significance of differences between groups was assessed via Student's t-test using SPSS (Statistical Package for the Social Sciences) 16.0 statistical software (SPSS Inc., Chicago, IL, USA). For multiple variable comparisons, data were analyzed by two-way ANOVA followed by Tukey's test. The correlation was analyzed via Pearson correlation. The P-values between catechins and gene expression were analyzed by pair comparisons between eight cultivated varieties. P-values,0.05 were considered significant.
Differences in catechin concentrations between spring and autumn teas
Six characteristic tea catechins were successfully extracted from bud leaves ( Figure 2). The most abundant catechins included the esterified catechins EGCG and ECG, which accounted for approximately 60% of the total catechin (TC), and only low amounts of nonesterified catechins were noted in spring tea (Table 3). EGCG was the most abundant catechin in tea plants harvested in spring and autumn, and EC was the least abundant ( Figure 3). These results are consistent with a previous report indicating that EGCG is generally the most abundant tea catechin, whereas C and EC concentrations are rather low. 40 Tea plants harvested in autumn had higher EGC concentrations, followed by EGCG, C, and GC, but lower EC and ECG concentrations than in spring tea ( Figure 3). With the higher EGC concentrations, esterified catechins, including EGCG and ECG, only accounted for approximately 50% of the TC of tea plants harvested in autumn, which was 10% lower than that in spring tea (Table 3).
To explore the effects of cultivar, season, and cultivar 3 season on catechin concentrations in tea plants, data were analyzed by two-way ANOVA followed by Tukey's test. As shown in Table 4, significant differences in catechin concentrations were identified in eight tea plant cultivars and two seasons (spring and autumn). A significant difference in cultivar 3 season was also noted. These results indicate that different cultivars and seasons affect catechin accumulation in tea plants.
Expression of relevant genes
The expression patterns of flavonoids (including catechin) and genes that are involved in the biosynthesis of catechins, such as PAL, C4H, CHS, CHI, F3H, F39H, F3959H, DFR, LAR, ANS, and ANR, were examined. The relative gene expression in spring tea and the correlation coefficients (r) are listed in Table 5. ANS expression exhibited a significantly negative correlation with EGC accumulation, but F39H and DFR expression were significantly positively correlated with EGCG accumulation. The relative expression levels of PAL, DFR, and LAR in tea plants harvested in spring were significantly positively correlated with increased TC, whereas F3H, F3959H, and ANS gene expression were negatively correlated with increased TC ( Table 5).
As shown in Table 6, EGC concentrations were closely associated with PAL, C4H, F3H, and DFR expression in autumn tea. A significantly positive correlation between PAL and DFR expression and EGCG accumulation was noted. PAL, C4H, F3H, DFR, and ANS expression levels in tea harvested in autumn were significantly positively correlated with increased TC. In contrast, F3959H expression was negatively correlated with increased TC.
From Tables 5 and 6, correlations between catechin content and PAL, F3H, F3959H, DFR, and ANS gene expression were observed in the bud leaves of eight varieties in spring and autumn. A positive correlation between TC and LAR expression (r50.707) was observed in spring tea, but a positive correlation was noted between C4H expression and TC levels in autumn tea. F3H and ANS expression were downregulated in autumn tea as compared with spring tea, and catechin accumulation was higher in autumn tea than in spring tea ( Figure 4). Catechin concentrations were negatively correlated with F3H and ANS expression levels in spring tea but positively correlated in autumn tea (Tables 5 and 6).
DISCUSSION
Differences in catechin concentrations between spring and autumn tea Catechins are the main astringent substances in tea and are important components of tea quality. Of the various catechins, ECs (including EC and its gallolyl derivatives EGC, ECG, and EGCG) constitute approximately 90% of the TCs in tea leaves. 41 In this study, the TC concentration increased by 2.29 mg g 21 in tea plants harvested in autumn (Table 3). Nagata and Sakai 42 reported that the order of catechin concentrations in tea leaves (C. sinensis var. sinensis) was EGCG. EGC.ECG from highest to lowest. In Assam tea (C. sinensis var. assamica) leaves, the order was EGCG.ECG.EC. These different results indicate that the composition of each catechin substance varies in different cultivars. EGC concentrations in the eight cultivars of tea plants harvested in autumn in this study were significantly higher than in those harvested in spring ( Figure 3). Thus, we hypothesized that the higher EGC and EGCG concentrations in autumn tea might be the central reason for its higher catechin levels (P,0.01). EGC, an epimer of EGCG at the C-2 position, is easily detected in tea infusions. In our research, tea harvested in autumn contains higher EGC concentrations ( Figure 3). Sano et al. 43 reported that EGC is produced from the epimerization of EGCG via heat treatment. This finding may explain why the EGC concentrations were higher in autumn tea than in spring tea. In addition, catechins are classified as dihydroxylated catechins or trihydroxylated catechins based on the number of hydroxyls in the B-ring. The ratio of dihydroxylated to trihydroxylated catechin ((EC1 ECG):(EGC1EGCG1GC)) and the level of TC can be used as indicators of better tea quality. 40 The higher the ratio, the better the tea quality. In this research, the ratio decreased gradually and ranged from 0.10 to 0.37 in autumn tea. Ortho-dihydroxy B-ring-substituted flavonoids may inhibit the generation of free radicals through the chelation of metal ions. 27,44,45 These unique properties of flavonoids with a catechin group in the B-ring may explain why the ratio of dihydroxy to trihydroxy B-ring-substituted catechins decreased in autumn tea. Based on Table 4, we concluded that season of tea harvest affects catechin accumulation in tea plants. These results indicate improved tea catechin quality in teas harvested in spring. In addition, the quality of taste of the tea harvested in spring was better than of that harvested in autumn, probably owing to the higher EGC content in autumn tea.
The relationship between gene expression and catechin accumulation Tea is an important commercial crop known for its flavonoid compounds, such as catechins, which are important in beverages and medicinal use. We found that cultivar, season, and cultivar 3 season Table 4). The results indicate that the PAL, F3959H, and DFR genes, which are involved in the catechin biosynthesis pathway, are positively correlated with catechin concentrations in tea bud leaves harvested both in spring and autumn. F3H and ANS expression decreased in autumn. These results indicate that the season plays a very important role in regulating the expression of genes related to catechin biosynthesis, such as F3H and ANS. Flavan-3-ols (also known as catechins) are mainly produced via the naringenin-chalcone R naringenin R dihydrokaempferol pathway. 11 Eungwanichayapant and Popluechai 30 reported that increased tea catechin concentrations are attributed to the increased expression of genes involved in catechin biosynthesis. As shown in Figure 1, PAL catalyzes the deamination of phenylalanine to produce trans-cinnamic acid, which is converted to p-coumaric acid via an oxidative reaction catalyzed by the cytochrome P450 enzyme C4H. Research indicates that the C4H gene is involved in the catechin pathway and that its expression is associated with catechin accumulation. 14,46 F39H, F3959H, and DFR catalyze the reduction of flavanones to leucoanthocyanidin. 12,15-18 C and EC are hydroxylated by the F3959H gene. [47][48][49] F3959H exhibits a 394959-hydroxylation pattern, such as GC, EGC, and EGCG. In addition, the 39,49-dihydroxylation of the B-ring, as is evident in quercetin, substantially increases the antioxidant activity of flavonoids as compared with B-ring monohydroxylated flavonols. DFR is a key enzyme in the catechin flavonoid pathway and catalyzes important steps in the control of metabolic fluxes that feed into biosynthetic pathway branches, thus producing anthocyanins and proanthocyanidins. 19 DFR is overexpressed in calluses harvested in sufficient light 50 and bud leaves 51,52 but downregulated in leaves grown in the absence of light. 24 Therefore, tea catechin biosynthesis is critically dependent upon the products of these enzymes. 46 Our research demonstrated that PAL, F3959H, and DFR exhibited the same positive correlation with TC concentrations in tea bud leaves harvested in spring and autumn (Tables 5 and 6). This result indicates that these three genes may serve as core factors in the control of catechin biosynthesis in tea plants regardless of the harvest season.
LAR is the only enzyme that has been found to catalyze the conversion of leucocyanidin to C in many plants. 53 A negative correlation between C content and LAR expression level was also observed in the bud leaves of the eight different tea varieties harvested, both in spring and autumn (Tables 5 and 6). We also found that LAR transcripts decreased and C and GC concentrations increased in autumn tea ( Figure 3).
TC concentrations in autumn tea were significantly higher than in spring tea given the greater EGC content in the eight tea plant cultivars harvested in autumn (Figure 3). High EGC concentrations in green tea have been reported by Eungwanichayapant and Popluechai. 30 Based on the correlation analysis of catechin levels and the expression of genes involved in catechin biosynthesis, we observed positive correlations between PAL, C4H, F3H, and DFR expression and EGC accumulation in tea plants harvested in autumn, with correlation coefficients of 0.710, 0.763, 0.884, and 0.707 (Table 6), respectively; only ANS was negatively correlated with EGC concentrations in tea plants harvested in spring (Table 5).
quantities of flavonoids and also plays an important role in resistance to biotic and abiotic stresses. 55 F3H expression is controlled and regulated by catechin levels. 49,56 In the present study, a positive correlation between EGC and TC concentrations and F3H expression was noted in the bud leaves of eight cultivars harvested in autumn, whereas a negative correlation was evident in spring.
Zhang et al. 57 demonstrated that F3H expression was increased in Reaumuria trigyna under drought as well as cold stresses. The materials were harvested in northern subtropical climate zones; a notable feature of these areas is the monsoon. The amount of sunshine and precipitation in winter and spring seasons is less than that in the summer and autumn seasons. These climatic characteristics potentially explain why F3H expression was markedly downregulated in the tea bud leaves of eight cultivars harvested in autumn ( Figure 4). Thus, EGC may be hydroxylated by F3H, which may serve as a key gene in the control of catechin concentrations in tea plants harvested in different seasons. Therefore, further studies are required to better understand the relationship between F3H gene expression and catechin accumulation in tea plants harvested in autumn.
The metabolic genes involved in EGC biosynthesis include ANS and ANR. ANS expression results in the accumulation of ECs and is a key enzyme at the branch points of catechin biosynthesis. Through the catalysis of ANS and ANR, leucoanthocyanin is transformed into 2,3-trans-flavan-3-ols, such as EC and EGC. 11 Hong et al. 32 demonstrated that ANS is the key enzyme at the branch points of catechin biosynthesis, which results in the accumulation of ECs. They also found that darkness reduced ANS expression and EGC accumulation. In our research, a positive correlation was noted between EGC and TC concentrations and ANS expression in the bud leaves of eight cultivars harvested in autumn, whereas a negative correlation was evident in spring. ANS expression was downregulated in autumn tea as compared with spring tea, and higher EGC concentrations were observed (Figures 3 and 4). In contrast, EC and ECG concentrations were lower in autumn tea (Figure 3). It is a novel finding that different harvest seasons exert different effects on the expression patterns of genes involved in the phenylpropanoid pathway. This finding supports the hypothesis that ANS may be a critical gene involved in catechin accumulation. The facilitation of EGC production necessitates the adjustment of EC and ECG biosynthesis in the opposite direction. This finding demonstrates how tea plants maintain the balance of phenylpropanoid metabolism in response to environmental cues. Studies in which the temperature and light duration are varied would be required to assess the exact function of ANS in tea plants harvested in autumn.
CONCLUSIONS
In summary, this study was the first to examine the expression of most of the catechin biosynthesis pathway genes (with the exception of FGS) and measure the changes in catechin concentrations in the leaves of eight tea plant cultivars harvested in spring and autumn. Our results explain the basic causes of the increased astringency of autumn tea. These results suggest that F3H and ANS are the most important genes involved in the catechin biosynthetic pathway in tea plants. In addition, negative correlations between F3H and ANS expression and catechin levels were identified in spring tea, whereas a positive correlation was observed in autumn tea. We hypothesize that the suppression of ANS and F3H expression potentially alters catechin biosynthesis in tea plants during spring and autumn. Catechin accumulation was markedly increased in autumn tea.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
|
2016-08-09T08:50:54.084Z
|
2015-04-01T00:00:00.000
|
{
"year": 2015,
"sha1": "dd1fd88ca7bf51d61dba337f9e514ba7fead2e99",
"oa_license": "CCBYNCSA",
"oa_url": "https://www.nature.com/articles/hortres201511.pdf",
"oa_status": "GOLD",
"pdf_src": "Nature",
"pdf_hash": "dd1fd88ca7bf51d61dba337f9e514ba7fead2e99",
"s2fieldsofstudy": [
"Agricultural And Food Sciences"
],
"extfieldsofstudy": [
"Biology",
"Medicine"
]
}
|
11063636
|
pes2o/s2orc
|
v3-fos-license
|
Reviewing and identifying amino acids of human, murine, canine and equine TLR4 / MD-2 receptor complexes conferring endotoxic innate immunity activation by LPS/lipid A, or antagonistic effects by Eritoran, in contrast to species-dependent modulation by lipid IVa
There is literature evidence gathered throughout the last two decades reflecting unexpected species differences concerning the immune response to lipid IVa which provides the opportunity to gain more detailed insight by the molecular modeling approach described in this study. Lipid IVa is a tetra-acylated precursor of lipid A in the biosynthesis of lipopolysaccharide (LPS) in Gram-negative bacteria. Lipid A of the prototypic E. coli-type is a hexa-acylated structure that acts as an agonist in all tested mammalian species by innate immunorecognition via the Toll-like receptor 4 (TLR4)/myeloid differentiation factor 2 (MD-2) receptor complex. In contrast, lipid IVa is proinflammatory in mouse cells (agonism) but it remains inactive to human macrophages and even antagonizes the action of potent agonists like E. coli-type lipid A. This particular ambivalent activity profile of lipid IVa has been confirmed in other mammalian species: in equine cells Lipid IVa also acts in a weak agonistic manner, whereas being inactive and antagonizing the lipid A-induced activation of canine TLR4/MD-2. Intriguingly, the respective TLR4 amino acid sequences of the latter species are more identical to the human (67%, 68%) than to the murine (62%, 58%) ortholog. In order to address the unpaired activity-sequence dualism for human, murine, canine and equine species regarding the activity of lipid IVa as compared to LPS and lipid A and, we review the literature and computationally pinpoint the differential biological effects of lipid IVa versus LPS and lipid A to specific amino acid residues. In contrast to lipid IVa the structurally related synthetic compound Eritoran (E5564) acts consistently in an antagonistic manner in these mammalian species and serves as a reference ligand for molecular modeling in this study. The combined evaluation of data sets provided by prior studies and in silico homology mapping of differential residues of TLR4/MD-2 complexes lends detailed insight into the driving forces of the characteristic binding modes of the lipid A domain in LPS and the precursor structure lipid IVa to the receptor complex in individual mammalian species.
yields lipid A still capable of inducing endotoxicity (15 in [3]), even if exceptions have been reported [4]. The innate immune system mediates very effective recognition of invading bacteria on a molecular level by receptor/sensor proteins localized at the cell surface and intracellular sites. Due to this high affinity binding of response triggering bacterial molecules at picomolar concentrations, practical laboratory work is driven to the cutting edge of what can be achieved technically concerning isolation, analysis, purification or contaminants detection. Hence, when interpreting bioactivities of LPS, lipid A and LPS/lipid A substructures it matters if they are obtained from natural sources or chemical synthesis [5]. As revealed in the last two decades, two accessory extracellular proteins, LPS-binding protein (LBP) and CD14 significantly contribute to the extreme sensitivity of mammlian innate immunity to LPS by specific extraction of a single LPS moiety from endotoxin aggregates or bacterial membranes and its transfer to the TLR4/MD-2 heterodimer [6,7].
In LPS of wild-type enterobacteria an inner and outer core region and the strain-specific O-specific chain have been defined in the polysaccharide region based on evolutionary variation. In a set of specific enterobacterial glycosyltransferase mutants diplaying a rough (R)-type colony form only partial poly/oligo saccharide structures are expressed.
CSBJ
Abstract: There is literature evidence gathered throughout the last two decades reflecting unexpected species differences concerning the immune response to lipid IVa which provides the opportunity to gain more detailed insight by the molecular modeling approach described in this study. Lipid IVa is a tetra-acylated precursor of lipid A in the biosynthesis of lipopolysaccharide (LPS) in Gram-negative bacteria. Lipid A of the prototypic E. coli-type is a hexa-acylated structure that acts as an agonist in all tested mammalian species by innate immunorecognition via the Toll-like receptor 4 (TLR4)/myeloid differentiation factor 2 (MD-2) receptor complex. In contrast, lipid IVa is proinflammatory in mouse cells (agonism) but it remains inactive to human macrophages and even antagonizes the action of potent agonists like E. coli-type lipid A. This particular ambivalent activity profile of lipid IVa has been confirmed in other mammalian species: in equine cells Lipid IVa also acts in a weak agonistic manner, whereas being inactive and antagonizing the lipid A-induced activation of canine TLR4/MD-2. Intriguingly, the respective TLR4 amino acid sequences of the latter species are more identical to the human (67%, 68%) than to the murine (62%, 58%) ortholog. In order to address the unpaired activity-sequence dualism for human, murine, canine and equine species regarding the activity of lipid IVa as compared to LPS and lipid A and, we review the literature and computationally pinpoint the differential biological effects of lipid IVa versus LPS and lipid A to specific amino acid residues. In contrast to lipid IVa the structurally related synthetic compound Eritoran (E5564) acts consistently in an antagonistic manner in these mammalian species and serves as a reference ligand for molecular modeling in this study. The combined evaluation of data sets provided by prior studies and in silico homology mapping of differential residues of TLR4/MD-2 complexes lends detailed insight into the driving forces of the characteristic binding modes of the lipid A domain in LPS and the precursor structure lipid IVa to the receptor complex in individual mammalian species. The prototypic E.coli Lipid A shows a hydrophobic region composed of six (hydroxy-) acyl chains of 12 and 14 carbon atoms (panel B). In addition, five experimental values of reduction in human monocyte activation due to the lack of the indicated structural elements are given (panel B). The tetra-acylated biosynthetic precursor Lipid IVa of mammalian LPS/lipid A and its synthetic analogue compound 406 are displayed (panel C) next to the tetraacyl compound Eritoran (E5564) (panel D). See text for details. More in detail, within this hexa-acylated structure a subset of four 3-hydroxymyristoyl (3-OH-C14) residues is attached directly to the β-D-glucosaminyl-(1,6)-β-Dglucosamine backbone by two amide and two ester bonds at positions 2/2´and 3/3´, respectively, and the 3-OH-groups of both of the ´primary´ residues at positions 2´and 3´on GlcN II are further esterified to a lauroyl (C12) and a myristoyl (C14) group, respectively. The two phosphate residues of the lipid A's backbone differ -glycosidic linkage to the reducing monomer (GlcN I) in position C1 of the disaccharide scaffold and the other is ester-bound to position 4' of the nonreducing pyranose unit (GlcN II) [9]. The (-) sign marks the 2 and 6 negative charges in lipid A and the inner core, respectively. Additionally given (Figure 1, panel B) are the approximate numerical values describing the relative reduction of human monocyte activation due to the lack of the indicated structural elements as measured by comparing the in vitro cytokine induction activities of the corresponding synthetic partial structures to complete lipid A (compound 506).
TLR4 / MD-2 receptor complexes
As compared to the ubiquitous activation of mammalian TLR4/MD-2 signaling by enterobacterial LPS or lipid A, particular lipid A substructures like the tetra-acylated biosynthetic precursor Lipid IVa or its synthetic analogue compound 406 act either as antagonists or weak receptor agonists in a species-dependent manner ( Figure 1, panel C). In contrast, the tetra-acyl compound Eritoran (E5564) acts as a TLR4/MD-2 receptor antagonist in all mammalian species investigated to date ( Figure 1, panel D).
The lipid A-specific interaction between the complex of the solenoid TLR4 ectodomain and MD-2 with LPS induces a rearrangement and dimerization to an "m-shaped" signaling complex of two TLR4/MD-2/LPS units [8] ( Figure 2). This LPS/lipid Ainduced formation of the dimeric (TLR4/MD-2/LPS)2 complex on the cellular surface constitutes a key step to activate the innate immune system in mammalian species [9,10]. In addition, X-ray crystal structures of human MD-2 either alone or in an 1:1 association with a partial structure of the human TLR4 ectodomain have shown that the tetra-acylated ligands lipid IVa and Eritoran also bind to the central binding cleft of human MD-2, but in a largely different and thus non-agonistic orientation as compared to the lipid A domain of LPS [11,12]. Furthermore, x-ray structural data of the radioprotective 105 kDa (RP105) ectodomain/ myeloid differentiation factor 1 (MD-1) complex representing a major negative feed-back-regulatory element of LPS-induced TLR4/MD-2-signaling have revealed a ligand-independent dimer formation of two RP105/MD-1 units in an inverse orientation as compared to the (TLR4/MD-2-LPS)2 complex [13,14].
In general, loss of one or both phosphate groups, underacylation and/or the replacement of single or multiple acyl residues within the characteristic pattern of five n-(hydroxy)myristoyl chains plus one nlauroyl residue by shorter acyl chains lead to partial or total loss of immunoactivity [5,[16][17][18][19][20][21][22][23][24][25][26][27][28][29][30]. The binding to TLR4/MD-2 is TLR4 / MD-2 receptor complexes completely lost when both of the two phosphate groups are removed indicating their importance for providing binding affinity in all partial structures of enterobacterial lipid A tested [16,24,31]. The activity is, however, preserved when a carboxyl group replaces the phosphate group on LPS [32]. This indicates a major contribution of negative charges in lipid A-binding to TLR4/MD-2 mediated by positively charged amino acids on TLR4 and MD-2. Another example of a lipid A derivative/partial structure with reduced overall negative charge is 4´-monophosphoryl lipid A (MPLA), which has been shown to display markedly reduced TLR4/MD-2 agonistic activity as compared to the diphosphorylated lipid A parent structure. Binding of these ligands to the receptor binding site leads to only partial activation of the TLR4/MD-2-connected intracellular signaling network of as compared to the full agonist (lipid A), thus resulting in incomplete signaling [28]. In comparison to lipid A, 4´monophosphoryl lipid A (MPLA) lacks one phosphate group and is therefore unable to contact some positively charged residues on the surface of both MD-2 and TLR4 [2,8]. Recently, it was observed that the presence or absence of the oligosaccharide core segment from Capnocytophaga canimorsus LPS modulates endotoxic potency (Table 1). The structural implications were also discussed based on molecular dynamics simulations of the liganded human MD-2 monomer with lipid A from C. canimorsus and E. coli [4].
Site-directed mutagenesis data of liganded TLR4/MD-2 complexes Various lipid A derivatives or analogous agents ( Figure 1) with a common amphipathic glycolipid scaffold have been described to represent LPS/lipid A-like activators (agonists) or TLR4/MD-2 signaling or inhibitors (antagonists) of LPS-induced cellular immunoactivities [2,25,28,49,50]. LPS receptor inhibitors are being developed as potential drugs for adjunct treatment of septic shock patients with Gram-negative septic infection (endotoxicity). Lipid IVa, a tetra-acylated lipopolysaccharide precursor in the biosynthesis of Escherichia coli or Salmonella lipid A, is an agonist in murine myeloid cells but it remains inactive to human macrophages and even antagonizes the action of potent agonists like E. coli-type hexaacylated LA. In contrast to lipid IVa, however, the synthetic compound Eritoran that also comprises of only four acyl chains is a potent TLR4/MD-2 receptor antagonist in human, murine and equine species and can be considered as an investigational drug against bacterial sepsis [2], (48,51,75 in [46]).
Site directed mutagenesis approaches have revealed involvement of species-specific residues in MD-2 [36] and TLR4 in agonistic/antagonistic activities of lipid IVa (Table 2) [8,37,51,52]. Hence these mutagenesis data indicate, that both MD-2 and TLR4 contribute to the species-specific response to lipid IVa [12,15,36,37,53,54]. For example Thr57, Val61, and Glu122 of mouse MD-2 may have an impact on activation of the murine receptor complex by lipid IVa [36]. TLR4 was also subject to a set of selective human/horse sequence conversion mutants like R385G The oligomerization state is different for the liganded and unliganded complexes and it also depend on the environment: In 2007 it was described that LPS binding induces the dimerization of hTLR4/MD-2 [55]. Crystallography (Table 3) showed unambiguously the dimeric structures of human (TLR4/MD-2/LPS)2 [48] and mouse (TLR4/MD-2/LPS)2 complexes [15]. In native gel electrophoresis experiments the complex was shown to be a dimer: m(TLR4/MD-2/Re-LPS)2. In contrast, when bound to lipid IVa the complex remained monomeric m(TLR4/MD-2/L4a) in solution according to the electrophoretic mobility shift data. However, the lipid IVa murine complex was dimeric in the crystal form: (TLR4/MD-2/L4a)2 [15]. The authors ascribe its missing dimerization in solution to weaker noncovalent forces in the dimerization interface, assuming that the dimerization would be enhanced in the membrane where movements are restricted to the cell surface to facilitate contacts. If this holds, then the structural events as observed by crystallography of mouse (TLR4/MD-2/L4a)2 and human MD-2/L4a monomer are in close keeping with the biological function of either agonism (mouse) or antagonism (in human cells).
As pointed out by Park and colleagues [8] the initial comparison of the human dimeric (TLR4/MD-2/Ra-LPS)2 crystal structure (PDB code: 3FXI) with the monomeric antagonist complexes of lipid IVa with human MD-2 (PDB code: 2E59) and Eritoran with a TLR4/MD-2 fragment (PDB code: 2Z65), revealed that the presence of the two additional (secondary) acyl residues in the lipid A domain of Ra-LPS is correlated with a relative upward shift of the diphosphorylated glucosamine backbone by approximately 5 Å towards the solvent area. This structural shift allows phosphate groups of LPS to contribute to receptor multimerization by forming ionic interactions with a cluster of positively charged residues in human TLR4 and MD-2. The results of the pairwise superimposition of this set of four crystal structures (Table 4) were visualized (Figure 2).
Dimerization and signaling
Mechanistically different aspects of signaling are in discussion: ligand-induced oligomerization, cytoplasmically driven selfassociation or agonistic dimerization [15,20,56]. LPS binding to the cell surface receptor TLR4 constitutes the extremely specific and effective agonistic stimulus for transmembrane signaling via connecting Toll/IL-1R endodomain (TIR) to mount an immediate immune response. Intriguingly, recent experiments on replacement of the TLR4 ectodomain with MD-2, CD14, monomeric fluorescent protein or a 24 kDa gyrase B fragment revealed a robust ligandindependent constitutive activation of signaling by the corresponding chimeric fusion proteins, comparable to the maximal stimulation of the receptor with LPS. As discussed by the authors this indication for an intrinsic dimerization propensity of the transmembrane and cytoplasmic domains of TLR4 and reveals a previously unknown function of the ectodomain in inhibiting spontaneous receptor dimerization, since the unliganded TRL4 receptor complex must be kept in an inactive state without release of undesired immune responses [57].
In 2009, Meng et al. published mutational observations about important residues on the TLR4/MD-2 protein complex which were animated through our SPL scripting in the 3D models (Table 5) [37]. Surprisingly, the equine residues are identical with human residues contrary to what could be expected from their activity differences concerning lipid IVa. It acts as an antagonist in human cell tests but as a full or partial agonist in mouse or horse systems, respectively.
TLR4 / MD-2 receptor complexes
More in detail, the backbones substructures of agonists were consistently found to display an orientation with the reducing glucosamine-1-phosphate unit facing the secondary (signaling) dimerization site at the open side of the MD-2 pocket, whereas the antagonists were consistently crystallized in the inverted backbone orientation, rotated (´flipped´) by 180° as compared to the agonistic ligands. Moreover, the hydrophilic backbone structures of all agonistic ligands have been found to be placed in a significantly upward shifted position relative to MD-2 surface as compared to the antagonists. Now, it is to say, that contrary to classical pharmacology, where drugs act as either (full or partial) agonists or antagonists, lipid IVa acts either as an antagonist or an agonist in a characteristic speciesdependent manner. Evidently, there is not a corresponding ´classical´ structure-activity relationship for lipid IVa, since the ligand structure is the same. However, a correlation has been revealed between the species-characteristic agonist versus antagonist activities of lipid IVa and the overall binding mode this particular ligand i.e. how deep lipid IVa is buried in the MD-2 pocket and which mode of backbone orientation is present along the binding cleft. Apparently, when this tetra-acylated ligand binds very deeply into the pocket, then the specific unit of one phosphate group (and/or any acyl equivalent nearby, KDO) is not exposed on the dimerization side of the TLR4/MD-2/L4a monomer in order to attract a second monomer. This species-related TLR4/MD-2/L4a monomer thus fails in attracting a second monomer to initiate dimerization and consecutive downstream signaling.
TLR4 / MD-2 receptor complexes
In this view, lipid IVa appears to be rather an imperfect agonist than an agent with dual activity. As a comparably lower affinity ligand, it can bind in species-dependent manner either in an agonistic orientation (with its more surface-exposed backbone as found in the murine receptor complex) or the antagonistic orientation (with the flipped or inversed di-phospho-di-glucosamine backbone as revealed for the monomeric human MD-2 complex).
The earlier reviewed literature attests that lipid IVa is an agonist in mouse but acts as an antagonist in human cells. It can be assumed, that binding of agonists lead to a dimerization of liganded Toll-like receptor 4 and myeloid differentiation factor 2 complexes which then triggers downstream signaling for anti-inflammatory response. Cytoplasmically-driven self-association was, however, also reported.
Based on the x-ray crystal structures of dimeric TLR4/MD-2*agonist complexes the presentation of the three protein chains in the region of the dimerization interface, i.e. TLR4 (chain A), MD-2 (chain B) and TLR4* (chain C, ´counter´ TLR4) as a triangular interaction zone ("wedge") is a most useful concept to analyze species differences and to predict mutational effects (Figure 3). In this structural selection the wedge-like area consists of three sides: two partial structures of TLR4 proteins contributing to the ´primary´ and ´seondary´dimerization interfaces and one MD-2 forming the connecting bottom line. On a molecular level the species-dependent activity profiles of lipid IVa are reflected by a concert of conserved and variable amino acids in their respective protein sequences in this wegde-shaped zone.
Dimerization and activity
The differential height of the phosphates above the wedge bottom (MD-2) directly reflects the activity changes between the species. Only a highly exposed phosphate group of the backbone is capable of linking two liganded TLR4/MD-2 units [8]. Hence, this site (labeled as "Pag" in Figure 3) is considered to play the key role for switching to agonism (Pag) from antagonism (Pan) and back. While the Pag site contacts the GlcN I phosphate group of all observed agonists, the Pan site interacts with the GlcN II phosphate group of all antagonists as the latter show a flipped backbone.
According to the binding models the equine residue pair eGly322A & eGlu344A at the TLR4 interface corresponds to hGlu321A & hGly343A or mLys319A & mLys341A of the human and murine systems, respectively (Figure 3). The former two pairs are interrelated in a homologous way. The murine pair differs by two cationic residues and stabilizes the phosphate group in its agonistic site. Moreover, this pair together with others (nonconserved hGly384 vs. mAla382 vs. eArg385) nicely explain the crystallographically observed shift in the wedge's leftmost phosphate groups from a MD-2 assisted antagonistic phosphate binding site (interacting hARG90C) toward the agonistic phosphate site.
TLR4 / MD-2 receptor complexes
Park et al. already mentioned the relevant phosphate bridging of liganded TLR4/MD-2 unit (chains A,C) toward a second or counter-TLR4 (chain B) in the wedge, a concert of interacting residues assists the dimerization interface ( Figure 3). It can be assumed that homodimerization is signaling relevant, since agonistic lipid IVa appears as (TLR4/MD-2/L4a)2 in mouse complexes [14] which parallels agonistic LPS bound to mouse and human crystal complexes [8,14]. Very close to the wedge lies a histidine-rich surface patch (never referred to) in the dimerization interface of the two TLR4 proteins (chain A and chain B of 3D template 3FXI [8], which is the counter TLR4, sometimes labeled as TLR4* in the literature). This dimerization interface probably plays a role in transmembrane signal transduction.
Conclusions
During eons only LPS has been relevant for evolutionary adaptation. During time any random mutations were kept in the genes if they were not detrimental to the LPS recognition (silent point mutations) with the present day consequence that all nonidentical residues in MD-2 and TLR4 sequences do not affect picomolar LPS recognition. However, this is not the case of synthetic lipid IVa which is also a biosynthetic interim in bacterial cells. The hitherto silent mutations for LPS in mammalians become relevant for lipid IVa. Mapping makes the interacting net of amino acids recognizable (Figure 3). The homodimerization process in mice is helped by a nonconserved nonpolar residue (mAla414B) sitting in opposition to what would be the antagonistic phosphate (Pan) on (counter) TLR4*. Thanks to the nonpolar alanine -anionic repulsion, the phosphate group becomes a theoretical "rejection group" and the phosphate group moves into its agonist position further up as an accessible alternative, which actually is the only solution in the wedge. In its linear elongated conformation the mArg434B can interact with the phosphate group in agonist position. This amino acid is not conserved in horse, but compensated through the presence of eArg385A which moves the phosphate group into an "in-between pose" between aforementioned agonistic phosphate in the upper left corner of the wedge and its antagonistic counterpart in its lower left corner.
Taken together the collected data in a more general view, binding of agonists to TLR4/MD-2 leads to formation of a dimeric (TLR4/MD-2/Ligand)2 complex that efficiently activates downstream signaling to activate the vertebrate innate immune response while antagonists act as competitive inhibitors of agonists by binding to the monomeric TLR4/MD-2 unit in a non-dimerizing and thus non-signaling, hence, unproductive manner. Drawn from the hitherto known crystal structures, there is a complex correlation between binding and biological function regarding the exact ligand positioning and in particular the stretched diphophosphorylated diglucosamine backbone spanning the width of the MD-2 unit reaching from one TLR4 to the other (counter) TLR4* in view of the changing agonistic versus antagonistic activities of complexed ligands. All agonist ligands have been shown to mediate a specific bridging of two TLR4/MD-2 subunits to dimeric complexes (TLR4/MD-2/Lig)2 in the crystal structures whereas the antagonists apparently do not provide the dimerization of TLR4/MD-2 complexes. A crucial finding is the recognition of a murine acidic residue in the otherwise basic vestibule of MD-2. It leads to a repulsion and phosphate shift into agonist position which enables the ligand backbone to bridge both TLR4/MD-2 units. The right corner of the wedge (Figure 3) is highly conserved and holds the GlcN II phosphate group of agonists or, GlcN1 phosphate of antagonists due to their flipped backbones. This is why liganded TLR4/MD-2 forms most likely the functional biological unit. Hence, it must be considered a "monomer" (strangely, a heteromonomer, so to speak). It constitutes the "attachment point" or devise for the association of another liganded TLR4/MD-2 unit to form the signaling dimer. . Schematic view of the wedge, a triangular space between the three interacting polypeptide chains, A, B and C corresponding to TLR4, secondary or counter-TLR4 (labeled TLR4*) and MD-2 proteins, respectively. The numbering of the TLR4 residues refers to the horse sequence. Due to deletions of a few TLR4 residues their numbers differ slightly at equivalent positions: up to equine position number 297: e=h=c=m+1; then from equine position 298 to 560: e = m+3 = h+1 = c+1 or h=m+2=e-1=c-1. For instance, comparison of the residues present in (equine) position 322A indicates repulsion effects for the agonistic phosphate group (Pag) of lipid IVa in human (Glu-321A) and canine (Asp-321A) complexes, but strong attraction for murine (Lys-319A) and less for equine (Gly-322A) systems. In mice, the positions 367B and 434B push the phosphate group into Pag. Position 367B destabilizes Pag as phosphate localization in human, equine and canine systems. On the wedge bottom, MD-2 has a cation-reach vestibule to accommodate (Pag-rejected) phosphate groups in Pan except for mice (mGLU122C). Surrounded by conserved residues, Pag/Pan always accommodates the other (second) phosphate group of lipid IVa. The Pan position is occupied by the GlcN II phosphate group of lipid IVa or Eritoran, while the GlcN I phosphate group of LPS / lipid A occupies Pag. In the highly conserved right corner of the wedge is the all species-shared Pag-Pan site holding the complementary phosphate groups of the backbones: either GlcN I phosphate group for lipid IVa and Eritoran, or GlcN II phosphate group for LPS and lipid A.
|
2017-06-10T17:43:49.357Z
|
2013-02-01T00:00:00.000
|
{
"year": 2013,
"sha1": "319a3a5e04bca5da50fefa8617c95bc95b6aa853",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.5936/csbj.201302012",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "319a3a5e04bca5da50fefa8617c95bc95b6aa853",
"s2fieldsofstudy": [
"Biology",
"Chemistry",
"Medicine"
],
"extfieldsofstudy": [
"Biology",
"Medicine"
]
}
|
246871579
|
pes2o/s2orc
|
v3-fos-license
|
Effect of Lung Protective Ventilation Combined With Flurbiprofen Axetil on Immune Function During Thoracoscopic Radical Resection of Lung Cancer
The decreased immune function of patients with lung cancer has always been the focus of clinical attention. However, the stress response caused by surgery, anesthesia and pain will further reduce the body's immune function and affect the prognosis of patients to a certain extent. It was found that both protective ventilation and flurbiprofen ester pretreatment could reduce the immunosuppression caused by stress response. In this study, 120 lung cancer patients treated with video-assisted thoracoscopic radical resection were divided into group A, group B, group C and group D, which were treated with conventional mechanical ventilation, lung protective ventilation, conventional mechanical ventilation + flurbiprofen axetil and lung protective ventilation + flurbiprofen axetil, respectively. The results showed that the levels of CD3+, CD4+, CD4/CD8+, and NK in groups A, B, and C were lower than T0 on T1, T2, and T3, while those indicators in group D were lower than T0 on T1 and T2 (P < 0.05). The above indicators in group D were higher than those in the other three groups on T1, T2, and T3 (P < 0.05). The above indicators were statistically significant compared with those in group A and group C, group B and group D, and group A and group B at T1, T2, and T3 (P < 0.05). The comparisons of CD3+, CD4+, CD4/CD8+, and NK among the four groups within different time groups, and the repeated - measures analysis of variance (repeated - measures ANOVA) showed that there were interactions among time, group, and between groups × within groups (P < 0.05). It was confirmed that lung protective ventilation combined with flurbiprofen axetil could alleviate the immunosuppression of patients undergoing thoracoscopic radical lung cancer, providing a new idea for clinical treatment.
INTRODUCTION
Lung cancer can be divided into two types of non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC), and 80% of patients with NSCLC. The 5-year survival rate of lung cancer patients is low, and in recent years, the onset age of lung cancer patients also tends to be younger (1). Thoracoscopic radical resection of lung cancer is currently an important means of treatment for lung cancer. But it needs to be performed under mechanical ventilation, which has a certain inhibitory effect on patients' autoimmunity, while narcotic drugs (especially opioids) also have an impact on immunity (2,3). However, immunity is closely related to the patient's ability to resist external pathogen infection, and the balance regulation of immune system also plays an important role in maintaining body homeostasis. Once the immune balance is out of balance, the body is very prone to infection, and autoimmune diseases. Therefore, alleviating the immunosuppressive effects of mechanical ventilation and anesthesia is of great significance for improving the prognosis of patients undergoing video-assisted thoracoscopic radical resection of lung cancer (4).
In recent years, studies have pointed out that lung protective ventilation and flurbiprofen ester can weaken the inhibition of the above factors on the immune system of patients. On this basis, we hereby studies the effect of combination of the two on immune function in patients undergoing radical lung cancer surgery, as reported below.
General Data
One hundred twenty patients with lung cancer who received thoracoscopic radical resection of lung cancer in our hospital from April 2017 to June 2020 were included as research objects and divided into 4 groups. The ages of group A, B, C, and D were (56. 15 The mean course of disease was (6.17 ± 1.52) months, (6.25 ± 1.37) months and (6.33 ± 1.29) months. There was no difference in the general data of the four groups(P > 0.05).
Inclusion Criteria
i) 18-69 years old; ii) The BMI was 18-28 kg/m 2 ; iii) Meeting the Class I-II criteria of the American Association of Anesthesiologists (5); iv) Sign informed consent form.
Exclusion Criteria
i) Combined with fever, cough and gastrointestinal ulcers; ii) Patients who took NSAIDs for a long time before entering the group; iii) Patients with combined history of consciousness disorder and mental disease; iv) Patients with pulmonary tuberculosis and bronchial asthma; v) Patients requiring blood transfusion during the operation; vi) Combined with kidney, liver, heart and other major organ dysfunction; vii) Taking glucocorticoids and immunosuppressive agents before operation; viii) Patients with coagulation abnormalities and severe endocrine diseases; ix) Allergic to drugs used in this study; x) The pathological type of lung cancer is not suitable for thoracoscopic lung cancer radical surgery.
Methods of Anesthesia and Mechanical Ventilation
Routine anesthesia: 0.5 mg penehycliane hydrochloride was intravenously injected 30 min before surgery, oxygen was inhaled via nasal catheter, peripheral venous access of the upper limb was opened, SpO 2 , HR and other indicators were detected. The non-operative radial artery puncture was performed under local anesthesia, and the invasive blood pressure was monitored. The lateral internal jugular vein puncture was completed under local anesthesia ultrasound guidance, and the CVP was maintained within the range of 5-10 cmH 2 O. Midazolam, etomidate, sufentanil and rocuronium were given intravenously at doses of 0.05 mg/kg, 0.2 mg/kg, 0.4 ug/kg and 0.8 mg/kg. Indwelling of the left double-lumen bronchocatheter was performed under laryngoscopy. Under the assistance of a laryngoscope, the left double-cavity bronchial catheter was placed. Localization was performed by fiberoptic bronchoscopy. Mechanical ventilation was performed with A5 anesthesia, and volumetric controlled ventilation mode was used. After the patient's position was changed, the indwelling position of the double lumen tube was observed again.
Group A underwent conventional ventilation: one-lung ventilation was performed at Vr8ml/kg and RR13-16 times/min; The bilateral lung ventilation rate was Vr10ml/kg, and RR10-12 times /min. Group B was treated with protective ventilation: Onelung ventilation was given at Vr6ml/kg and RR14-16 times/min. Two-lung ventilation was given at Vr8ml/kg and RR12-14 times /min. One-lung ventilation was performed with PEEP5 cmH 2 O, oxygen flow rate of 1-2 L/min, I: E ratio of 1:2, and FIO 2100%. PETCO 2 was maintained at 35-45 mmHg. Anesthesia was maintained by target controlled infusion of remifentanil and propofol, with target plasma concentrations of 2-4 ng/mL and 2-4 mg/mL, respectively. During infusion, the dosage and infusion speed were adjusted according to the arterial pressure to maintain the arterial pressure fluctuation to be ≤20% of the preoperative level. Before skin incision, 0.2 ug/kg sufentanil was given intravenously, while 0.05 mg/kg atracurium besylate was given intermittently during the process, to maintain the Narcotrend index within the range of 37-64. Also, 6 mL kg/h compound sodium lactate was given intravenously during the operation. All patients stopped drug administration at the completion of the operation. After the patients were conscious and the muscle strength recovered, the double-lumen endobronchial tube was removed. Meanwhile, the same scheme of analgesic pump was used for analgesia within 24 h after the operation. In addition, in the groups C and B, flurbiprofen axetil 2 mg/Kg was given intravenously 5 min before anesthesia induction.
Observation Indicators
T0-t4 was used to represent preoperative, post-operative, postoperative 24 h, 72 h, and 7 d. At the above time, the expression of CD3+, CD4+, CD8+ and NK cells was detected by FC500 flow cytometry, and CD4+/CD8+ was calculated. 2 ml venous blood was taken in the fasting state in the morning, and heparin anticoagulant blood was taken (1: 9) 100P1 was added with monoclonal antibody, kept away from light for 12 min at room temperature, centrifuged, washed twice by PBS, and then added with 0.5 mL 1% paraformaldehyde. Machine detection was performed. Homotype negative control was performed for each sample at the same time. The number of cells per sample was 10,000, and the percentage of positive cells was calculated.
Statistical Methods
All data were processed with SPSS 22.0 statistical software, and GraghPad prism 8 was used to make statistical graphs. Measurement data are expressed as mean ± standard deviation (X± s), the comparisons of four groups at different times were performed with repeated measures analysis of variance and F test. The count data between groups were expressed in percentage (%) and tested by "x 2 ". The difference is statistically significant when P < 0.05.
Comparison of CD3+ Expression
CD3+ in Groups A, B, and C was lower than T0 on T1, T2, and T3, while it was lower than T0 on T1 and T2 in Group D; T1, T2, and T3 in group D were higher than those of the other three groups from T1 to T3. The CD3+ expressions on T1, T2, and T3 in group A and group C, group B and group D, and group A and group B were all statistically significant (P < 0.05). Repeated measures analysis of variance: F time = 121.201 (P < 0.001); F Group = 76.951 (P < 0.001); F Time×grouping = 65.150 (P < 0.001, Table 1).
Comparison of CD4+ Expression
CD4+ in group A, B and C was lower than T0 at T1, T2, and T3, and lower than T0 at T1 and T2 in group D. The CD4+ of group D was higher than that of the other three groups at T1, T2, and T3. The expression of CD4+ in group A and GROUP C, group B and group D, and group A and group B was statistically significant at T1, T2, and T3 (P < 0.05). The comparison of CD4+ in group A and group B at T1, T2, and T3 was statistically significant (P < 0.05). Anova of repeated measures: F time =89.113 (P < 0.001); F Group = 89.658 (P < 0.001); F Time×grouping =41.625 (P < 0.001, Table 2).
Comparison of CD4+/CD8+ Expression CD4+/CD8+ in groups A, B and C were lower than T0 on T1, T2, and T3, while CD4+/CD8+ in group D was lower than T0 on T1 and T2. Group D had higher CD4+/CD8+ values on T1, T2, and T3 than the other three groups. The comparisons of T1, T2, and T3 between group A and group C, group B and group D, and group A and group B were statistically significant (P < 0.05). Repeated measures analysis of variance: F time = 69.067 (P < 0.001); F Group = 49.167 (P < 0.001); F Time×grouping = 29.117 (P < 0.001, Table 3).
Comparison of NK Expression
NK in group A, B and C was lower than T0 at T1, T2, and T3, and NK in group D was lower than T0 at T1 and T2. NK in group D was higher at T1, T2, and T3 than in the other three groups. There were statistically significant differences between group A and GROUP C, group B and group D, group A and group B at T1, T2, and T3 (P < 0.05). Anova of repeated measures: F time = 59.621 (P < 0.001); F Group = 39.651 (P < 0.001); F Time×grouping = 23.780 (P < 0.001, Table 4).
Deficiencies of Conventional Mechanical Ventilation
Mechanical ventilation is an important supportive treatment in ICU. It can not only effectively maintain the airway patency of patients, improve oxygenation and ventilation, but also prevent carbon dioxide accumulation and hypoxia in the body, thereby enabling the body to avoid respiratory failure caused by basic lesions. However, many studies have confirmed that within 2-4 h of conventional mechanical ventilation, the susceptibility of patients to bacteremia is significantly higher than that of patients without mechanical ventilation. Excessive mechanical ventilation results in the accumulation of cytokines, white blood cells, and neutrophil-dependent tissue damage, resulting in cell activation and release of mediators, leading to alveolar inflammation (6). In addition, the observation of NK cell expression during conventional mechanical ventilation (10 ml/kg tidal volume) in infants without pulmonary diseases undergoing cardiac catheterization also showed that the activity of NK cells in peripheral blood began to decrease 2 h after the operation. The subjects of this study were lung cancer patients, and the results showed that the levels of CD3+, CD4+, CD4/CD8+, and NK in the four groups at T1 were lower than those at T0, which confirm that routine mechanical ventilation can adversely affect the patient's immune system.
Application of Lung Protective Ventilation
Lung protective ventilation strategies include appropriate PEEP and low tidal volumes (7). In animal experiments, the expression of NK cells in peripheral blood of mice with different mechanical ventilation schemes was compared and analyzed after 4 h of ventilation. It was found that high tidal volume ventilation could cause significant immunosuppression, and the decline degree of NK cells in mice with high tidal volume and without PEEP was more significant than that with high tidal volume and PEEP. And the combination of low tidal volume and PEEP could alleviate the immunosuppression caused by mechanical ventilation (8). In this study, the expression levels of the above indicators in group A were lower than those in group B from T1 to T3 (P < 0.05). This is due to compared with conventional mechanical ventilation, lung protective ventilation can alleviate alveolar-capillary barrier damage and inhibit inflammatory response. It was reported that 90% of patients with general anesthesia can appear atelectasis. During general anesthesia, low Comparison with T0, a P < 0.05. (11). In addition, if there are no contraindications, the use of positive end-expiratory pressure and lung recruitment can also help prevent end-expiratory lung volume loss and small airway closure during anesthesia (12). Although 10 ml/kg tidal volume was mostly used in clinical practice in the past, anesthesiologists would reduce tidal volume during single ventilation. Moreover, many studies have pointed out that a tidal volume of 4-5 mL/kg can better protect lung tissue while fully satisfying the gas exchange (13). The tidal volume selected for lung protective ventilation in this study belongs to the safe range (14).
Flurbiprofen Ester Alleviates Immunosuppression
Flurbiprofen ester is a non-steroidal analgesic drug, which can inhibit coX-2 release, prostaglandin synthesis, inflammatory factor release and other mechanisms through selective aggregation in surgical incision and inflammatory tissue, and exert targeted analgesic effect. It can reduce the dose of opioids, and is currently mainly used for cancer pain treatment, postoperative analgesia and preemptive analgesia et al. (15,16). In addition, compared with tramadol or morphine, flurbiprofen had the weakest immunosuppressive effect during postoperative analgesia (17). Previous studies have indicated that postoperative analgesia with flurbiprofen axetil can alleviate postoperative immunosuppression, and protect the immune function of cancer patients (18,19). Zhang et al. (20) pointed out that the decrease of CD4+, CD3+, CD4+/CD8+ and NK cell activity in sufentanil combined with flurbiprofen ester was lower than that in sufentanil alone. Anova of this study showed that group and time had impact on each indicator (P < 0.001), which suggesting that flurbiprofen ester could relieve immunosuppression caused by anesthesia or surgery.
In summary, lung protective ventilation combined with flurbiprofen axetil in video-assisted thoracoscopic lung cancer radical surgery can alleviate immunosuppression and facilitate postoperative recovery, which is worthy of promotion.
DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.
ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Committee of Zhoushan Hospital of Wenzhou Medical University. The patients/participants provided their written informed consent to participate in this study.
|
2022-02-17T14:29:13.387Z
|
2022-02-17T00:00:00.000
|
{
"year": 2022,
"sha1": "acdf49447fe5186ad5c3dedfb146db3b47b4a49c",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Frontier",
"pdf_hash": "acdf49447fe5186ad5c3dedfb146db3b47b4a49c",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
212946939
|
pes2o/s2orc
|
v3-fos-license
|
Scientific Landscape of Smart and Sustainable Cities Literature: A Bibliometric Analysis
: The smart sustainable city (SSC) is a concept created in response to problems and challenges arising from rapid urbanization. This is a relatively new term that is developing dynamically, which is confirmed by the growing number of publications over recent years. For this reason, this article presented an up-to-date comprehensive bibliometric analysis to describe and assess the scientific landscape of smart and sustainable cities literature. The analysis was based on two bibliographic sources—the Web of Science Core Collection and the Scopus database. It covers publications on the SSC, as well as documents describing the smart city (SC) and the sustainable city (SuC) concepts separately. VOSviewer and Biblioshiny were selected as software tools for the bibliometric analysis. Based on the descriptive bibliometric analysis, quantity and quality indicators were determined separately for the SC, SuC, and SSC concepts, while the network analysis mapped and covered the level of multi-faceted scientific cooperation in the field of the SSC research. The analysis results were intended to familiarize scholars and practitioners with the most prolific authors, sources, institutions, and countries in the analyzed scientific field, to identify the most influential research channels and impact from authors, sources, countries, and research topics, to determine major clusters of the SSC research and also to provide valuable information for further investigation.
Introduction
A city is a place where a large number of people live and work. It is a human settlement, which differs from others not only by its relatively great size but also by its functions entrusted by public administration authorities. According to the United Nations data, the number of city dwellers has increased significantly over the last century. In 2018, it totaled 55% of the world's population. It is expected that this percentage will continue to grow (reaching 68% in 2050), which will be the result of the ongoing gradual relocation of residents from rural to urban areas in parallel with the increase in the overall global population. Almost 90% of this increase will occur in peripheral countries [1].
People are moving to urban areas, hoping for better education and better job opportunities and seeking a higher standard of living. The growing number of city dwellers creates many problems and challenges that cities now have to face. These problems and challenges include: • growing demand for land designated for urban development, which results in a reduction in green areas and agricultural land, • growing environmental threats due to increased air pollution and bigger amounts of liquid and solid waste produced by people living in urban areas, • increasing demand for resources (water, food, fossil energy resources, etc.) due to the growing needs of urban residents, • increasing demand for public services, such as sanitation, public transport, health care, and education, • growing number of vehicles, causing congestion and traffic jams, as well as greater air pollution, • increasing the application of new technologies in the development and management of cities, which can improve the living conditions of some city residents, but can also socially exclude those incapable of adopting new solutions.
All these problems and challenges of city life are well-known, and yet people still want to move from rural to urban areas (especially in peripheral countries). In addition, the bigger number of city residents, combined with increased wealth of urban communities, will only aggravate the situation. Therefore, it has become essential to answer the question of how cities should develop and how they should be managed to meet the challenges of urbanization.
Following the sustainable development concept introduced at the end of the 20th century, the concept of sustainable city development was created. According to it, a 'sustainable city' (SuC) is a city organized in a way which 'enables all its citizens to satisfy their own needs and to enhance their well-being without damaging the natural environment or endangering the living conditions of other people, now, or in the future' [2]. Including the principles of sustainable development in the process of designing, building, and managing cities aims to ensure the economic development of cities and to reduce inequalities between citizens, to improve the quality of life and also the quality of the environment [3]. It should be noted that the United Nations Agenda 2030 for the 11th sustainable development goal (SDG 11) recommends making cities more sustainable, resilient, and inclusive [4].
The growing number of urban residents has caused a significant increase in the demand for urban services, which, in many cases, cannot be sufficiently and effectively provided using the existing infrastructure. For this reason, cities began to search for and introduce new intelligent technologies to fulfill their functions better. This resulted in the implementation of the 'smart city' (SC) concept. According to it, a city is smart if it uses clever solutions to thrive and improve the quantity and quality of public services provided to its residents [5,6]. The implementation of new intelligent technologies is considered a key factor in solving urban problems, and for that reason, the United Nations Agenda 2030 in the 9th sustainable development goal (SDG 9) recommends increased application of the information and communication technology (ICT) to build a resilient infrastructure with investments in irrigation, energy generation, and information and communication technologies [4]. However, it should be noted that although the implementation of new technologies can bring many benefits to cities, it can also negatively affect the quality of the environment or the quality of life of inhabitants [7]. It is, therefore, very important to find such innovative solutions and advanced technologies that will allow cities of the 21st century to be both smart and sustainable [8].
The above-mentioned approach has become the pillar of the 'smart sustainable city' (SSC) concept. According to it, the 'smart sustainable city is a city that: (1) meets the needs of its present inhabitants without compromising the ability of other people or future generations to meet their needs, (2) does not exceed environmental limitations in resource sharing and pollution assimilation, and (3) uses new information and communication technologies and advanced technologies to support the tasks entrusted to it' [9]. The concept has mainly developed in ecologically and technologically advanced countries [10], but given the fact that in the coming years the largest increase in urban areas will take place in peripheral countries, it is important that these countries should be able to benefit from the experience of developed countries in creating smart sustainable cities. Unfortunately, the literature and practice present varied definitions and dimensions of the SSC, which makes it difficult to understand the concept unequivocally. For this reason, it is necessary to conduct a thorough analysis to identify the most interesting research directions of the SSC concept development. Although several articles reviewing the literature on the SSC have been published in recent years, they focus primarily on reviewing the SSC definitions and dimensions [6,10,11], trying to find the answer to a research problem defined in relevant articles [12,13] or carrying out a bibliometric analysis in the field of the SC [14][15][16] or the SuC [17,18], based on a selected single bibliographic database. There are also publications presenting the results of a bibliometric analysis carried out in the field of the SSC [19], but they focus on general results from a few years ago, which undoubtedly needs expansion and updating due to the significant increase in the number of the latest SSC publications. Therefore, this paper presented a comprehensive quantitative evaluation of the SSC scientific landscape, based on the development of the SC and the SuC literature, in particular. The detailed and longitudinal bibliometric analysis of the SSC research filled the gap identified in rigorous, systematic, and objective examination of the global patterns and trends of research in this field from multiple perspectives. It was based on two bibliographic sources: the Web of Science (WoS) Core Collection and the Scopus database. Three types of bibliometric indicators were used: quantity indicators (for measuring productivity), quality indicators (for measuring the impact), and structural indicators (for measuring the connections) [20]. A network analysis (i.e., co-keyword analysis, co-authorship analysis, and co-citation analysis) was also performed, and major clusters of the SSC research were identified.
The results achieved in the analyzed domain would enable researchers and practitioners: • to know the most prolific authors, sources, institutions, and countries in the analyzed scientific field, • to identify the most influential research channels and impact from authors, sources, countries, and research topics in the SSC literature, • to determine how the SSC publications are clustered, • to provide valuable information for further investigation and to determine publication strategies.
The structure of this paper is as follows: Section 2 is an overview of the SC, SuC, and SSC definitions. The research methodology is described in Section 3. Section 4 presents the results and discussion of the bibliometric analysis. Finally, Section 5 offers conclusions and indicates possible directions for future research and limitations.
Towards Smart and Sustainable City-Literature Review
The SSC concept is derived from the combination of the SC and the SuC. It is conceptually difficult to describe, mainly due to the diversity of definitions of the SC and the SuC. Therefore, to present the essence of the SCC, it is necessary to explain the meaning of the SC and the SuC concepts.
The term 'smart city' (SC) was first used in the early 1990s in connection with the growing significance of new information and communications technologies (ICT's) and modern infrastructures within cities. It has gained greater attention since 2008 due to the launch of the IBM Smarter Planet project [21]. Since then, the concept has evolved, resulting in a variety of SC definitions. Being a multifaceted, wide-ranging, and rather fuzzy notion, it is used in ways that are not always consistent. According to Hall et al. [22], the SC 'monitors and integrates the conditions of all of its critical infrastructures ( . . . ), can better optimize its resources, plan its preventive maintenance activities, and monitor security aspects while maximizing services to its citizens'. Giffinger et al. [23] claimed that the SC is 'a city well performing in a forward-looking way in economy, people, governance, mobility, environment, and living, built on the smart combination of endowments and activities of self-decisive, independent, and aware citizens'. Harrison et al. [21] considered the SC as 'an instrumented, interconnected, and intelligent city, connecting the physical infrastructure, the IT infrastructure, the social infrastructure, and the business infrastructure to leverage the collective intelligence of the city'. Kourtit and Nijkamp [24] pointed out that SC is 'the result of knowledge-intensive and creative strategies, aiming at enhancing the socio-economic, ecological, logistic, and competitive performance'. Yigitcanlar et al. indicated that the SC is 'an ideal model to build cities based on a systemic approach, which includes sustainable and knowledge-based development activities to generate desired outcomes and future for all humans and non-humans' [13,25]. It seems to be evident that there is no commonly shared definition of the SC. However, there used to be two mainstream approaches to the concept: the approach based on ICT and technology, in general, and the people-oriented approach [26]. Nevertheless, the SC notion technocentricity has been criticized recently [13], and it is emphasized that the development of the SC should be based on a co-creative approach where citizens and other stakeholders shape SC services in a collaborative manner [27]. It must be noted that several similar terms are often used interchangeably with the smart city. These include: 'digital cities', 'virtual cities', 'cyber cities', 'networked cities', 'intelligent cities', 'knowledge cities', 'wisdom cities', 'ubiquitous cities', 'real-time cities', and 'hybrid cities', which combine these names [10]. However, these terms refer to more specific aspects of the city [28]. In this way, the SC concept is wider. It often includes the terms [29] and focuses on other issues [30].
The term 'sustainable city' (SuC) also emerged in the early 1990s. As no single definition of 'sustainable development' has been accepted, the SuC does not have one common explanation either. In addition, there are several similar notions often used interchangeably, including sustainable urban, city sustainability, urban sustainability, sustainable urban development, sustainable city development. According to Ewers and Nijkamp [31], the SuC is a city, which has the 'potential to reach qualitatively a new level of socio-economic, demographic, and technological output, which in the long run reinforces the foundations of the urban system'. Girardet [2] claimed that the SuC 'is organized to enable all its citizens to meet their own needs and to enhance their well-being without damaging the natural world or endangering the living conditions of other people, now, or in the future'. The United Nations defines the SuC as a city, which 'is built on social development, economic development, environmental management, and urban governance to ensure 'low ecological footprint' and eliminate the transfer of economic, social, and environmental hazards to other locations and future generations' [32]. Hiremath et al. [33] described the SuC as a city that has achieved a balance between urban development and environmental protection. In particular, it is a city that, now and in the future, can satisfy the basic needs of its inhabitants, benefiting all sectors of society [34]. Wang et al. pointed out that the SuC means 'improving the quality of life in a city, including ecological, cultural, political, institutional, social, and economic components without leaving a burden on the future generations' [18]. In general, the SuC integrates environmental, economic, and social considerations to achieve goals of sustainable development. It should provide a safe and healthy urban environment where both people and nature can develop. It must be noted that there are several terms similar to a sustainable city, including, for example, 'eco-cities', 'green cities', or 'resilient cities'. However, these notions refer to more specific aspects of the city, so SuC seems to be a wider, more complex concept.
The 'smart sustainable city' (SSC) is a new phenomenon, and this notion became widespread in the mid-2010s. Being relatively new, the phenomenon has been less explored than the SC and SuC concepts. In general, the SSC is 'a city that is supported by a pervasive presence and massive use of advanced ICT, which, in connection with various urban domains and systems and with how these intricately interrelate, enables cities to become more sustainable and provide citizens with a better quality of life' [35]. Dhingra and Chattopadhyay determined the SSC specific goals expected to be accomplished in an adaptable, reliable, scalable, accessible, and resilient manner [36]. Therefore, the SSC represents and involves 'inherently complex socio-technical systems of all sorts of innovation systems' [35]. In addition, Bibri [37] described the SSC as 'a holistic urban development approach, which seeks to explicitly bring together the sustainable city and the smart city as urban endeavors in ways that address and overcome the key shortcomings of both classes of cities in terms of their contribution to the goals of sustainable development'. The above-mentioned definitions indicate that the field of the SSC is inherently interdisciplinary, comprising technological, social, environmental, economic, cultural, and philosophical perspectives. This is due to the fact that it covers multidimensional issues related to the development of sustainability awareness, rapid urban growth, and technological development. However, research on the SSC is in its early stages, and it needs thorough clarification and operationalization of its characteristics. Nevertheless, the SSC means that SC must contribute to sustainability, including the environmental, economic, and social goals of sustainable development. In this regard, Kramers et al. [38] suggested that the SSC concept should be understood 'as a way of emphasizing initiatives when smartness is (also) used to promote sustainability'. In addition, Höjer and Wangel [9] claimed that the SuC becomes smart when it is supported by ICT.
Materials and Methods
The structure of knowledge and the development of research in the fields of smart and sustainable cities were assessed using bibliometric analysis. The analysis covered publications in the SSC field, publications describing the SC concept and the SuC notion separately, as well as publications discussing concepts that some authors connect with SC or SuC, such as the digital city or the green city. The choice of such a scope of the analysis resulted from the fact that the SC and the SuC concepts had a significant impact on the development of the SSC concept, and, therefore, they had to be included in the in-depth analysis of the scientific landscape of the SSC.
The analysis presented in this article was carried out in the following phases: Phase 1: Study design; Phase 2: Data collection; Phase 3: Initial data analysis; Phase 4: Descriptive bibliometric analysis; Phase 5: Network analysis; Phase 6: Conclusions and directions for future research.
In the first phase of the study, online bibliographic databases were reviewed for the possibility of using them for the research presented in this article. Two bibliographic sources were selected: the Web of Science (WoS) Core Collection and the Scopus database. These databases provide reliable and relevant information on scientific work and are most often selected for bibliometric analyses. All types of scientific papers collected in the WoS and Scopus databases were chosen for the analysis. Then, the bibliometric analysis software tools were reviewed. As the intention was to combine information collected in the WoS and Scopus databases, VOSviewer and Biblioshiny were selected as software tools for the bibliometric analysis. In the next step, a list of the search query words and the Boolean operators were identified. Aiming to conduct an in-depth analysis, the following query words were selected: • in the extensive scientific field of the SC concept: "smart city/cities", "intelligent city/cities", "digital city/cities", "virtual city/cities" "cyber city/cities", "networked city/cities", "knowledge city/cities", "wisdom city/cities", "ubiquitous city/cities", "real-time city/cities", "hybrid city/cities"; • in the extensive scientific field of the SuC concept: "sustainable city/cities", "sustainable urban", "resilient city/cities", "eco-city/cities", "ecocity/ecocities", "city sustainability", "green city/cities", "urban sustainability", "sustainable urbanization", "sustainable city development", "sustainable urban development"; • in the extensive scientific field of the SSC concept: "smart sustainable city/cities", "sustainable smart city/cities", "smart and sustainable city/cities", "sustainable and smart city/cities", "sustainable and smart urban", "smart and sustainable urban", "sustainable development of smart city/cities", "sustainable development", "sustainability".
The words 'AND' and 'OR' were used as the Boolean operators, resulting in a total of 36 combinations of specific query wording. The lists of query wording instances with relevant Boolean operators are presented in Tables 1-3. In the last step of this phase, the time span of the analysis was determined. Initially, no starting publication date was entered in the search. However, due to the fact that the first publication in the analyzed field (regarding the SuC concept) appeared in 1983, the timespan was narrowed down to 1983-2020.
In the second phase of the study, the analysis input data was collected. The final item of the data was retrieved on 30 October 2019 from the WoS and Scopus databases. Various wording combinations in the publication subject (TS) and the publication title (TI) for the WoS database and in the publication TITLE-ABS-KEY and in the publication TITLE for the Scopus database were used to identify scientific publications in the extensive field of the SC, SuC, and SSC concepts. The obtained results are presented in Tables 1-3. The third phase of the study consisted of a preliminary review of the collected data. This was limited to the records directly related to the SC and the SuC. In the next step, the publications included in the identified extensive scientific field of the SSC were screened (i.e., titles and abstracts) to indicate the items directly related to the SSC concept. In addition, the papers defining the SC, SuC, and SSC concepts were selected to show the development of individual concepts and their impact on the development of the SSC concept (this is presented in Section 2).
The database created in this way became the basis for descriptive bibliometric analysis (the fourth phase of the study). The analysis was carried out separately for scientific papers, describing the concepts of the SC, the SuC, and the SSC. For each of these concepts, the number of publications and the total number of citations, the main research areas and research categories, the most productive authors, sources, countries, and organizations publishing scientific works were determined. The decision to conduct a separate descriptive analysis for each concept resulted from the fact that the concepts appeared in different years, which affects the number of published articles or the number of citations and makes it difficult to perform a joint analysis of all data.
In the fifth phase of the study, a network analysis was performed. To achieve the purpose of the article, a decision was made to perform the network analysis for the SSC concept data collected from the WoS and Scopus databases. This analysis was conducted using the VOSviewer (version 1.6.11) (Centre for Science and Technology Studies, Leiden University, The Netherlands, 2019) and Biblioshiny (based on R version 3.6.1, Bibliometrix package version 2.2.1)(University of Naples Federico II, Naples, Italy, 2016) software tools. VOSviewer is powerful at showing networks of structures composed of many elements based on a distance-based visualization approach. The created networks consist of nodes and lines. Depending on the subject of the analysis, the nodes are represented, e.g., by keywords, sources, countries, or authors. The nodes are grouped into clusters (i.e., each node is assigned to exactly one cluster). The lines represent the relationships between them, e.g., co-occurrence, co-citation. The number of obtained clusters can be determined (limited) by a clustering parameter (e.g., cluster minimum size). However, to achieve the highest homogeneity of individual clusters, this was not the case in the performed analyses. The clustering technique used by the VOSviewer is discussed by Waltman, Van Eck, and Noyons [39]. The technique requires a relevant algorithm for solving an optimization problem. For this purpose, VOSviewer uses the smart local moving algorithm introduced by Waltman and Van Eck [40]. Biblioshiny is a proper tool for presenting general statistics and relations between the most important scientific collaboration units, using the Three Fields Plot [41]. In the first step of this phase, the data obtained from the two databases were converted into an appropriate format, which could be processed using the selected software tool. This made it possible to create a common database. Then, the data was cleaned by removing duplicates and misspelled elements. This enabled the determination of the final number of publications covered by the network analysis (539 items). The network analysis was divided into three parts: (1) Word, keyword, and co-keyword analyses; (2) Scientific collaboration mapping; (3) Top authors, sources, and keywords relations. Word, keyword, and co-keyword analyses are methods of describing and visualizing the structure of scientific fields of a particular group of publications [42]. The level of scientific collaboration is measured, among others, by citation and co-citation analyses [41]. Therefore, scientific collaboration was mapped and analyzed using the results of the co-authorship network of countries and the citation network of authors, countries, and sources. Top authors, sources, and keywords relations were developed using the Three Fields Plot of the Biblioshiny package. The tool makes it possible to visualize the main items of three selected fields (e.g., authors, authors' keywords, sources) and show how they are related using a Sankey diagram. As for the network analysis, it was assumed that the networks would contain a maximum range of information. Therefore, the boundary conditions were always set to the minimum (1 occurrence/citation). Only in the network of keywords' co-occurrence, was the threshold set to 3 (occurrences) to make the network legible and draw attention to the most significant elements, relations, and structures thereof.
The final phase of the study included the presentation of conclusions on academic trends and directions for future research in the SSC field.
All the phases of the study presented in this article are shown in Figure 1.
The most cited article on the SC in the WoS database, with 1425 citations, was 'Internet of things for smart cities' written by Zanella et al. [45]. In the second place was 'Smart cities in Europe' written by Caragliu et al. [5], cited 722 times. The study by Botta et al. [46] 'Integration of cloud computing and internet of things: A survey' appeared in third place with 509 citations. The most cited publication on the SC in the Scopus database, with 2264 citations, was 'Fog computing and its role in the internet of things' [47]. Interestingly enough, this paper was not indexed in the WoS database. In the second place in the Scopus database was 'Internet of things for smart cities' [45], with 1917 citations. 'Smart cities in Europe' [5] took third place with 913 citations. Tables A1 and A2 in Appendix A show the most cited publications on the SC in the WoS and Scopus databases. It should be noted that most studies on the lists with the highest number of citations were from the period of 2011-2016.
In the next step, the main research areas and categories of studies on the SC were identified, together with sources and authors with the most substantive contributions to the literature. Moreover, the most productive countries and organizations in the field were indicated. The analysis results showed that: The most productive authors, sources, countries, and organizations in the SC literature field are presented in Tables A3 and A4 in Appendix A.
The most cited article on the SuC, indexed in the WoS and Scopus databases, was 'The role of urban parks for the sustainable city' by Chiesura [50] (with 830 citations in the WoS and 945 citations in the Scopus database). The study by Kennedy et al. [51] titled 'The changing metabolism of cities' was in second place in the WoS (cited 522 times) and in the third place in the Scopus database (577 citations). The second-place Scopus-indexed item was 'Urbanization in developing countries: Current trends, future projections, and key challenges for sustainability' by Cohen [52] (636 citations). Interestingly enough, this paper was not indexed in the WoS database. The article 'Rethinking sustainable cities: Multilevel governance and the 'urban' politics of climate change' by Bulkeley and Betsill [53] The analysis of the main research areas and categories of studies on the SuC, including the most productive sources, authors, countries, and organizations in the field, indicated that: The most prolific organizations with regard to the SuC publications indexed in the WoS database were the University of London (39 records), the Royal Institute of Technology (31), and the University of California (28). According to the Scopus database, the most active were the University of Melbourne (33), the Royal Institute of Technology (32), and University College London (32).
The most productive authors, sources, countries, and organizations in the SuC literature field are presented in Tables A7 and A8 in Appendix A.
Descriptive Bibliometric Analysis of the SSC Publications
The analysis indicated 1117 and 1804 publications listed in the WoS and the Scopus database, respectively, covering the extensive scientific field of the SSC (Table 3). The items (i.e., titles and abstracts) were screened to determine publications directly related to the SSC concept. The process resulted in 355 publications (including 195 articles, 116 proceedings papers, and 29 book chapters) indexed in the WoS database and 489 Scopus-indexed items (including 239 articles, 167 proceedings papers, and 49 book chapters). Table 6 presents indicators of publication activity related to the SSC literature based on items indexed in the WoS and Scopus databases. It contains the number of publications (i.e., annual and cumulative) for the period of 2005-2020 and the number of citations that these items have received.
The first-ever publication on the SSC, indexed in the WoS and Scopus databases, was a book chapter titled 'Smart and sustainable city-a case study from Hong Kong' by Lau et al. [54]. It was published in June 2005, which was correctly recorded in the WoS database. According to the Scopus database, the publication was from 2008.
The most cited article on the SSC recorded in the WoS and the Scopus database (with 163 and 217 citations, respectively) was the paper by Lee et al. [55] titled 'Towards an effective framework for building smart cities: Lessons from Seoul and San Francisco'. The study 'Smart and digital city: A systematic literature review' by Cocchia [14] appeared in second place in the WoS database with 160 citations. Interestingly enough, this publication was not indexed in the Scopus database. The article 'Sustainable-smart-resilient-low carbon-eco-knowledge cities: Making sense of a multitude of concepts promoting sustainable urbanization' by De Jong et al. [56] ranked third in the WoS database (158 citations) and second in the Scopus database (205 citations). In the third place in the Scopus database was the study 'Programming environments: Environmentality and citizen sensing in the smart city' by Gabrys [57] (180 citations). Tables A9 and A10 in Appendix A show the most cited publications on the SSC in the WoS and Scopus databases. It should be added that most studies on the lists with the highest number of citations were from the period of 2014-2017. Table 6. The number of publications and total citations related to the SSC literature.
•
The most active countries in the SSC literature field, according to the WoS database, were Italy (39 indexed records), Spain (37), the United Kingdom (36), and China (36). According to the Scopus database, the most productive countries in this field were Italy (52), the United Kingdom (51), and the United States (47).
•
The most prolific organizations in the SSC academic field included in the WoS database were the Royal Institute of Technology (18 items), the Norwegian University of Science and Technology (11), and Queensland University of Technology (7). According to the Scopus database, the most productive entities were the same, but in a different order: the Norwegian University of Science and Technology (28), the Royal Institute of Technology (15), Queensland University of Technology (10).
The most productive authors, sources, countries, and organizations in the SSC literature field are presented in Tables A11 and A12 in Appendix A.
Word, Keyword, and Co-Keyword Analyses
This subsection presents the results of the most frequently used words and keywords, the author keyword occurrence dynamic, and the co-occurrence network of author keywords in the SSC literature indexed in the WoS and Scopus databases.
The analysis of the most frequent words and keywords was based on the number of relevant wording occurrences in publications in the following types of categories: author keywords, words in titles, and words in abstracts. The analysis results showed that 1666 author keywords were used in the SSC literature. However, as shown in Table 7, there were only two author keywords that occurred more than a hundred times. Although the key topics of the papers under consideration were issues related to the SSC, due to the wide diversification of the concept forms used by authors, total occurrences of author keywords were much fewer than the number of publications. The authors used the terms, such as 'smart city', 'smart cities', 'sustainable city', 'sustainable cities'. The main term 'smart and sustainable cities' appeared in keywords as 'smart sustainable city/cities', 'sustainable smart city/cities', 'sustainable and smart city/cities'. It was decided to show them in the original version separately as used by respective authors. The analysis revealed that the most frequent keywords were as follows: 'smart city', 'smart cities', 'sustainability', 'sustainable city', 'sustainable development', and 'smart sustainable cities'. In the next step, the dynamic of the time-dependent occurrence of author keywords was investigated. It could be seen in Figure 2 that the number of all main-term occurrences per year increased over time, but some of them grew more dynamically than others. The terms with the highest increase in occurrences over time were: 'smart city', 'smart cities', 'sustainability', 'sustainable cities', and 'sustainable development'.
To visualize the research hotspots in the SSC scientific field, the co-occurrence of author keywords was analyzed. The keyword co-occurrence threshold was set to three, and 122 keywords out of 1666 were classified as visualization items. The keyword co-occurrence network is presented in Figure 3. The size of the circles corresponded to the number of occurrences of the represented keywords. The larger the circle, the more the author keyword had been co-selected in the SSC literature. The topic similarity and its relative strength were demonstrated by the distance between the elements of individual pairs. Different colors of the circles were assigned to individual keyword clusters. The network in Figure 3 illustrates 16 distinct clusters, representing individual subfields of the research areas in the SSC literature. Table A13 in Appendix A presents the main parameters of the top 10 occurrences of author keywords in the SSC literature. The links between particular keywords indicated the number of papers in which the keywords co-occurred. Analyzing Figure 3 and Table A13 in Appendix A, it could be concluded that:
•
The core topics in the SSC literature with the highest total link strength were: 'smart city/cities', 'sustainable city/cities', 'sustainable development', 'ICT', 'big data analytics', and 'urban sustainability'.
The obtained results demonstrated a considerable diversity of the co-occurrence of author keywords in individual publications in the SSC literature. This proved the multi-faceted and multi-dimensional character of this particular scientific field.
Mapping Scientific Collaboration
Regional collaboration and research hotspots could be provided with the visualization of co-authorship occurrences of countries. The network illustrating such co-occurrences between countries in the SSC literature is presented in Figure 4. The size of the circles represented the number of occurrences of documents. The collaboration strength was demonstrated by the distance between circles in individual pairs. Different colors of circles were assigned to individual collaboration camps. In addition, Table A14 in Appendix A lists the main parameters of the top 10 occurrence countries. The total link strength of a country indicated the number of papers in which the authors of a given publication represent the two countries involved.
Analyzing the created network, it could be seen that: • there were 66 countries represented by authors collaborating in the SSC literature; • the United Kingdom and the United States were the countries with the greatest total link strength when it comes to an international collaboration of authors; • Italy, Spain, India, and Norway had a high number of publications, but their international collaboration strength was relatively low.
The next step was the country-by-country citation analysis. Figure 5 shows the citation network of countries, illustrating mutual international citations in the SSC literature. The size of the relevant figures depended on the number of publications in a given country. The distance between two circles representing two different countries reflected the number of mutual citations of authors coming from them. The more often authors from one country cited authors from another, the shorter was the distance between two circles representing two different countries in the presented network. The links number (i.e., the number of other countries linked to the analyzed country by citations) and the total link strength (i.e., the number of all citations of authors coming from the country) for the top 10 countries are listed in Table A15 in Appendix A. Analyzing the created network, it could be seen that: • there were 50 countries whose publications in the SSC literature had been cited by foreign authors; • Italy, the United States, and the Netherlands were characterized by the closest academic relationship; • there were several scientific communities that had published in the SSC literature, and the biggest two were the following groups of countries: (1) Italy, Australia, the United States, and the Netherlands and (2) Norway, Sweden, the United Kingdom, and Finland; • despite the smaller number of publications, the international significance of countries, such as Sweden, Australia, Brazil, Finland, South Korea, Greece, Taiwan, and Canada, was relatively high (i.e., high total link strength relative to the number of publications).
Collaboration and significance of individual sources in the creation of the SSC knowledge were covered by the source-citation analysis. The source was understood as a journal, a book, or proceedings indexed in the WoS and Scopus databases. Figure 6 shows the network of mutual citations of different sources of the SSC literature. The size of the circles in the figure depended on the number of publications of a given source. The distance between two circles representing two different sources reflected the number of their mutual citations. The more often the publications of a given source cited the other source publications, the shorter was the distance between the two sources in the network presented in the figure. The links number (i.e., the number of other sources related to the analyzed source through citations) and the total link strength (i.e., the number of the source total citations) for the top 10 sources are listed in Table A16. Analyzing the created network, it could be seen that: • there were 99 sources whose publications in the SSC literature had been cited by another source and had created the network; • there were two groups of significant sources with strong mutual relations: (1)
Top Authors, Sources, and Keywords Relations
The relations between the main authors' keywords, authors, and sources were visualized using the Three Fields Plot. In this case, the relevant elements were represented in the diagram by rectangles with a different color. The height of the rectangles depended on the value of the sum of the relations arising between the element that the rectangle represents (one of the elements in the authors' keyword, author, and source diagram) and the diagram of other elements. The more relations the element had, the higher was the rectangle representing it. Figure 7 presents the diagram for research in the SSC literature, focusing on relations between the main authors' keywords, authors, and sources. The analysis demonstrated in which sources the authors of SSC publications had published most frequently and which research topics of the SSC concept they had explored. The research topics were understood here as the authors' keywords. The analysis of the top authors, sources, and keywords indicated that there were three authors (i.e., Simon Elias Bibri, John Krogstie, and Tan Yigitcanlar) and five sources (i.e., Sustainable Cities and Society, Cities, Sustainability, the Journal of Cleaner Production, and the Urban Book Series) having strong relations with the SSC literature main research topics ('smart sustainable cities', 'smart cities', and 'smart city', in particular). In addition, Figure 8 shows the relations between the main authors' keywords, sources, and cited sources. This diagram demonstrated the structure of top sources of publications (as the source and cited source) in developing the knowledge in the main topics in the SSC literature. The topics were represented by the authors' keywords, and the source or cited source significance was represented by the number of relevant relations. The analysis showed that the key sources of publications based on their contribution to the exploration of the SSC literature's main topics were Sustainable Cities and Society, the Journal of Cleaner Production, Sustainability, Cities, and the Urban Book Series.
Conclusions and Directions for Future Research
Considering that, in the coming years, the number of city inhabitants worldwide would continue to grow, it is extremely important to find solutions that would allow cities to meet the challenges of urbanization. One of them is the SSC concept, but its implementation requires changes at all levels of the city [11]. Therefore, scholars and policymakers responsible for urban development should have extensive theoretical knowledge in this field. They should also familiarize themselves with the achievements and failures of cities that have already implemented the SSC concept to benefit from their positive experience and avoid repeating their mistakes.
For this reason, this article presented the scientific landscape of the SSC based on a bibliometric analysis of 1983-2020 publications recorded in the WoS and Scopus databases. Apart from SSC publications, the analysis also covered publications in the field of the SC and the SuC because the SC and SuC concepts influenced the development of the SSC notion significantly. Because the bibliometric analysis enabled a systematic and transparent review process based on the statistical measurement of science, scholars, or scientific activities, it was possible to identify how the SSC concept and the research trends have changed over the years and what research directions in the SSC field should be considered in the future.
The literature review indicated that the SSC notion is based on a blend of two complex concepts, the SC and the SuC. In general, it could be stated that the SC focuses on the interconnectedness and advancement of the physical, IT, social and business infrastructure to support the city's collective intelligence, while the SuC integrates environmental, economic, and social considerations to meet the goals of sustainable development. The SSC integrates the two concepts of urban development, but according to different authors, these are approached differently, depending on the starting point for creating the SSC. For some, the SSC is a sustainable city that should become smart because thanks to intelligent solutions, it would be able to meet the challenges of urbanization and achieve the goals of sustainable development [38,58,59]. Others suggest that the starting point for the SSC creation is smart city, which should not focus on technology and smart solutions only but should also take account of the impact of these solutions on the development of the social, economic, and environmental capital [7,12]. As a result, the scope of the research is very wide, and the directions of the SSC concept development are varied and cover diverse scientific areas.
The bibliometric analysis results proved that the number of publications on the SC, SuC, and SSC concepts increased year by year. However, there were substantially fewer publications on the SSC than on the other two concepts. Undoubtedly, this was due to the fact that the SSC concept is the newest, and the authors of the highest number of publications in the field started their publishing activity only in 2014. This might point to a great current interest in this research area on the one hand, and to its enormous potential on the other.
The descriptive bibliometric analysis indicated that studies on the SSC covered diverse research areas, such as Social Sciences, Computer Science, and Engineering, in particular. The most productive sources of publications on the SSC included Sustainable Cities and Society, Sustainability, the ACM International Conference Proceeding Series, and the Journal of Cleaner Production. Indisputably, the most prolific author in this area was Simon Elias Bibri. It must be noted that the publishing activity of the above-mentioned researcher had increased significantly in recent years. However, the SSC scientific landscape was mainly shaped by authors from the countries, such as Italy, the United Kingdom, the United States, Spain, and China. In addition, considering the affiliations of scientists focusing on the SSC concept, a few organizations should be distinguished. This especially concerned the Royal Institute of Technology, the Norwegian University of Science and Technology, and the Queensland University of Technology.
The network analysis resulted in more interesting findings. The co-occurrence network of authors' keywords revealed that research in the SSC area primarily concerned the following topics: 'smart city/cities', 'sustainability', 'sustainable city/cities', 'sustainable development', and 'smart sustainable cities'. However, the clusters obtained from the network analysis of authors' keywords demonstrated a considerable diversity of the co-occurrence of various topics (i.e., author keywords) in individual publications in the SSC literature. For example, the 'smart sustainable city' keyword co-occurred, in particular, with such other keywords as 'smart city', 'sustainability', 'innovation', 'smart mobility', and 'renewable energy'. The 'smart sustainable cities' keyword co-occurred with 'urban sustainability', 'big data analytics', 'context-aware computing', 'cloud computing', and 'environmental sustainability'. In contrast, 'sustainable smart city' co-occurred with 'sustainable urban development', 'eco-city', 'climate changes', 'urbanization', and 'green city'. This proved the multi-faceted and multi-dimensional character of this particular scientific field. Moreover, it seemed to be an interesting topic for further investigation based on a thorough review of the SSC literature.
The results of the countries' cooperation network analysis proved that there were two countries (i.e., the United Kingdom and the United States), which are definite leaders in the field of co-authorship. However, the citation network of countries showed that other countries (i.e., Norway, Italy, and Sweden) have had the closest relations in this kind of collaboration. The citation network of sources analysis revealed that there were two groups of significant sources cooperating in the SSC literature: (1) Sustainability, the Journal of Cleaner Production, the IOP Conference Series, and Lecture Notes in Computer Science and (2) Sustainable Cities and Society, the Urban Book Series, and the Journal of Big Data.
The bibliometric analysis identified the most frequent topics (i.e., author keywords) and also the most productive authors, sources, countries, and organizations in the SSC literature. Indicating the sources with the highest productivity and citations, these results could be used by future authors of SSC publications to adopt an appropriate publication strategy. Moreover, the analysis provided valuable information about the most active countries and academic organizations, as well as the most influential authors in the field of the SSC, which could become the basis for establishing future collaboration. The results of the analysis could also be useful for practitioners and decision-makers in cities because they could indicate publications, which are the most influential in terms of the SSC concept development, as evidenced by the number of citations that these publications receive.
In addition, the analysis results proved that, so far, most publications indexed in the WoS and Scopus databases have focused on showing the SSC concept theoretical foundations. There were also publications describing practical aspects of the SSC concept implementation, but they usually related to the implementation of smart solutions in cities (e.g., [58][59][60]). The publications presenting the experience of cities in the implementation of the SSC concept were few (e.g., [61][62][63]), and an increase in its number could only be observed in the last two years. It should be noted, however, that most of them were based on the experiences of cities in developed countries. The number of publications concerning the SSC concept implementation in peripheral countries is very limited (e.g., [36,64]). Therefore, it seems important in the future to give more attention to research showing good experiences in the implementation of smart solutions that make it possible for cities to develop in a manner that could also be called sustainable.
The analysis of the SSC literature indicated some important research gaps and issues with a potential future contribution. In general, there are growing expectations that smart cities will drive sustainable development. For example, Martin et al. [61] identified the key tensions between smart city visions and the goals of sustainable urban development. Höjer and Wangel specified the main challenges for the SSC concept of practical use [9]. In addition, based on an extensive interdisciplinary literature review, Bibri and Krogstie [10] delivered a comprehensive overview of the existing gaps in research on the SSC of the future. It should be noted that there are promising recent research areas concerning a holistic approach to the SSC concept [65][66][67].
Although every effort has been made to perform the bibliometric analysis in the best and most accurate manner, the research has some limitations. The analysis presented in the article was based on WoS-and Scopus-indexed publications only. For this reason, it could not be assumed as fully complete as there might be other important SSC publications not included in the two databases. However, the WoS and Scopus databases alone ensure the highest quality standards. In addition, it must be noted that the number of publications exploring the SSC concept is expected to increase dynamically in the future. Therefore, the obtained results should be treated with caution because they might become obsolete rather fast. Nevertheless, the aim of this study was comprehensive quantitative up-to-date evaluation of the SSC scientific landscape to identify the most prolific authors, sources, institutions, and countries, to indicate the most influential research channels and impact from authors, sources, countries, and research topics, as well as to provide valuable information for further investigation and determine publication strategies. Table A14. Main parameters of the top 10 countries (ranked by the total link strength) in the co-authorship network of countries in the SSC literature.
|
2020-01-23T09:07:13.549Z
|
2020-01-21T00:00:00.000
|
{
"year": 2020,
"sha1": "266e5abcd9ada8eb0a1a02aff9db43aac82c1b8a",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2071-1050/12/3/779/pdf",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "2abf87ca2f19b9a2b257e0083b0d0447a406caf9",
"s2fieldsofstudy": [
"Computer Science"
],
"extfieldsofstudy": [
"Engineering"
]
}
|
215679723
|
pes2o/s2orc
|
v3-fos-license
|
Itraconazole Inhibits Enterovirus Replication by Targeting the Oxysterol-Binding Protein
Itraconazole (ITZ) is a well-known antifungal agent that also has anticancer activity. In this study, we identify ITZ as a broad-spectrum inhibitor of enteroviruses (e.g., poliovirus, coxsackievirus, enterovirus-71, rhinovirus). We demonstrate that ITZ inhibits viral RNA replication by targeting oxysterol-binding protein (OSBP) and OSBP-related protein 4 (ORP4). Consistently, OSW-1, a specific OSBP/ORP4 antagonist, also inhibits enterovirus replication. Knockdown of OSBP inhibits virus replication, whereas overexpression of OSBP or ORP4 counteracts the antiviral effects of ITZ and OSW-1. ITZ binds OSBP and inhibits its function, i.e., shuttling of cholesterol and phosphatidylinositol-4-phosphate between membranes, thereby likely perturbing the virus-induced membrane alterations essential for viral replication organelle formation. ITZ also inhibits hepatitis C virus replication, which also relies on OSBP. Together, these data implicate OSBP/ORP4 as molecular targets of ITZ and point to an essential role of OSBP/ORP4-mediated lipid exchange in virus replication that can be targeted by antiviral drugs.
Itraconazole acted on a novel target, the oxysterol-binding protein (OSBP), a protein that has an essential role in lipid transfer.
INTRODUCTION
The family Picornaviridae contains many important human and animal pathogens. The genus Enterovirus includes poliovirus (PV), coxsackievirus (CV), echovirus, several numbered enteroviruses (e.g., enterovirus-71 [EV71]), and human rhinovirus (HRV). Except for PV, no vaccines are available to prevent infections with enteroviruses and no antiviral drugs are available for treatment. Other important human picornaviruses include hepatitis A virus and human parechovirus (HPeV). Well-known animal pathogens are foot-and-mouth disease virus and encephalomyocarditis virus (EMCV).
The genome of enteroviruses consists of a 7.5 kb singlestranded RNA molecule of positive polarity [(+)RNA]. It encodes a single polyprotein that is proteolytically processed by the viral proteases into the structural proteins (VP1-VP4) and the nonstructural proteins (2A-2C and 3A-3D). The viral genome is replicated by assemblies of viral and host proteins located on intracellular membranes termed replication organelles (ROs). The ROs are formed as a result of virus-induced remodeling of secretory pathway membranes, which most likely starts at the Golgi complex (Hsu et al., 2010), eventually resulting in a complex network of tubulovesicular membranes (Belov et al., 2012;Limpens et al., 2011). Viral modification of lipid homeostasis is thought to play a major role in RO formation. Viral proteins 2BC and 3A play a major role in the membrane rearrangements by recruiting essential host factor for enterovirus replication to ROs, such as phosphatidylinositol-phosphate-4-kinase III beta (PI4KIIIb), a Golgi-localized lipid kinase that generates phosphatidylinositol-4 phosphate (PI4P) (Arita, 2014;Hsu et al., 2010). The functional importance of elevated PI4P levels at ROs remains to be established. The viral RNA-dependent RNA-polymerase, 3D pol , binds PI4P in vitro, but it is unknown whether this is important for its recruitment and/or activation in infected cells (Hsu et al., 2010). Alternatively, the PI4P lipids may participate in RO formation by facilitating the recruitment of PI4P-binding host proteins with membrane-modifying properties.
Cholesterol is a critical membrane component that determines membrane fluidity and regulates the formation and function of membrane-bound complexes of lipids and proteins. Several (+) RNA viruses, such as hepatitis C virus (HCV), dengue, and West Nile virus, remodel the cellular cholesterol landscape to make intracellular host-cell membranes conducive for efficient genome replication (Rothwell et al., 2009;Wang et al., 2014). Enterovirus-induced rearrangements of secretory pathway membranes into the tubulovesicular RO structures may also depend on alterations in cholesterol homeostasis. Recent data suggest that enteroviruses stimulate clathrin-mediated endocytosis to transport cholesterol from the plasma membrane and extracellular medium to ROs (Ilnytska et al., 2013). However, other intracellular cholesterol trafficking pathways may also be subverted by enteroviruses to create their ROs.
Recently, oxysterol-binding protein (OSBP) was shown to play a key role in the transport of cholesterol and PI4P between the endoplasmic reticulum (ER) and Golgi (Mesmin et al., 2013). OSBP links ER and trans-Golgi membranes at so-called ER-Golgi membrane contact sites (MCSs) and shuttles sterol into the Golgi and PI4P back to the ER, where it is hydrolyzed by the PI4P-phosphatase Sac1. This cholesterol/PI4P exchange cycle drives the delivery of sterol in the trans-Golgi and self-regulates the localization of OSBP on the Golgi. OSBP and the OSBP-related proteins (ORPs) constitute a family of related proteins that, based on gene structure and sequence, can be subdivided into six subfamilies. OSBP and its closest relative, ORP4 (also called OSBP2), belong to subfamily I. All ORPs have a lipid-binding domain that was initially thought to be specific for sterols. However, recent structural analysis suggests that ORPs have the ability to bind PI4P and a second lipid that is either a sterol or a nonsterol ligand. Many ORPs, including OSBP, have an FFAT-motif that is recognized by ER-resident VAP receptors and an N-terminal pleckstrin homology (PH) domain for binding PI4P, through which they are linked to a variety of organelles. Although the functions of most ORPs are not very well understood, it has become clear that ORPs execute diverse functions in lipid sensing, lipid transport, and cell signaling (Raychaudhuri and Prinz, 2010;Weber-Boyvat et al., 2013).
We set out to identify novel inhibitors of enterovirus replication by screening the NIH Clinical Collection (NCC), a library of US Food and Drug Administration (FDA)-approved drugs that have a history of use in clinical trials for treatment of a wide variety of diseases. Similar collections of FDA-approved drugs have proven to be rich sources of undiscovered bioactivity and therapeutic potential. We identified itraconazole (ITZ) as a broad-spectrum inhibitor of enterovirus replication. ITZ is a well-known antifungal drug that inhibits CYP51, a cytochrome P450 required for sterol biosynthesis (Lestner and Hope, 2013). In addition, ITZ exerts anticancer activity by inhibiting angiogenesis-through disturbing mTOR signaling and vascular endothelial growth factor receptor 2 (VEGFR2) trafficking-and the Hedgehog (Hh) signaling pathway (Kim et al., 2010;Nacev et al., 2011;Xu et al., 2010). ITZ has been found to be efficacious in patients with several cancer types in multiple phase 2 clinical studies (Antonarakis et al., 2013;Kim et al., 2014;Rudin et al., 2013). We demonstrate that known targets of ITZ cannot explain the antiviral activity of ITZ. Instead, evidence is presented that OSBP and ORP4 are novel targets of ITZ and that direct binding of ITZ to OSBP, which localizes at ROs, disrupts its lipid-shuttling function, and accounts for the antiviral effect of ITZ.
ITZ Is an Inhibitor of Enterovirus and Cardiovirus
Replication We performed a screen of the NCC to identify novel inhibitors of CVB3 replication. Like all enteroviruses, CVB3 kills its host cell and thereby causes a ''cytopathic effect'' (CPE). We screened the NCC by microscopically observing which compounds prevented the development of CPE in a multicycle replication assay and identified ITZ ( Figure S1) as one of the hits. To determine its spectrum of antiviral activity, we tested ITZ against a representative panel of picornaviruses in a multicycle CPE-reduction assay. ITZ exhibited antiviral effect against all enteroviruses examined (belonging to several species) with 50% effective concentration (EC 50 ) values between 0.3 mM and 1.6 mM (Table S1). In addition, EMCV, a Cardiovirus genus member, was inhibited by ITZ. In contrast, equine rhinitis A virus (ERAV; Aphthovirus genus member) and HPeV-1 (Parechovirus genus member) were insensitive to ITZ. To exclude the possibility that the antiviral activity was due to toxic side effects, we determined the effect of ITZ on virus production during a single replication cycle. Similar to the multicycle CPE-reduction assay, ITZ was active against CVB3, EV71, HRV14, and EMCV, but not ERAV, in a single replication cycle ( Figure 1A) without apparent toxicity ( Figure 1B). ITZ also inhibited Saffold virus (SAFV) replication, a human cardiovirus ( Figure 1A). Thus, ITZ exerts broad antiviral activity against enteroviruses and cardioviruses.
ITZ Inhibits Viral RNA Genome Replication Next, we determined the effect of ITZ on translation and replication of transfected CVB3 and EMCV RNAs, namely a subgenomic replicon of CVB3, in which (part of) the capsid-coding region is replaced by a firefly luciferase gene, or a genomic RNA of EMCV, in which a Renilla luciferase gene is inserted upstream of the coding region. As positive controls, we used guanidine-HCl and dipyridamole, well-known and potent inhibitors of CVB3 and EMCV replication, respectively. Two hours after transfection of the RNAs, when no RNA replication has taken place yet (van Kuppeveld et al., 1995), luciferase levels were unaffected, indicating that ITZ does not inhibit viral genome translation (Figure 1C). However, at later time points, luciferase production by both replicons was decreased, demonstrating that ITZ affects RNA replication. Importantly, ITZ did not affect viral polyprotein synthesis and processing ( Figure S2).
Inhibition of Virus Replication Is Independent of Known
Targets of ITZ ITZ is widely used as an antifungal drug that inhibits the fungal enzyme CYP51. ITZ has also been shown to have some inhibitory activity toward the human CYP51 (hCYP51) and the related cytochrome P450 CYP3A4. In addition to ITZ, other azole family antifungal drugs, including posaconazole, ketoconazole, fluconazole, and voriconazole ( Figure S1), also inhibit hCYP51 and CYP3A4 with slightly lower or similar potency as ITZ (Warrilow et al., 2013;Zhang et al., 2012). We tested whether these drugs exert antiviral activity using recombinant viruses RLuc-CVB3 and RLuc-EMCV, which carry the Renilla luciferase gene upstream of the coding region. At 10 mM, only posaconazole inhibited replication of RLuc-CVB3 and RLuc-EMCV. The remaining azoles did not display any antiviral activity at concentrations up to 100 mM . Similar results were obtained in a multicycle CPE-reduction assay (not shown). These results ruled out the possibility that inhibition of hCYP51 or CYP3A4 underlies the antiviral activity of ITZ and its structurally related analog posaconazole.
As ITZ also inhibits the Hedgehog (Hh) signaling pathway, most likely by interfering with the function of the G proteincoupled receptor-like protein Smoothened (Kim et al., 2010), we tested several Smoothened antagonists in the viral luciferase assays. The Smoothened antagonists KAAD-cyclopamine, Sant-1, and Sant-2 (Chen et al., 2002;Taipale et al., 2000) had no effect on the replication of RLuc-CVB3 or RLuc-EMCV ( Figures 2B and S3D), indicating that the antiviral activity of ITZ is not mediated by its inhibition of the Hh pathway.
The antiangiogenic activity of ITZ has been attributed at least in part to its inhibition of the mTOR signaling pathway through disruption of the shuttling of cholesterol between plasma membrane and late endosomes/lysosomes, thereby inducing the accumulation of cholesterol in the endolysosomal system (Xu et al., 2010). We found that cholesterol, stained with filipin, was redistributed not only by ITZ and posaconazole but also by ketoconazole (which does not inhibit virus replication) in two human cell lines (HAP1 [ Figure 2C] and HeLa R19 cells [ Figure S3E]). Moreover, the mTOR inhibitor rapamycin had no effect on picornavirus replication (Beretta et al., 1996;Wong et al., 2008). Together, these results suggest that inhibition of virus replication by ITZ or posaconazole is not due to disruption of endosomal cholesterol shuttling or the cholesterol-related mTOR inhibition.
In addition to the aforementioned molecular and pathway targets of ITZ, ITZ has been reported to disturb N-glycosylation (Nacev et al., 2011). However, the N-glycosylation inhibitor tunicamycin did not affect poliovirus (Doedens et al., 1997) or CVB3 replication (our data not shown). ITZ has also been shown to antagonize the estrogen receptor a (ERa) (Cheng et al., 2012). But ERa agonist b-estradiol did not affect CVB3 ( Figure 2D) or EMCV replication ( Figure S3F). Finally, ITZ has been reported to target p-glycoprotein, UDP-glucuronosyltransferase, and ERb, none of which are likely to mediate the antiviral activity of ITZ, because these are as potently inhibited by ketoconazole (Cheng et al., 2012;Walsky et al., 2012;Wang et al., 2002b), which did not affect virus replication.
Mutations in 3A that Confer Resistance to PI4KIIIb
Inhibitors Also Confer Resistance to ITZ, but ITZ Does Not Inhibit PI4KIIIb Activity As a first step to identifying the antiviral target of ITZ, we studied its effect on replication of CVB3 mutant viruses that we previously selected for resistance against other inhibitors. CVB3 carrying mutation 3A[H57Y] -which confers resistance to PI4KIIIb inhibitors (e.g., PIK93, enviroxime, GW5074) (van der Schaar et al., 2012)-proved less sensitive to ITZ than wild-type (WT) CVB3 in both a single-cycle replication assay ( Figure 2E) and a multicycle CPE-reduction assay ( Figure 2F). Other mutations in 3A that were shown to protect against PI4KIIIb inhibitors (i.e., V45A and I54F) (van der Schaar et al., 2012), also provided cross-resistance to ITZ ( Figure 2G). Similarly, mutation A70T in PV 3A, which was also shown to protect against PI4KIIIb inhibitors (Arita et al., 2009), protected PV against ITZ ( Figure 2H). These results imply a link between 3A, PI4P lipids, and the mechanism of antiviral action of ITZ.
To determine whether ITZ inhibits PI4KIIIb activity, we transiently transfected cells with a genetically encoded PI4P sensor, i.e., the GFP-tagged PH domain of FAPP1 (FAPP1-PH-GFP). Localization of this sensor specifically depends on activity of PI4-KIIIb (Balla et al., 2005;van der Schaar et al., 2012). In control cells, FAPP1-PH-GFP overlapped with the Golgi-localized PI4-KIIIb ( Figure 2I). Upon treatment with a PI4KIIIb inhibitor, PIK93, FAPP1-PH-GFP was redistributed to the cytosol. ITZ, however, did not decrease FAPP1-PH-GFP localization. In fact, ITZ caused a small increase in the amount of Golgi-localized FAPP1-PH-GFP, which was more apparent in a cell line stably expressing this PI4P sensor (which showed a more homogenous and moderate expression level) ( Figure S4A). Also upon staining PI4P with a specific antibody, a PI4KIIIb inhibitor, BF738735 (van der Schaar et al., 2013), decreased PI4P levels, whereas ITZ increased PI4P levels ( Figure S4B).
CVB3 replication is not completely blocked by ITZ (see e.g., Figures (legend continued on next page) in treated cells. Both in untreated and ITZ-treated cells infected with CVB3 (visualized using 3A antibody), PI4P levels (visualized using PI4P antibody) were clearly increased compared to uninfected cells ( Figure S4C), indicating that ITZ also does not inhibit PI4KIIIb activity in infected cells.
ITZ Inhibits Virus Replication by Targeting OSBP and ORP4
Having ruled out PI4KIIIb as a target of ITZ, we next turned to signaling steps downstream of PI4P, i.e., proteins that bind to PI4P lipids. To assess whether any of the known PI4P-binding proteins could be a target of ITZ, we performed a target identification by small interfering RNA (siRNA) sensitization (TISS) assay (Arita et al., 2010). TISS encompasses siRNA knockdown of candidate target proteins to potentiate the biological effect of a low concentration of a compound. Among a number of PI4Pbinding, Golgi-localized proteins, knockdown of OSBP, but not any of the other PH domain-containing proteins, enhanced the inhibitory effect of a low concentration (1.25 mM) of ITZ on PV replication ( Figure 3A), implying OSBP as a possible antiviral target of ITZ. We further assessed this possibility by several experiments. First, the OSBP antagonist OSW-1 (Burgett et al., 2011) potently inhibited CVB3 replication ( Figure 3B), confirming that pharmacological targeting of OSBP can inhibit enterovirus replication. As for ITZ, the 3A[H57Y] mutation in CVB3 provided resistance against OSW-1 ( Figure 3B). Akin to ITZ, OSW-1 inhibited all enteroviruses tested as well as EMCV, but not ERAV (data not shown). Importantly, OSW-1 did not affect endolysosomal cholesterol distribution ( Figure 3C), supporting our previous conclusion that this effect unlikely explains the antiviral effect of ITZ. Second, similar as for PV (Wang et al., 2014), siRNA knockdown of OSBP inhibited replication of EV71 and HRV2 (Figure 3D). CVB3 replication was also inhibited by OSBP knockdown, but this difference was not statistically significant, in line with the lower sensitivity of CVB3 than EV71 to ITZ ( Figure 1A). Third, overexpression of OSBP restored replication of CVB3 and EV71 in the presence of ITZ or OSW-1 ( Figure 3E), confirming that inhibition of viral replication by ITZ and OSW-1 is mediated through OSBP. Overexpression of PI4KIIIb failed to rescue replication, and OSBP overexpression did not provide rescue against PI4KIIIb inhibitors (data not shown), indicating the specificity of the experimental setup.
Besides OSBP, OSW-1 also targets ORP4 (Burgett et al., 2011). Knockdown of ORP4, but none of the other ORPs, also sensitized PV to ITZ ( Figure 3F), and overexpression of ORP4 counteracted the inhibitory effect of OSW-1 on CVB3 and EV71 replication ( Figure 3G). We also attempted to test the effect of ORP4 depletion. Although in the TISS assay, ORP4 knockdown potentiated the effect of ITZ, we were not able to achieve robust knockdown (>75% at mRNA level), and therefore we cannot conclude unambiguously whether ORP4 is important for virus replication. Problems with ORP4 knockdown were also observed by others and are likely due to an essential role of ORP4 in cell proliferation and survival (Charman et al., 2014). Collectively, these results indicate that both OSBP and ORP4 are novel targets of ITZ and are involved in its mechanism of antiviral action.
ITZ Inhibits In Vitro HCV Replication
Replication of HCV also requires OSBP and is inhibited by OSW-1 (Wang et al., 2014). In line with our findings for enteroviruses, we found that ITZ and posaconazole, but not the other selected azoles, inhibited HCV replication in cell culture (Figure S6). EC 50 values for inhibition of HCV replication by ITZ were comparable to those obtained for the enteroviruses (Table S1). Together, our data clearly demonstrate that ITZ inhibits OSBP function and that viruses from different families that depend on OSBP function can be inhibited by ITZ. Importantly, not all (+)RNA viruses are sensitive to inhibition of OSBP. Dengue virus replication was recently observed to be insensitive to OSW-1 (Wang et al., 2014), and we also showed that replication of mouse hepatitis virus (a coronavirus) is insensitive to OSW-1 and ITZ (data not shown).
Treatment with ITZ Results in Relocalization of OSBP to the Golgi Complex
After having established that OSBP and, possibly, ORP4 are novel targets of ITZ, we wanted to study how ITZ targets these proteins. Because of available tools, we focused on OSBP for the remainder of this study. In line with published data (Burgett et al., 2011), OSW-1 caused a massive recruitment of overexpressed GFP-OSBP to the Golgi apparatus (as marked by staining endogenous PI4KIIIb) ( Figure 4A). A similar relocalization was observed for endogenous OSBP ( Figure 4B). ITZ and posaconazole, but not the azoles that lacked antiviral activity, redistributed OSBP in a manner that is similar to OSW-1. Live-cell imaging was performed to study the dynamics of GFP-OSBP relocalization by the compounds. Before addition of the compounds, GFP-OSBP primarily localized in the cytosol with a Golgi pattern faintly visible. A few minutes after the addition of the compounds, (C) HAP1 cells were treated for 6 hr with 10 nM OSW-1 or 10 mM ITZ, fixed, and stained with filipin. (D) HeLa R19 cells were transfected with constructs encoding OSBP or EGFP (negative control) for 24 hr, infected with RLuc-CVB3 at MOI 0.25 or EV71 at MOI 1, and treated with 10 mM (CVB3) or 3 mM (EV71) ITZ, 3 nM OSW-1, or DMSO. Renilla luciferase levels were determined at 7 hr p.i. (CVB3) or virus titers at 10 hr p.i. were determined by endpoint titration (EV71). (E) HeLa R19 cells were transfected with siRNAs against OSBP, PI4KIIIb (positive control), or a scrambled siRNA for 2 days and infected with CVB3, EV71, or HRV2 at MOI 1. Virus titers at 10 hr p.i. were determined by endpoint titration. Knockdown efficiency was determined by quantitative PCR and immunofluorescence ( Figure S5), and an MTS assay was used to test for effects on cell viability. (C) HeLa R19 cells were transfected with GFP-OSBP and treated with DMSO, 10 mM of ITZ, or 10 nM OSW-1, and the relocalization of OSBP was imaged by livecell confocal laser scanning microscopy. Cells were imaged overnight. During the first 30 min, images were taken as fast as possible (1.5 min intervals), then intervals were stepwise increased to 30 min from 3.5 hr onward. Representative groups of cells are shown. The images are frames from Movie S1. (D) Quantification of the relative GFP-OSBP signal at the Golgi apparatus in five cells for each condition from (C). Error bars indicate SEM. Scale bar corresponds to 10 mm. See also Movie S1.
(legend on next page) GFP-OSBP fluorescence at the Golgi was clearly increased in cells treated with either ITZ or OSW-1 and continued to increase at the expense of the cytoplasmic signal ( Figures 4C and 4D; Movie S1). OSW-1 was previously reported to disrupt the structure of the Golgi apparatus (Burgett et al., 2011), which we also observed from 30 to 60 min onward as GFP-positive punctae that became more numerous over time ( Figure 4C). In ITZtreated cells, the Golgi pattern became affected only hours later and appeared less dispersed than that in OSW-1 treated cells.
ITZ Directly Inhibits Lipid Shuttling by OSBP
To investigate whether ITZ can block the lipid transfer activity of OSBP, we used a set of in vitro liposomal assays (Mesmin et al., 2013) (Supplemental Experimental Procedures) to measure the transport of dehydroergosterol (DHE) ( Figure 5A) and PI4P (Figure 5D) between ER-like and Golgi-like liposomes. ITZ inhibited the sterol-transfer activity of purified OSBP in a dose-dependent manner with a 50% inhibitory concentration (IC 50 ) of 200 nM ( Figure 5B). At 1 mM, ITZ and posaconazole, but not the other selected azoles, strongly inhibited DHE transfer transport in this liposomal assay, although they were less potent than the known OSBP ligand 25OH ( Figure 5C). We also observed a dose-dependent inhibition of PI4P transfer by ITZ (IC 50 = 4 mM) ( Figure 5E). Posaconazole slightly inhibited PI4P, whereas the other azoles showed no activity ( Figure 5F). For unknown reasons, a stimulatory effect of 25OH on PI4P transfer was observed, which depended on the 2% cholesterol content of the ER-like liposomes. The IC 50 values suggest that ITZ is more potent toward sterol than PI4P transfer. Importantly, for technical reasons, the sterol and PI4P-shuttling assays are performed under different conditions and therefore cannot be directly compared. Further investigations would be needed to establish whether ITZ indeed more potently inhibits sterol than PI4P shuttling.
ITZ may inhibit the lipid transfer functions of OSBP directly by inhibiting the function of the ORD, which transfers the lipids, or indirectly by disrupting the binding of OSBP to the liposomes. To investigate whether ITZ inhibits binding of OSBP to the liposomes, we studied whether it interferes with the interactions between (1) the FFAT-motif and VAP-A on the ER-like liposomes and (2) the PH-domain and PI4P on the Golgi-like liposomes. To this end, we performed liposomal float-up experiments using a recombinant fragment of OSBP containing the PH domain and FFAT motif (amino acids 76-408; PH-FFAT) ( Figures 5G and 5I). In the presence of VAP-A, PH-FFAT bound to the ER-like liposomes, and this interaction was not disrupted by 1 mM ITZ ( Figure 5H). The interaction of PH-FFAT with PI4P-containing Golgi-like liposomes was not disrupted by 10 mM ITZ either (Figure 5J). Likewise, VAP-A interaction with PH-FFAT recruited to Golgi-like liposomes was also insensitive to 10 mM ITZ (Figure 5J). Together, the liposomal float-up assays show that ITZ does not interfere with the binding of OSBP to the liposomes via VAP-A and PI4P.
To establish whether ITZ inhibits the lipid transfer activity of the ORD, we made use of a previously established assay (Mesmin et al., 2013). Limited tryptic proteolysis of OSBP cleaves OSBP into three major fragments; a 43 kDa fragment containing the PH-domain and FFAT-motif, and two fragments of 35 kDa and 20 kDa that are derived from the ORD. Previously, it was shown that the ORD-derived fragments retain lipid transfer activity, also in the absence of the inactive 43 kDa fragment (Mesmin et al., 2013). We found that ITZ still inhibited both DHE ( Figure 5K) and PI4P ( Figure 5L) transfer by OSBP that had been subjected to tryptic proteolysis ( Figure S7A). These results suggest that ITZ inhibits both the sterol-and PI4P-transfer activities of OSBP by targeting the ORD.
ITZ Binds Directly to OSBP
The inhibitory effect of ITZ on OSBP function in a minimal in vitro system implied that ITZ directly inhibits OSBP. To biochemically define the binding in more detail, we measured binding of ITZ to GFP-OSBP using microscale thermophoresis (MST). Each molecule or complex distributes differently in a temperature field, depending on size, charge, and the hydration shell. Binding of ITZ to OSBP will affect the hydration shell and thereby its thermophoretic behavior. ITZ altered the thermophoretic profiles of purified GFP-OSBP ( Figures S7B and S7C) in a dose-dependent manner, indicating direct binding. Normalization and fitting of data from three independent measurements demonstrated that ITZ binds to OSBP with a K D of 430 nM ( Figure 5M). The monophasic shape of the binding curve indicates that there is likely only a single binding site for ITZ on OSBP, although our data cannot rule out that there are two sites with nearly identical K D 's.
OSBP Localizes to ROs in a PI4P-Dependent Manner
To test whether OSBP plays a role in formation and/or maintenance of the ROs, we examined its localization in infected cells. In uninfected cells, OSBP is mainly distributed throughout the cytosol with some OSBP localized to the Golgi apparatus (Figure 6A), where it colocalized with PI4KIIIb and the trans-Golgi network marker TGN46 (data not shown). In infected cells, OSBP localization was markedly changed, i.e., the Golgi pattern was lost and OSBP appeared in dispersed structures throughout (M) The interaction of ITZ with GFP-tagged OSBP was investigated using MST. Data from three separate measurements were normalized and plotted, and a sigmoidal dose-response curve was fitted. Shown are mean values ± SEM. Statistical significance for the drug-treated conditions was calculated compared to the ''no drug'' control. See also Figure S7. the cytoplasm where it colocalized with viral protein 3A as a marker for ROs (Manders' coefficient for overlap of 3A with OSBP: 0.36) (Figures 6A and 6B). To examine whether OSBP localized to ROs in a PI4P-dependent manner, cells were infected with CVB3 and replication was allowed to progress uninhibited for 4.5 hr before the PI4KIIIb inhibitor BF738735 was added for 30 min, after which cells were processed for microscopy. Inhibition of PI4KIIIb decreased colocalization of OSBP with 3A (Manders' coefficient: 0.11), whereas treatment with ITZ increased colocalization of OSBP with the RO-marker 3A (Manders' coefficient: 0.62) (Figures 6A and 6B). The enhanced recruitment of OSBP to ROs upon ITZ treatment is reminiscent Figure 6. ITZ Affects OSBP Localization in Infected Cells (A) BGM cells were infected with CVB3 at MOI 10. At 4.5 hr p.i., cells were treated for 30 min with DMSO as vehicle control, 1 mM BF738753 (BF; a PI4KIIIb inhibitor), or 10 mM ITZ. At 5 hr p.i., cells were fixed, processed for immunofluorescence with antibodies against OSBP and viral protein 3A, and imaged using confocal laser scanning microscopy. (B) Manders' coefficients for overlap of 3A with OSBP were calculated for DMSO (12 cells), BF (7 cells) and ITZ (10 cells). Shown are means ± SEM. Asterisks indicate statistical significance compared to DMSO-treated controls. Scale bars correspond to 10 mm.
of the enhanced Golgi-localization of OSBP upon treatment of ITZ or other OSBP inhibitors, and may therefore be caused by an inhibition of PI4P removal from ROs.
ITZ Inhibits PI4P and Cholesterol Shuttling at ROs
To test directly whether ITZ inhibits the PI4P shuttling function of OSBP at ROs, cells were infected with CVB3 and replication was allowed to progress uninhibited for 3 hr. Then ITZ or BF738735 were added for 1 hr, cells were processed for microscopy, and PI4P intensity at ROs was quantified. ITZ treatment caused a strong increase in PI4P signal at the ROs (50% increase), whereas BF738735 treatment reduced it by 50% ( Figures 7A and 7B), in line with the effects of these drugs on OSBP recruitment ( Figure 6A). No such effects on PI4P were observed upon treatment with guanidine, an inhibitor of the viral 2C protein, which was included to rule out that the observed effects were merely due to an inhibition of replication. Thus, these results demonstrate that in infected cells, ITZ prevents the removal of PI4P from ROs, which is comparable to our observations in uninfected cells (Figures 2I, S4A, and S4B).
To test whether ITZ also inhibits cholesterol shuttling to ROs, cells were infected and treated similar as described above, cholesterol was visualized by filipin staining, and colocalization of filipin with 3A was quantified using a Pearson's correlation coefficient. In DMSO-treated cells, filipin partially overlapped with 3A (Pearson's 0.53). ITZ significantly reduced the colocalization of filipin with 3A (Pearson's 0.38), indicating that ITZ inhibited the redistribution of cholesterol to the ROs (Figures 7C and 7D). Similarly, BF738735, which reduces the localization of OSBP to ROs (Figure 6A), also inhibited cholesterol shuttling to ROs (Pearson's 0.35), whereas guanidine did not decrease the overlap between filipin and 3A. Thus, we demonstrate that OSBP is recruited to ROs through the action of PI4KIIIb and that ITZ inhibits both the PI4P and the cholesterol-transfer functions of OSBP in infected cells.
DISCUSSION
Enteroviruses alter cellular lipid homeostasis and remodel hostcell membranes into replication organelles by usurping a number of host proteins, such as PI4KIIIb (Arita et al., 2011;Hsu et al., 2010). However, as yet little is known about the underlying mechanisms and the identity of other host factors involved. Elucidation of the mechanism of action of inhibitors of virus replication has proven instrumental in obtaining novel insights into the mechanisms of viral replication. In this study we identified ITZ, a widely used antifungal drug that is currently also being explored as an anticancer agent, as a novel, broad-spectrum inhibitor of enteroviruses, cardioviruses, and HCV. We show that none of the well-established targets of ITZ (i.e., hCYP51, mTOR, VEGFR2, Hh) explains its antiviral activity. Instead, we identified the PI4P-binding proteins OSBP and ORP4 as novel targets of ITZ through which the antiviral effect is mediated.
OSBP is a master regulator of lipid homeostasis at MCSs between the ER and the trans-Golgi apparatus. It exchanges cholesterol and PI4P between these membranes and has been proposed to control MCS stability (Mesmin et al., 2013). OSBP is the prototype member of the family of ORPs, a group of proteins whose cellular functions have remained poorly understood. We identified OSBP and ORP4 as targets of ITZ. Pharmacologic inhibition, siRNA knockdown, and rescue of replication by overexpression demonstrate the importance of these proteins for virus replication. Furthermore, OSBP localized to ROs in a PI4-KIIIb-and PI4P-dependent manner. ITZ directly bound purified OSBP and inhibited both the cholesterol and PI4P-transport activities of OSBP in vitro (in liposomal assays). Also in living (uninfected) cells, ITZ inhibited the transport function of OSBP (i.e., transport of cholesterol from ER to Golgi and transport of PI4P from Golgi to ER), leading to an increase in PI4P levels at the Golgi, thereby causing the accumulation of OSBP. Likewise, in infected cells, ITZ increased PI4P levels on ROs, again leading to an enhanced recruitment of OSBP, and inhibited the accumulation of cholesterol on ROs. Thus, we demonstrate that ITZ inhibits the lipid-shuttling functions of OSBP not only in vitro but also in both infected and uninfected cells.
The enteroviral proteins 2BC and 3A play a critical role in RO formation by recruiting PI4KIIIb, which leads to the accumulation of PI4P lipids on ROs (Arita, 2014;Arita et al., 2011;Hsu et al., 2010). We here show that OSBP is subsequently recruited to ROs via PI4P. Our data indicate that at ER-RO MCSs, OSBP exchanges PI4P for cholesterol, either newly synthesized in the ER or originating from a lipid droplet storage pool and being mobilized through the ER, leading to an accumulation of cholesterol at the ROs (Arita, 2014). Our findings are in agreement with those of a recent paper that suggested that OSBP shuttles cholesterol to HRV ROs based on the inhibitory effects on HRV replication of OSBP knockdown and 25OH treatment (Roulin et al., 2014). The finding that the levels of cholesterol are elevated at the expense of cholesterylesters (i.e., the form in which cholesterol is stored in lipid droplets) in enterovirus-infected cells (Ilnytska et al., 2013;Roulin et al., 2014) suggests that stored cholesterol is mobilized for transport to ROs. In addition, uptake of cholesterol by endocytosis has been suggested to contribute to the accumulation of cholesterol at ROs (Ilnytska et al., 2013). The role of cholesterol accumulation at ROs is far from established. Cholesterol is of profound importance for membranes properties such as membrane fluidity and formation of lipid microdomains, and it is thereby likely important for the membrane rearrangements and deformations underlying RO formation. In addition, cholesterol alterations have been suggested to affect viral polyprotein processing efficiency (Ilnytska et al., 2013).
The activity of OSBP is also important for the homeostasis of other lipids. At ER-Golgi MCSs, it acts in concert with the PI transfer protein Nir2, which supplies PI for PI4P synthesis at Golgi membranes, and CERT, which transfers ceramide to Golgi for sphingomyelin synthesis, thereby generating diacylglycerols (DAGs) (Peretti et al., 2008). Importantly OSBP ligands, e.g., 25OH and OSW-1, change the localization of CERT and modify sphingomyelin synthesis (Burgett et al., 2011;Perry and Ridgway, 2006). As an inhibitor of OSBP-mediated lipid shuttling, ITZ may thus not only affect the accumulation of cholesterol but also perturb the homeostasis of other lipids, such as sphingomyelin and DAGs. Whether and how this contributes to the inhibition of RO formation and/or function remains to be established.
Our study and the work by Arita et al. (2013) implicate a role for ORP4 in addition to OSBP in enterovirus replication. Unfortunately, little is known about the biological function of ORP4. Roles for ORP4 are proposed in organization of the cytoskeletal vimentin network, cell proliferation and survival, and sterol transfer (Charman et al., 2014;Wang et al., 2002a). However, unlike OSBP, ORP4 does not localize to the Golgi under normal conditions or in response to a ligand such as 25OH (Charman et al., 2014;Wang et al., 2002a). It therefore seems unlikely that ORP4 transports cholesterol between the ER and Golgi in a similar manner as OSBP. How ORP4 overexpression can counteract the inhibitory effect of ITZ on virus replication thus remains to be established. It is possible that OSBP-ORP4 heteromultimers (Wyles et al., 2007) are important for virus replication, but this requires further investigation. Besides OSBP and ORP4, other ORPs did not appear to be targeted by ITZ, though they may still be important for virus replication.
ITZ has been shown to inhibit angiogenesis (via mTOR and VEGFR2) and growth of Hh-dependent cancer cells, but the exact molecular mechanisms of the antitumor activities of ITZ await elucidation. It remains to be established whether OSBP inhibition contributes to the anticancer activities of ITZ via these pathways. OSBP overexpression, which we showed to counter the antiviral activity of ITZ, failed to prevent the inhibitory effects of ITZ on mTOR and Hh signaling (not shown). These observations suggest that ITZ does not inhibit these antitumor pathways through OSBP, but we cannot exclude that the overexpression approach can only neutralize the antiviral effect of ITZ. Therefore, more work is needed to establish whether or not ITZ exerts its antitumor activities via OSBP and/or ORP4. OSW-1 and several other natural products were recently reported to inhibit the growth of cultured human cancer cell lines through OSBP and ORP4 and therefore collectively termed ORPphilins (Burgett et al., 2011). Our data that ITZ targets OSBP and ORP4 justify classifying ITZ as a novel ORPphilin. It is plausible that ITZ inhibits OSBP/ORP4-dependent cancer cell growth and survival in a manner independent of, and in addition to, mTOR, VEGFR2, and Hh. Recently, two inhibitors of PV replication were shown to target OSBP and ORP4 (Arita et al., 2013), although binding to OSBP has yet to be shown, and may therefore also classify as ORPphilins.
In conclusion, we identified ITZ as a broad-spectrum inhibitor of enterovirus, cardiovirus, and HCV replication that exerts its antiviral activity through the novel targets OSBP and ORP4, presumably by inhibiting the lipid-shuttling functions of OSBP. Together, our study provides insight into enterovirus replication and presents ITZ, OSW-1, and other ORPphilins as potential novel inhibitors to treat enterovirus infections.
EXPERIMENTAL PROCEDURES
Details about published and standard methods (cell culture, plasmids, virus infections, replicon transfections, the TISS assay, rescue experiments, analysis of viral polyprotein processing, siRNA experiments, immunofluorescence microscopy, and liposomal assays) are provided in Supplemental Experimental Procedures.
Compound Library Screen
The NIH Clinical Collection was purchased from the NIH. The 446 highly druglike compounds were screened for inhibitors of CVB3 using reduction of CPE as readout. Subconfluent monolayers of Buffalo green monkey kidney (BGM) cells in 96-well plates were infected with 10 CCID 50 of CVB3 per well, compounds were added to a final concentration of 10 mM, and the level of CPE was visually assessed after 2 days of incubation at 37 C when full CPE had developed in the infected, untreated control wells.
Live-Cell Imaging
For live-cell imaging experiments, HeLa R19 cells were transfected with pEGFP-hOSBP; treated with ITZ, OSW-1, or solvent control (DMSO); and imaged using a Nikon A1R confocal laser scanning microscope. Images were processed and quantified using the Nikon NIS-Elements software. For additional details, see Supplemental Experimental Procedures.
Microscale Thermophoresis
The interaction between ITZ and recombinant GFP-hOSBP-SII (human OSBP with an N-terminal GFP and a C-terminal Strep-tagII) was investigated by MST using a NanoTemper Monolith NT.115 instrument and the NTAnalysis software (NanoTemper Technologies).
Statistical Analyses
Data are presented as means ± SEM. Replication data were analyzed by pairwise comparisons of conditions using one-tailed Student's t test. Statistics significance is indicated as *p < 0.05, **p < 0.01, or ***p < 0.001.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures, seven figures, one table, and one movie and can be found with this article online at http://dx.doi.org/10.1016/j.celrep.2014.12.054.
|
2020-04-10T20:44:14.226Z
|
2015-01-29T00:00:00.000
|
{
"year": 2015,
"sha1": "5d3cad66ff27b7d70f1895c653f79d63379af069",
"oa_license": "CCBYNCND",
"oa_url": "http://www.cell.com/article/S2211124714011218/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "2ea03f350c47cb3fe3da7b5af6b36f2a5fb03e14",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Medicine",
"Biology"
]
}
|
157659671
|
pes2o/s2orc
|
v3-fos-license
|
Performance Analysis of Steam Power Plants Using Ideal Reheat-Rankin Cycle
In this paper, a hypothetical examination has been done to assess the execution of the power plants that are chipping away at Reheat-Rankin cycle. The execution of cycle was dissected for various (warm, evaporator, condenser weights) values and also warm temperature qualities to demonstrate its impact on cycle warm proficiency. In this work, the heater weights qualities was accepted limited between (10to 26 MPa), the pressure proportion (warm stage weight to evaporator weight) was expected fluctuated in wide range from (0.1 to 1.0), while the condenser weight was accepted shifted between (5 to 25 kPa). And, a variety in warm temperature esteem was done between scopes of (400-600oC) at low weight turbine. The outcomes demonstrate that the warm productivity is considerably upgraded when the pressure proportion lies between (0.25-0.35) and the ideal proficiency is gotten when the pressure proportion and evaporator weight are equivalent to( 0.33 and 26MPa) separately .
INTRODUCTION
Fundamental steam control plants work in view of Rankin cycle. Numerous methods are being utilized to expand the effectiveness of the steam cycle plant at various conditions. Most extraordinary of these procedures are warming as well as decreasing the irreversibility .The warming procedure builds the proficiency of cycle by raising the mean temperature of the warmth expansion prepare by expanding the steam temperature at the principal organize turbine outlet furthermore, expanding the productivity of the extension procedure in the steam turbine [1]. The impacts of evaporator temperature, weight, and the irreversibility's on the execution of single and twofold warm irreversible Rankin cycles are researched. The impacts of first and second stage warm weights on the general execution and proficiency increment rates for the single and twofold warmed Rankin cycles are additionally examined [2]. A thermodynamic examination of a solitary warm Rankin cycle for the steam control plants in light of the first and second laws of thermodynamic has been performed. The vitality and energy efficiencies for various framework parameters have been concentrated, for example, heater temperature, kettle weight, mass part proportion and work yield, [3]. The vitality and energy investigation of a perfect Rankin cycle with warm had been exhibited notwithstanding break down the framework segments independently and to distinguish and measure the locales having biggest vitality and energy misfortunes at cycle. The impact of changing the condenser weight on this examination will likewise be introduced [4]. Another specialists were clarified the endeavors of progression in the best in class in steam turbine innovation, which is identified with the change in the thermodynamic proficiency by expanding the temperature and weight at which warmth is added to the power cycle. The steam turbine was explored for ultrasupercritical power plants and reported warmth rate change of single and twofold warm steam cycles with ultrasupercritical steam conditions on their work [5]. Energy investigation and parametric examination have been conveyed for supercritical Rankin cycle with warm, open and shut sustain water radiators of higher power era for present day steam control plants at supercritical, ultra supercritical and progressed ultra-supercritical temperatures and weights [6]. The vitality and energy examination of a warm regenerative vapor control cycle has been portrayed. The plant comprises of one evaporator sustain pump, one supercritical kettle, two steam turbines with a warmed in the middle of, two nourish water radiators and a condenser. The vitality and energy adjust think about has been completed for every segment of the plant. Vitality effectiveness, energy proficiency and the irreversibility comes about got from the reproduction has been displayed as diagrams [7]. The streamlining of warm regenerative warm power plants has been broke down for the subcritical weight territory, [8][9]. The exegetic examination and streamlining has been accomplished for the supercritical Rankin cycle, [10]. In this examination, the execution of the steam control plant chipping away at Reheat-Rankin cycle has been broke down for various (heater weight, condenser weight, and warm 4 to 1: Isobaric warm dismissal (Condenser), an isobaric procedure, in which the working liquid change over from steam to fluid stage at consistent weight. In the examination of a warmth motor cycle, it is expected that the liquid stream takes after the progression condition relevant to relentless state conditions. All the power plants are expected to keep running under consistent state conditions. The beginning and closing down working condition is exempted where deviations from an enduring state can't be maintained a strategic distance from. With reference to the Rankin cycle that appeared in Figs. (1a) and (1b), the control of the stream can be practiced by a valve. Fractional nearer of the valve would diminish the stream of steam in the turbine and power yield. In the perfect Rankin cycle, working liquid takes after reversible an adiabatic or isentropic way in the turbine and is subjected to lower weight and temperature in the condenser. The condenser is an expansive shell-and-tube sort warm exchanger. This is situated beside the turbine so as to get a vast stream rate of low weight steam. This steam in the condenser goes under a stage change from vapor to fluid water. Outside cooling water is pumped through a huge number of tubes in the condenser to transport the warmth of buildup of the steam far from the plant. After leaving the condenser, the condensate is at a low temperature and weight. Evacuation of this condensate might be considered as keeping up the low weight in the condenser constantly.
The stage change thusly relies on upon the exchange of warmth to the outside cooling water. The dismissal of warmth to the surroundings by the cooling water is basic to keep up the low weight in the condenser. Applying the consistent stream first law of thermodynamics to the consolidating steam empowers [11]: The estimation of (Qcond.) is negative on the grounds that h4>h1. In this manner, predictable with sign tradition, (Qcond.) speaks to a surge of warmth from the consolidating steam. This warmth is consumed by the cooling water going through the condenser tubes.
Applying the consistent stream first law of thermodynamics condition (1) to the steam generator with shaft work equivalent to zero, then the steam generator warm exchange is given by the accompanying condition [12]
RANKIN WITH REHEAT CYCLE ANALYSIS:
In basic Rankin cycle, after the isentropic development in turbine, steam is specifically nourished into condenser for buildup prepare, however in warm framework, two turbines (high weight turbine and low weight turbine) are utilized for enhancing productivity. Steam, after development from high weight turbine, is sent again to heater and warmed till it comes to superheated condition. It is then left to grow in low weight turbine to achieve condenser weight. The cycle comprises of six procedures as appeared in Figs. (2a) and (2b). 1 to 2: Isentropic pressure (in a pump) An isentropic procedure, in which the entropy of working liquid stays consistent. 2 to 3: Constant weight warm expansion (in a kettle) An isobaric procedure, in which the weight of working liquid stays consistent. 3 to 4: Isentropic development (in a high weight turbine). 4 to 5: Reheat steam (in a heater). 5 to 6: Isentropic development (in a low weight turbine). 6 to 1: Constant weight warm dismissal (in a condenser). The Rankin cycle has been altered to deliver more yield work by presenting two steam turbine stages utilizing middle of the road warming. Essentially, in this work the changed Rankin cycle, the full development of steam is hindered in the high-weight turbine and steam is released after incomplete extension. This fumes steam is gone through an icy warm line to the steam generator where it picks up warmth while going through hot tubes. This warmed steam is provided to a low weight turbine for full development and achieves the condenser weight. The changed Rankin T-S graph demonstrates some expansion in the zone which is accomplished on account of the presentation of the warming of steam. These outcomes in low weight turbine development work. Consequently, warm builds work out. This results in low pressure turbine expansion work. Thus, reheat increases work output because of low pressure turbine expansion work. The use of reheat also tends to increase the average temperature at which heat is added. If the steam from the low-pressure turbine is superheated, the use of reheat may also increase the average temperature at which heat is rejected. It may reduce or increase the thermal efficiency depending on the specific cycle conditions, thermal (heat addition and heat reduction temperature) or mechanical (condenser vibrations and air leakage, pump vibrations, turbine blade vane deflection etc.). Indeed the work net of the reheat cycle is the algebraic sum of the work of the two turbines and the pump work and the total heat addition is the sum of the heat added in the feed-water and reheat passes through the steam generator.
III. RESULTS AND DISCUSSIONS
Steam is the most widely recognized working liquid utilized as a part of vapor power cycles as a result of its numerous alluring attributes, for example, minimal effort, accessibility, and high enthalpy of vaporization. Steam warming is a critical component in steam-control plants. The principle target of warm is to build the power yield and, under specific conditions, the warm productivity of the plant, in this manner enhancing plant execution. In a steam control plant, bolster water radiators permit the sustain water to be raised to the immersion temperature steadily. This minimizes the inescapable irreversibility's connected with warmth exchange to the working liquid. The execution of the warm power plant was investigated with a subcritical and supercritical heater. This kind of heater is working at weights and temperatures of (11MPa and 500oC)for subcritical kettle, or more (22MPa and 600oC) for supercritical evaporator [13]. Additionally the warm temperature for low weight turbines was found around to 500oC.This temperature was exhibited in many power plants boilers, for example, cutting edge pummeled coal-let go kettle that intended to item steam at 540oC for both superheat and warm [14]. And in addition, the examinations manage condenser weight (turbine backpressures) between (5-25 kPa). Turbine backpressures for the most part range from (3.5 kPa to 16 kPa) [14]. The power plant was broke down utilizing the above information. The designing condition solver or (EES, v.8.0) was utilized to discover the warmth stream and work done of every cycle component, and the general cycle warm effectiveness by utilizing the above conditions. Presently, Fig. (3) Shows the impact of pressure proportion (the proportion of warm weight to kettle weight) on the general proficiency of Reheat Rankin Cycle for various heater weights values (10, 14, 18, 22, and 26 MPa) at consistent condenser weight (10 Kpa), and also steady delta turbine temperatures (high and low weight turbines bays) at (500oC). Be that as it may, the effectiveness of the cycle increments bit by bit with a light increment in pressure proportion till coming to crest estimation of (44.12%) at a pressure proportion of (0.33) at kettle weight of (PB = 26 MPa).While the estimations of most noteworthy efficiencies were discovered lower this esteem at other heater weights and lower pressure proportions. The most extreme ://dx.doi.org/10.24001/ijaems.3.4.4 ISSN: 2454-1311 www.ijaems.com Page | 309 estimation of the proficiency on account of heater weight (PB= 10 MPa), which is discovered equivalent to (41.71%) at a pressure proportion equivalent to (0.25). That way to acquire most extreme effectiveness at low kettle weights; it must be lessen the pressure proportion of the cycle to get a higher proficiency for this cycle.
Fig. 3: Overall efficiency as capacity of correlation proportion at various heater weight values.
In Figs. (4) and (5) the impact of warm temperature of second turbine gulf temperature (low weight turbine) has been researched at temperatures near the (high weight turbine) first turbine bay temperature (400-600oC), at a few heater weights (10, 14, 18, 22, and 26MPa) and steady condenser weight (10 kPa). The reason for these figures, to examine the cycle proficiency and dryness rubbing (X6) at consistent pressure proportion (0.2) with keeping the turbine channel temperature of high weight turbine near (500oC). From Figs. (4) and (5) plainly the expansion in warm temperature at all evaporator weights will prompt to enhance the cycle productivity and expanding the estimation of dryness erosion. In this work, the nature of dryness part has been explored to keep up a nature of the dryness grinding around 90% in the turbine fumes to maintain a strategic distance from the disintegration in sharp edges close to the turbine exit, furthermore to build the vitality transformation effectiveness and working work of turbine as specified in writing [15]. /dx.doi.org/10.24001/ijaems.3.4.4 ISSN: 2454-1311 www.ijaems.com Page | 310 low weight turbines) at (500oC). It appears that lessening in the cycle pressure proportion prompts to increment in dryness portion for all kettle weights values. The impact of condenser weight on general effectiveness has been examined at various evaporator weight values as clear.
In Fig. (7).It clears that, as the condenser weight diminishes, causes to general effectiveness increments for all kettle weight values. Along these lines, the diminishing of the condenser working weight naturally brings down the temperature of the steam, and brings down the temperature at which warmth is rejected. Along these lines the general impact of diminishing the condenser weight is an expansion in the warm productivity of the cycle [11]. It appears to be likewise, as the evaporator weight diminishes causes to lessening in dryness portion as appeared.
In Fig. (8), which adversely impacts on the execution of the low weight turbine. For instance it can be lessen the estimation of condenser weight to (5kPa) when the heater weight is alter to (10MPa), while can't be decrease condenser weight under (13kPa) when the kettle weight is equivalent to (26MPa) on the grounds that the disintegration of the turbine at the last stage edges, and in addition the pulverization in the condenser weight increment the misfortune in active vitality of the steam stream because of dragging of dampness particles that have less speed from steam. Thusly, the condenser ideal weight territory is restricted between (6.9 kPa and 13.8 kPa)as specified in writing [4]. Fig. (9) Shows the aggregate system, add up to information warm stream, and warm proficiency as capacity of pressure proportion for a heater weight esteem of(10 MPa). In this figure, the warm proficiency, the aggregate system, and aggregate info warm stream for any pressure proportion can be effortlessly assessed.
IV.
CONCLUSIONS A hypothetical examination and investigation for Reheat-Rankin cycle has been accomplished for various (evaporator, condenser weights, and warm temperatures) to demonstrate its impact on the warm power plants general effectiveness. The effectiveness of a basic Rankin cycle is enhanced by utilizing moderate warm process. Henceforth, the accompanying conclusions have been considered as the synopsis of this work: 1. Increment the warm proficiency of the cycle with increment in estimations of kettle weight at any estimation of the pressure proportion. Additionally unmistakably, the expansion of warm proficiency was finished by increment in pressure proportion values for any evaporator weight esteem until it spans to pinnacle esteem then begins to diminish consistently with ceaselessly increment in the warm temperature values. 2. Increment the estimations of warm productivity with increment in warm stage temperature for any evaporator weight esteem, yet this develop in its esteem turns out to be immediately when the estimation of the kettle weight increments. 3. The estimation of dryness portion increments with increment in warm stage temperature, however in the other hand this estimation of dryness part diminishes with increment the pressure proportion, and it is reductions moreover as kettle weight increments. 4. The warm productivity diminishes with increment in condenser weight; however this lessening in its esteem achieves the lower values when the evaporator weight is equivalent to (10 MPa). 5. The dryness portion ascending with increment in condenser weight at specific kettle weight esteem. Be that as it may, this ascending in its esteem gets.
|
2018-12-27T04:23:00.711Z
|
2017-04-01T00:00:00.000
|
{
"year": 2017,
"sha1": "f02237ebdd2a31e54e0c3751f9c06f417e204278",
"oa_license": null,
"oa_url": "https://doi.org/10.24001/ijaems.3.4.4",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "7a08ef977e6c58f46efdb728110a3926ad4ef310",
"s2fieldsofstudy": [
"Engineering",
"Environmental Science"
],
"extfieldsofstudy": [
"Environmental Science"
]
}
|
1675345
|
pes2o/s2orc
|
v3-fos-license
|
Macular Bruch´s Membrane Length and Axial Length. The Beijing Eye Study
Purpose To assess whether macular Bruch´s membrane gets lengthened in axial myopia. Methods Using the enhanced depth imaging mode of spectral-domain optical coherence tomography and examining a subgroup of participants of the population-based cross-sectional Beijing Eye Study, we measured the length of Bruch´s membrane (“MacBMLength”) from the fovea to the temporal edge of parapapillary gamma zone, and the distance between the fovea and the temporal optic disc border. Parapapillary gamma zone was defined as the parapapillary region without Bruch´s membrane. We additionally measured ocular biometric parameters and assessed non-ophthalmologic variables. Results Measurements of MacBMLength were performed on 322 individuals. MacBMLength (mean: 3.99±0.33 mm; range: 3.17–4.93 mm) was not significantly associated with any systemic parameter or ocular biometric parameter. Gamma zone width (mean: 0.18±0.30mm; range: 0.00–2.61mm) was associated (multivariate analysis; correlation coefficient r:0.80) with longer axial length (P<0.001; standardized correlation coefficient beta: 0.60; non-standardized correlation coefficient B:0.11; 95%CI: 0.09,0.14) and with longer fovea-optic disc border distance (P<0.001; beta:0.28; B:0.19; 95%CI:0.14,0.25), but not with MacBMLength (P = 0.42). Fovea-temporal disc border distance (mean: 4.16±0.44mm; range: 3.17–5.86mm) was associated (overall correlation coefficient: 0.68) with longer axial length (P<0.001; beta: 0.36; B: 0.10; 95%CI: 0.06, 0.13), after adjusting for flatter anterior chamber depth (P = 0.003; beta:-0.14; B:-0.14; 95%CI: -0.23,-0.05) and wider parapapillary gamma zone (P<0.001; beta:0.42; B:0.62; 95%CI:0.44,0.81). Conclusions In contrast to parapapillary gamma zone width and fovea-disc border distance, MacBMLength was not significantly associated with axial length. Axial elongation associated increase in fovea-disc distance may predominantly occur through development or elongation of parapapillary gamma zone, while macular Bruch´s membrane may mostly be independent of axial elongation.
Introduction
Axial myopization is associated with an elongation of the globe diameters, more in the sagittal direction than in the horizontal direction or the vertical direction. It leads to a change in the shape of the eye, from a mostly spherical shape in the case of emmetropic globes to an axially elongated shape in highly myopic eyes. Based upon measurements of choroidal and scleral thickness, histomorphometric studies have revealed that the axial elongation associated thinning of the two outer ocular layers is found mostly posterior to the equator and that it is more marked the closer to the posterior pole [1][2][3][4]. If the axial elongation associated changes occur predominantly at the posterior pole, one may infer that the size of the macula increases. If the macular region enlarges postnatally and assuming that the retinal photoreceptors do not multiply after birth or after the end of the second year of life, one may conclude that the density of the retinal photoreceptors in the macular region decreases and that the inter-photoreceptor distance subsequently increases with longer axial length. This would imply a decreased visual acuity in axially myopic eyes versus in emmetropic eyes. It would be contradictory to the results of previous population-based studies in which, after exclusion of highly myopic eyes, best corrected visual acuity was not significantly associated with axial length in a multivariate analysis [5]. Since the retinal photoreceptors rest on the retinal pigment epithelium which forms Bruch s membrane as its basal membrane, the density of the foveal photoreceptors will depend on a potential lengthening of macular Bruch´s membrane in axially myopic eyes.
In particular in some myopic eyes, Bruch´s membrane does not reach the temporal optic disc border but leaves a parapapillary temporal region free of Bruch´s membrane [6][7][8][9][10][11][12][13][14][15][16]. That region has recently been called "parapapillary gamma zone" [14][15][16]. We therefore conducted the current study to investigate the association between the distance between the fovea and the end of Bruch´s membrane in direction to the optic disc (macular Bruch´s membrane length or "MacBMLength"), parapapillary gamma zone, the fovea-disc border distance, and axial length. The width of parapapillary gamma zone was calculated as difference of fovea-disc border distance minus MacBMLength.
Methods
The Beijing Eye Study 2011 is a population-based investigation performed in a rural and an urban region of Beijing. The study protocol was approved by the Medical Ethics Committee of the Beijing Tongren Hospital and informed written consent was obtained from all study participants. Out of 4403 eligible individuals fulfilling the inclusion criterion of an age of 50 + years, 3468 (78.8%) individuals participated. The mean age was 64.6 ± 9.8 years (median, 64 years; range, 50-93 years). The study has been described in detail previously [17,18]. Exclusion criterion for the present study was the presence of macular defects in Bruch´s membrane as described in detail previously [19,20]. Additional inclusion criterion for the present investigation was the availability of measurements of the MacBMLength and of the fovea-temporal optic disc border distance. All participants underwent an interview with a structured questionnaire, biochemical blood examinations, measurement of blood pressure and body height and weight, and a comprehensive ophthalmic examination. We measured visual acuity and performed a slit lamp examination of the anterior and posterior segment of the eye and a digital photography of the cornea, lens, macula and optic disc (fundus camera Type CR6-45NM; Canon Inc., Tokyo, Japan). Spectral domain optical coherence tomography (SD-OCT, Spectralis, Heidelberg Engineering Co., Heidelberg, Germany) with enhanced depth imaging modality was carried out after pupil dilation [21]. Thirty-one OCT sections were obtained, which covered a 30°x 30°large rectangle centered around the fovea. Using the enhanced depth imaging mode, seven additional sections (each comprising 100 averaged scans) were obtained in a 5°x 30°large rectangle centered onto the fovea. Using the scan running through the fovea (which was identified as the point of the maximal foveal depression), we measured the length of Bruch´s membrane from the fovea to the end of Bruch´s membrane in direction of the optic nerve head. Using the fundus photographs, we measured the distance between the fovea and the temporal border of the optic nerve head. In the parapapillary region, we differentiated between an alpha zone characterized by the presence of Bruch´s membrane with irregular retinal pigment epithelium, beta zone characterized by the presence of Bruch´s membrane and absence of retinal pigment epithelium, and gamma zone characterized by the absence of Bruch´s membrane (Figs 1 and 2). We corrected the image magnification caused by the optic media of the eye and by the fundus camera using the Littmann-Bennett method [22][23][24]. Statistical analysis was performed using a commercially available statistical software package (SPSS for Windows, version 22.0, IBM-SPSS, Chicago, IL, USA). The study population was divided in subgroups by axial length, and within each subgroup, we randomly selected participants to be included into the present study. We examined the distribution of the MacBMLength and of the fovea-disc border distance using the Kolmogorov-Smirnov test and determined the mean ± standard deviations of the parameters. We calculated the length of the parapapillary region without Bruch´s membrane (gamma zone) as the difference of the foveadisc border distance minus the MacBMLength. We then performed a univariate analysis of the associations between the MacBMLength, the fovea-disc border distance or the width of gamma zone with other ocular and systemic variables. Finally, we conducted a multivariate regression analysis, with the MacBMLength, the fovea-disc border distance or gamma zone width as the dependent variable and all those parameters as independent variables which were significantly associated with dependent parameter in the univariate analysis. From the list of independent parameters we then dropped step by step those parameters which were no longer significantly associated. 95% confidence intervals (CI) were presented. All P-values were two-sided and were considered statistically significant if the values were smaller than 0.01.
Results
Out of the 3468 subjects, measurements of MacBMLength and of the fovea-temporal optic disc border distance were performed for 322 individuals (173 women). Mean age was 62.9 ± 8.8 years (median: 61 years; range: 50-90 years), mean axial length was 24.2 ± 1.7 mm (median: 23.9 mm; range: 21.2 to 30.9 mm), and mean refractive error was -3.08 ± 3.06 diopters (median: -2.50 diopters; range: -20.00 to +2.00 diopters). Due to the selection of the study participants, the group of subjects with measurements of MacBMLength and of the fovea-temporal disc border distance as compared with the group of individuals without these measurements was significantly (P<0.001) younger (62.8 ± 8.8 years versus 64.8 ± 9.9 years), had a significantly (P<0.001) longer axial length (24.3 ± 1.7 mm versus 23.1 ± 1.0 mm) and was significantly (P<0.001) more myopic (-3.08 ± 3.04 diopters versus 0.09 ± 1.72 diopters), while both groups did not differ significantly in gender (P = 0.34).
Discussion
MacBMLength was not significantly associated with any systemic parameter or ocular biometric parameter except for, if at all, a weak relationship with longer axial length (P = 0.02; r 2 : 0.02) (Fig 3). The value of the correlation coefficient indicated that 2% of the variation in MacBMLength could be explained by a variation in axial length. Parapapillary gamma zone width was strongly associated with longer axial length (P<0.001; r:0.60) but not with MacBM-Length (P = 0.42). The fovea-temporal disc border distance was strongly associated with longer axial length (P<0.001; beta: 0.36) after adjusting flatter anterior chamber depth (P = 0.003; beta: -0.14) and wider parapapillary gamma zone (P<0.001; beta: 0.42). The results suggest that the axial elongation associated increase in the fovea-disc border distance predominantly occurred through a development or elongation of parapapillary gamma zone, while macular Bruch´s membrane was mostly independent of axial elongation. One may infer that in axial elongation with a mostly constant macular Bruch´s membrane length, the density of retinal photoreceptors in the fovea remained unchanged.
The result of our study on an increase in the fovea-optic disc distance agrees with previous investigations [9,25]. It may be explained by the finding that myopic axial elongation leads to changes which are more marked the closer to the posterior pole [1][2][3][4]. Histomorphometric investigations revealed that changes associated with primary high axial myopia are markedly less detectable anterior to the equator [1][2][3][4]. It is in contrast to eyes with secondary high axial myopia due to congenital glaucoma, in which scleral thinning occurs posterior and anterior to the equator and also includes a widening and thinning of the cornea.
The width of gamma zone as assessed in our study (0.18 ± 0.30 mm) roughly corresponded to the mean distance from the temporal disc margin to the edge of Bruch´s membrane as measured in the study by Park and colleagues (0.21 ± 0.20 mm) [6].
The finding of our investigation on an axial length dependent increase in the width of parapapillary gamma zone is in agreement with previous histomorphometric studies and clinical studies [8,9,[14][15][16]. In these studies, one had differentiated between alpha zone, characterized by the presence of Bruch´s membrane and presence of an irregularly structured retinal pigment epithelium; beta zone defined by the presence of Bruch´s membrane and absence of retinal pigment epithelium and which was located between the peripheral alpha zone and the optic disc border, or between alpha zone and gamma zone if gamma zone was present; and gamma zone, defined by the absence of Bruch´s membrane and located directly at the peripapillary ring which forms the border optic nerve head [14,26]. In the preceding studies as well as in the present investigation, gamma zone strongly increased with longer axial length, while it was not, or only marginally, associated with glaucoma. In contrast, parapapillary beta zone was significantly associated with glaucoma, while it was not, or not markedly, associated with axial length [14][15][16]. Also in our study, gamma zone increased with longer axial length, with the association starting mostly beyond an axial length of 25 mm (Fig 4). These results agree with findings obtained in other previous investigations. On the images of 72 healthy adults, Chui and colleagues measured the fovea-temporal optic disc border distance to be 4.22 mm ± 0.46 and the distance between the foveola and the temporal edge of the peripapillary crescent to be 3.97 mm ± 0.25 mm [9]. The difference between both measurements (approximately corresponding to parapapillary gamma zone in pour study) was similar to the width of gamma zone as determined in our study. In Chui´s study (as in our study), only the fovea-temporal optic disc border distance was significantly correlated with axial length. As axial length increased by 10% in Chui´s investigation, the fovea-temporal optic disc border distance increased by 13%, while the outer neural retina only expanded by 4%. Chui and colleagues concluded that retinal stretching might not mirror scleral growth. Nonaka and coworkers examined 61 highly myopic ( -6 diopters) eyes without myopic retinopathy [8]. They found that the distance from the fovea nasal margin of a socalled PPA-ß zone (presumable gamma zone) correlated with axial length (P<0.05; correlation coefficient r: 0.26), and that the width of the PPA-ß zone correlated with axial length (P<0.05; r: 0.32).
The parapapillary gamma zone as it was called in our study has previously been addressed in previous investigations [6][7][8][9][10][11][12][13][14][15][16][27][28][29][30][31]. Applying Fourier-domain optical coherence tomography, Park and coworkers found that within a clinical parapapillary beta zone atrophy the end of Bruch´s membrane did not reach the border of the optic nerve head in all eyes [6]. This zone without Bruch´s membrane would be the equivalent of gamma zone in our study. Hayashi and associates reported that a parapapillary beta zone as defined by these authors either included a Bruch's membrane which was straight or downward-curved or which showed a downward-bending slope without Bruch´s membrane [29]. This latter region without Bruch s membrane was called "gamma" zone in our investigation. Lee and coworkers reported similar observations [7].
In contrast to the width of parapapillary gamma zone and to the total fovea-disc border distance, the length of macular Bruch´s membrane, defined as the distance between the fovea and the temporal edge of parapapillary gamma zone was mostly independent of axial length (Fig 3). It suggests that the axial elongation associated increase in the fovea-disc border distance led to the appearance or enlargement of a papillary gamma zone. This finding makes one infer that the distance between the retinal photoreceptors in the macular and foveal region was not markedly dependent on axial length as long as highly myopic eyes with secondary macular Bruch´s membrane defects were excluded [15,18]. Correspondingly, a recent multivariate analysis revealed that within non-highly myopic eyes (i.e. eyes with an axial length of less than 26 mm), better best corrected visual acuity was significantly associated with thicker subfoveal choroid (P<0.001) in general and a subfoveal choroid thicker than 30 μm (P<0.001) in particular, while it was not significantly with axial length, after adjusting for younger age (P<0.001), higher level of education (P<0.001), taller body stature (P<0.001), higher body mass index (P = 0.005), and absence of major ocular diseases such as glaucoma [5]. That finding also agrees with the results of the study by Li and associates who used an adaptive optics scanning laser ophthalmoscopy under optimized wavefront correction and measured the cone photoreceptor diameter in 18 healthy eyes with axial lengths ranging between 22.86 to 28.31 mm [32]. Ocular biometry and an eye model were used to estimate the retinal magnification factor. They found a significant decrease in cone density (P<0.05) with increasing axial length at an eccentricity of 0.30 or more mm, but not closer to the fovea. These results were contradicted the findings obtained in the investigation by Kitaguchi and colleagues who applied an adaptive optics fundus camera and examined 19 healthy individuals with a mean axial length ranging between 23.4 and 28.0 mm. They found that the average cone spacing in the moderate-to high-myopia group (4.71 ± 0.44 μm) was significantly greater (P = 0.002) than in the normal and low-myopia groups (3.90 ± 0.47 μm). The cone spacing was significantly correlated with the axial length (r: 0.77; P<0.001) [33]. Also, Chui reported that cone photoreceptor packing density (cells per square millimeter) was significantly lower in myopic eyes than in emmetropic eyes [34]. At a given location, there was considerable individual variation in cone photoreceptor packing density, although more than 20% of the variance could be accounted for by differences in axial length.
The finding that parapapillary gamma zone width was, but that macular Bruch´s membrane length was not, significantly associated with longer axial length showed that the axial elongation associated increase in the fovea-disc border distance was predominantly due to the development or widening of parapapillary gamma zone. As shown in Fig 5, the steepness of the regression line of the association between the fovea-disc border distance and the width of parapapillary gamma zone was 0.95, indicating that 95% of the increase in fovea-disc border distance was due to the increase in the gamma zone width. This can be regarded as the widening of the physiological defect in Bruch´s membrane. This primary Bruch´s membrane defect forms the inner layer of the optic nerve head. The latter can be considered to be a three-layered hole, with Bruch´s membrane hole or opening as the inner layer, the choroidal hole or defect as the middle layer, and the scleral hole as the outer hole. One may speculate that in the early years of life, all three holes are aligned to each other. If axial elongation occurs, the papillary Bruch´s membrane opening may move in direction to the foveal center while the scleral hole may stay behind, leading to an overhanging of Bruch´s membrane on the nasal disc side and a corresponding lack of Bruch´s membrane on the temporal side of the optic nerve head [35]. The lack of Bruch´s membrane on the temporal side of the optic nerve head is equivalent to the development of parapapillary gamma zone. The peripapillary ring is the continuation of the optic nerve pia mater and marks the peripheral border of the lamina cribrosa within the optic nerve head [26]. In agreement with the finding that the length of macular Bruch´s membrane did not enlarge with longer axial length agrees with a recent histomorphometric study that the thickness of Bruch´s membrane, in contrast to a scleral and choroidal thickness, did not decrease with longer axial length [36]. It suggests that the myopic axial elongation was not associated with a stretching and lengthening of Bruch´s membrane in the macular region, at least not in the region extending from the fovea into nasal direction. The question then arises, why some eye with high axial myopia can develop secondary Bruch´s membrane defects in the macular regions as described recently [15,20].
The constancy of macular Bruch´s membrane length and the increase in the fovea-disc border distance by an enlargement of parapapillary gamma zone may imply that Bruch´s membrane is not strongly fixed on the underlying sclera through the choroid. Such a theory of a sliding Bruch´s membrane has also been discussed in a previous study in which the parapapillary region, presumably gamma zone, showed a marked decrease in size after a profound reduction in intraocular pressure for several months had occurred [37]. The concept of a Bruch´s membrane sliding by the swamping choroid on the sclera does not contradict a firm and stable relationship between Bruch´s membrane, the retinal pigment epithelium and the photoreceptors as basis for a stable visual function. The concept of a sliding tissue layer has also been suggested previously by Chui and associates, who discussed that the existence of a difference between the photoreceptor margin and retinal pigment epithelium margin in some eyes may suggest that a tissue slippage occurs within the retina during eye growth [9]. The difference between both models is that Chui suggested a slippage within the retina, while in the present study, a slippage was postulated to occur between Bruch´s membrane and the sclera The mean value of the fovea-disc border distance of 4.16 ± 0.44 mm as found in our study correlated with the findings obtained in previous investigations, if one takes into account that most of the previous studies measured the distance from the fovea to the center of the optic disc. In 51 preterm and full-term infants, mean optic disc-fovea distance was 4.4 ± 0.4 mm, without difference between the preterm infants and the full-term infants [38]. In 183 diabetic patients without retinopathy, van de Put and coworkers found a mean disc-fovea distance of 4.72 ± 0.27 mm, and a longer disc-fovea distance was associated with more myopic refractive error [25]. The disc-fovea distance in 27 prematurely born children at an age of 10-11 years was 4.74 ± 0.29 mm, as reported by Knaapi and coworkers [39]. In contrast to our study, Knaapi and associates did not find significant associations between the disc-fovea distance and refractive error (or axial length), although one child with axially high myopia had an aboveaverage disc-fovea distance of 6.35 mm. Lee and colleagues measured the distance between the temporal optic disc margin to the fovea distance in 88 patients with normal-tension glaucoma [40]. They found that in patients with a sparing of the central field the disc margin-fovea distance (3.88 ± 0.28 mm) was longer (P = 0.002) than in the group with glaucomatous central visual field loss (3.64 ± 0.40 mm). Interestingly, parapapillary atrophy (beta zone) was wider (P = 0.03) in the group with sparing of the central visual field and longer disc margin-fovea distance. It fits with our observation, that a longer disc-fovea distance was significantly associated with larger parapapillary gamma zone.
We had excluded eyes with secondary macular Bruch´s membrane defects in our study, since the irregular arrangement of the parts of Bruch´s membrane made it impossible to reliably measure the length of Bruch´s membrane. Future studies may address whether in these eyes the distance between the optic disc and the foveal region is enlarged, then pointing to a potential association between this type of secondary Bruch´s membrane defects and scleral staphylomata at the posterior pole.
Our results should be interpreted with some limitations in mind. First, our study had a lower age limit of 50 years so that the findings of our study cannot directly be transferred on younger individuals. Second, our study included only a randomly selected group of individuals within each subgroup of axial length. The mean values of the MacBMLength and the width of parapapillary gamma zone are therefore not normative data of a population-based study population. Third, we assumed that the length of macular Bruch's membrane ("MacBMLength"), measured as distance between the fovea and the end of Bruch´s membrane in direction to the optic disc, was a surrogate for the photoreceptor density in the macular region. That assumption however, has not been proven. Fourth, the macula is a three-dimensional concave structure, centered on the fovea. In our analysis, only a single linear measurement from the fovea to the optic disc was taken into account. The interpretation to commenting on macular photoreceptor density has therefore to be taken with caution, as our study did not look at changes temporal, superior, and inferior to fovea center. In particular in eyes with posterior staphyloma, this point may be relevant. Fifth, the present study included Chinese individuals. Since ocular dimensions may differ between ethnicities, the measurements obtained in our study population may not directly be transferred onto other populations. Sixth, it was assumed that the length of the macular Bruch's membrane was a surrogate for the density of retinal pigment epithelium cells and macular photoreceptors. That assumption however, may be too simplistic, since only the inner part of Bruch´s membrane is produced by the retinal pigment epithelium.
In conclusion, the length of the macular Bruch´s membrane was not significantly associated with axial length. In contrast, larger width of parapapillary gamma and longer total fovea-disc border distance were strongly correlated with longer axial length. It suggests that the axial elongation associated increase in the fovea-disc distance predominantly occurs through a development or elongation of parapapillary gamma zone, while the macular Bruch´s membrane is mostly independent of axial elongation. One may infer that in axial elongation with a mostly constant macular Bruch´s membrane length, the density of retinal photoreceptors in the fovea remains unchanged.
|
2018-04-03T03:16:45.168Z
|
2015-08-28T00:00:00.000
|
{
"year": 2015,
"sha1": "4ee511ea8f99f4304da6bbd2ede2d321d6431782",
"oa_license": "CCBY",
"oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0136833&type=printable",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "4ee511ea8f99f4304da6bbd2ede2d321d6431782",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine",
"Physics"
]
}
|
247934362
|
pes2o/s2orc
|
v3-fos-license
|
Nanoporous Cauliflower-like Pd-Loaded Functionalized Carbon Nanotubes as an Enzyme-Free Electrocatalyst for Glucose Sensing at Neutral pH: Mechanism Study
In this work, we propose a novel functionalized carbon nanotube (f-CNT) supporting nanoporous cauliflower-like Pd nanostructures (PdNS) as an enzyme-free interface for glucose electrooxidation reaction (GOR) in a neutral medium (pH 7.4). The novelty resides in preparing the PdNS/f-CNT biomimetic nanocatalyst using a cost-effective and straightforward method, which consists of drop-casting well-dispersed f-CNTs over the Screen-printed carbon electrode (SPCE) surface, followed by the electrodeposition of PdNS. Several parameters affecting the morphology, structure, and catalytic properties toward the GOR of the PdNS catalyst, such as the PdCl2 precursor concentration and electrodeposition conditions, were investigated during this work. The electrochemical behavior of the PdNS/f-CNT/SPCE toward GOR was investigated through Cyclic Voltammetry (CV), Linear Sweep Voltammetry (LSV), and amperometry. There was also a good correlation between the morphology, structure, and electrocatalytic activity of the PdNS electrocatalyst. Furthermore, the LSV response and potential-pH diagram for the palladium–water system have enabled the proposal for a mechanism of this GOR. The proposed mechanism would be beneficial, as the basis, to achieve the highest catalytic activity by selecting the suitable potential range. Under the optimal conditions, the PdNS/f-CNT/SPCE-based biomimetic sensor presented a wide linear range (1–41 mM) with a sensitivity of 9.3 µA cm−2 mM−1 and a detection limit of 95 µM (S/N = 3) toward glucose at a detection potential of +300 mV vs. a saturated calomel electrode. Furthermore, because of the fascinating features such as fast response, low cost, reusability, and poison-free characteristics, the as-proposed electrocatalyst could be of great interest in both detection systems (glucose sensors) and direct glucose fuel cells.
Introduction
Glucose electrooxidation reaction (GOR) has been extensively studied over the years for several applications such as in the development of glucose electrochemical sensors and biosensors for blood sugar monitoring [1][2][3] or the energy supply of bioelectronics-based devices by enzymatic and non-enzymatic glucose fuel cells [4][5][6]. GOR has also been involved in microbial fuel cell technology that converts glucose chemical energy to electrical energy [7]. As a reaction, GOR was first studied hundreds of years ago [8]; however, the last ten years have again attracted the scientific community's interest due to its potential application in the aforementioned fields [9]. These applications have led to increasing demand for the development of inexpensive electrocatalysts that have high sensitivity, high showing great potential and stability for use as a catalyst in glucose sensors [35]. These studies suggest that the shape and size of the Pd particles have a significant effect on electrocatalytic activity. In comparison with the chemical methods for catalyst preparation, electrodeposition is a flexible technique allowing the rapid coating of conductive surfaces with thin films of catalysts of different sizes (micro-and nano-structures) and forms (wires, rods, tubes, ribbons, etc.) [36][37][38][39]. For instance, Jiang et al. reported the electrodeposition of cauliflower-like palladium nanoparticles used as catalysts for the reductive dechlorination of 4-chlorophenol in the field of pollutants removal [40]. To the best of our knowledge, the electrocatalytic activity of electrodeposited Pd for glucose electrooxidation has not been investigated to date, although this preparation method provides the opportunity to control the shape and size of deposited catalysts through electrolysis parameters (applied potential, the concentration of the metallic precursors, temperature, pH, duration of the electrolysis, etc.) [41].
In this study, SPCE surfaces modified with carboxylic acid f-CNTs were used as carbonbased supporting materials for the electrodeposition of PdNS catalysts. The electrocatalytic activity of the deposited PdNS catalysts was investigated for the direct glucose oxidation in a neutral buffered solution. The effect of the metallic precursor concentration and electrodeposition potential on the morphology evolution of electrodeposited Pd nanostructures were investigated using scanning electron microscopy (SEM). The electrocatalytic activity of PdNS catalysts toward GOR was characterized by Cyclic Voltammetry (CV) and Linear Sweep Voltammetry (LSV). A correlation between the morphology of PdNS catalysts and their electrocatalytic activity was observed. In addition, a mechanism for GOR was proposed based on the LSV response of electrocatalysts and the potential-pH diagram for the palladium-water system. Amperometry measurements were also carried out to assess analytical features of electrodeposited PdNS as non-enzymatic catalysts for glucose sensing in a neutral environment.
Apparatus and Electrochemical Techniques
CV, LSV, and amperometric measurements were carried out using an electrochemical instrument, PalmSens BV (Houten, The Netherlands), connected to a computer and controlled by software named PSTrace5.8. A conventional three-electrode electrochemical system was used. The SPCE acted as a working electrode with a diameter of 3 mm. The reference electrode was a saturated calomel electrode (SCE) saturated with 3 M KCl, while a bare of stainless steel was used as a counter electrode. Electrode surface morphology was examined with scanning electron microscopy (SEM, JEOL model JSM-7401F, (JOEL, Croissy, France)).
SPCE Production
Screen-printed carbon electrodes (SPCEs) were home-produced with a 245 DEK (Weymouth, UK) screen-printing machine. Graphite-based ink (Electrodag 423 SS) from Acheson (Milan, Italy) was used for printing the working and counter electrodes. The substrate was a flexible polyester film (Autostat HT5) obtained from Autotype Italia (Milan, Italy). The homemade electrodes were produced in foils of 48 units. The diameter of the working electrode was 3 mm, resulting in a geometric area of 0.07 cm 2 . These electrodes were obtained as a gift from Prof. Fabiana Arduini at Tor Vergata University (Rome, Italy).
Preparation of f-CNT/SPCE
A dispersion of f-CNT was prepared by adding 10 mg of f-CNT powder into 10 mL of a dispersing agent (0.125% (v/v) Nafion in DMF) and sonicated for 3 h. A small volume (6 µL) of the dispersion was then drop-cast onto the working electrode surface of the SPCE in three steps of 2 µL. Afterward, the solvent was allowed to dry at 45 • C for 15 min, and an f-CNT "film" was left on the electrode surface (f-CNT/SPCE).
Electrodeposition of PdNS Catalyst
The Pd electrocatalyst was electrochemically deposited on the f-CNT/SPCE surface at a fixed deposition potential of −0.2 V vs. SCE for 300 s in a solution of 0.05 M HClO 4 and 0.25 M H 2 SO 4 containing the PdCl 2 precursor, which resulted in the PdNS/f-CNT/SPCE sensor. The same procedure was followed to prepare the PdNS catalyst onto the SPCE surface without being previously modified with f-CNTs, resulting in the PdNS/SPCE sensor. A general schematic illustration of the preparation steps of the enzyme-free electrocatalystbased glucose sensor is described in Figure 1.
SPCE Production
Screen-printed carbon electrodes (SPCEs) were home-produced with a 245 DEK (Weymouth, UK) screen-printing machine. Graphite-based ink (Electrodag 423 SS) from Acheson (Milan, Italy) was used for printing the working and counter electrodes. The substrate was a flexible polyester film (Autostat HT5) obtained from Autotype Italia (Milan, Italy). The homemade electrodes were produced in foils of 48 units. The diameter of the working electrode was 3 mm, resulting in a geometric area of 0.07 cm 2 . These electrodes were obtained as a gift from Prof. Fabiana Arduini at Tor Vergata University (Rome, Italy).
Preparation of f-CNT/SPCE
A dispersion of f-CNT was prepared by adding 10 mg of f-CNT powder into 10 mL of a dispersing agent (0.125% (v/v) Nafion in DMF) and sonicated for 3 h. A small volume (6 µL) of the dispersion was then drop-cast onto the working electrode surface of the SPCE in three steps of 2 µL. Afterward, the solvent was allowed to dry at 45 °C for 15 min, and an f-CNT "film" was left on the electrode surface (f-CNT/SPCE).
Electrodeposition of PdNS Catalyst
The Pd electrocatalyst was electrochemically deposited on the f-CNT/SPCE surface at a fixed deposition potential of −0.2 V vs. SCE for 300 s in a solution of 0.05 M HClO4 and 0.25 M H2SO4 containing the PdCl2 precursor, which resulted in the PdNS/f-CNT/SPCE sensor. The same procedure was followed to prepare the PdNS catalyst onto the SPCE surface without being previously modified with f-CNTs, resulting in the PdNS/SPCE sensor. A general schematic illustration of the preparation steps of the enzyme-free electrocatalyst-based glucose sensor is described in Figure 1.
Characterization of Electrocatalytic Activity for GOR
The electrocatalytic activity of electrodeposited catalysts was investigated by CV and LSV measurements at a well-defined potential window, with 8 mV as step potential and a scan rate of 10 mV s −1 in 0.1 M PBS at pH 7.4 containing an appropriate amount of glucose, as indicated in CV experiments and further electrochemical measurements. Chronoamperometry measurements were also carried out with a PdNS/f-CNT/SPCE sensor using amperometric batch analysis in 10 mL of a stirred (300 rpm) solution of 0.1 M PBS (pH 7.4) with a selected applied potential. When a stable baseline current was reached, the glucose was added successively, and the responses were recorded. All electrochemical measurements were performed at room temperature (25 °C) with three replicates.
Characterization of Electrocatalytic Activity for GOR
The electrocatalytic activity of electrodeposited catalysts was investigated by CV and LSV measurements at a well-defined potential window, with 8 mV as step potential and a scan rate of 10 mV s −1 in 0.1 M PBS at pH 7.4 containing an appropriate amount of glucose, as indicated in CV experiments and further electrochemical measurements. Chronoamperometry measurements were also carried out with a PdNS/f-CNT/SPCE sensor using amperometric batch analysis in 10 mL of a stirred (300 rpm) solution of 0.1 M PBS (pH 7.4) with a selected applied potential. When a stable baseline current was reached, the glucose was added successively, and the responses were recorded. All electrochemical measurements were performed at room temperature (25 • C) with three replicates. Figure S1A). The CV characterization/polarization of electrodeposited PdNS in 0.5 M H 2 SO 4 solution showed, for both electrodes, the typical electrochemical response of Pd surfaces. Indeed, CV curves exhibited anodic oxidation and cathodic reduction peaks corresponding respectively to Pd oxide formation during anodic scanning and a reduction in oxidized Pd in the reverse scan ( Figure S1B). It was observed that the oxidation and reduction peak currents increased with the presence of f-CNT on the electrode surface. This is due to a higher electroactive surface area of f-CNT/SPCE compared to the same geometric surface of bare SPCE. These results were consistent with previously reported studies [34,42].
Results and Discussion
The electrocatalytic behavior of PdNS-modified electrodes toward GOR was investigated using CV in 0.1 M PBS (pH 7.4), both in the absence ( Figure 2A) and in the presence of 20 mM glucose ( Figure 2B). In the glucose-free solution, an intense cathodic reduction peak was observed for both electrodes at around 0.0 V vs. SCE attributed to the reduction of palladium oxide species. The addition of glucose (20 mM) showed the anodic current peak of glucose oxidation at around 160 mV vs. SCE. This oxidative peak (*) appeared during the cathodic scan of both SPCE-and f-CNT/SPCE-based PdNS. This peak might be attributable to the oxidation of dehydrogenated D-glucose (previously adsorbed via dehydrogenation) or the oxidation of products adsorbed on the Pd catalyst during the direct scan (anodic scan) [27,43]. The current intensity of the anodic peak obtained with PdNS/f-CNT/SPCE was more than twice as high as that obtained with PdNS/f-SPCE, indicating that electrocatalytic activity of the PdNS catalyst toward GOR is greatly enhanced by the insertion of f-CNTs that synergize with the PdNS catalyst. Therefore, these f-CNTs were chosen for coating SPCE in the rest of this study.
Effect of f-CNTs
The electrodeposition of PdNS on bare SPCE and f-CNT/SPCE was performed at −0.2 V vs. SCE for 1000 s in a solution of 0.05 M HClO4 and 0.25 M H2SO4 containing 1 mM Pd 2+ ( Figure S1A). The CV characterization/polarization of electrodeposited PdNS in 0.5 M H2SO4 solution showed, for both electrodes, the typical electrochemical response of Pd surfaces. Indeed, CV curves exhibited anodic oxidation and cathodic reduction peaks corresponding respectively to Pd oxide formation during anodic scanning and a reduction in oxidized Pd in the reverse scan ( Figure S1B). It was observed that the oxidation and reduction peak currents increased with the presence of f-CNT on the electrode surface. This is due to a higher electroactive surface area of f-CNT/SPCE compared to the same geometric surface of bare SPCE. These results were consistent with previously reported studies [34,42].
The electrocatalytic behavior of PdNS-modified electrodes toward GOR was investigated using CV in 0.1 M PBS (pH 7.4), both in the absence ( Figure 2A) and in the presence of 20 mM glucose ( Figure 2B). In the glucose-free solution, an intense cathodic reduction peak was observed for both electrodes at around 0.0 V vs. SCE attributed to the reduction of palladium oxide species. The addition of glucose (20 mM) showed the anodic current peak of glucose oxidation at around 160 mV vs. SCE. This oxidative peak (*) appeared during the cathodic scan of both SPCE-and f-CNT/SPCE-based PdNS. This peak might be attributable to the oxidation of dehydrogenated D-glucose (previously adsorbed via dehydrogenation) or the oxidation of products adsorbed on the Pd catalyst during the direct scan (anodic scan) [27,43]. The current intensity of the anodic peak obtained with PdNS/f-CNT/SPCE was more than twice as high as that obtained with PdNS/f-SPCE, indicating that electrocatalytic activity of the PdNS catalyst toward GOR is greatly enhanced by the insertion of f-CNTs that synergize with the PdNS catalyst. Therefore, these f-CNTs were chosen for coating SPCE in the rest of this study.
Effect of Electrodeposition Potential
The applied potential is an important factor in controlling the morphology and size of metallic electrodeposited particles [36]. Therefore, the electrodeposition of the PdNS
Effect of Electrodeposition Potential
The applied potential is an important factor in controlling the morphology and size of metallic electrodeposited particles [36]. Therefore, the electrodeposition of the PdNS catalyst on the f-CNT/SPCE surface was performed under different applied potentials Figure 3 (see also all voltammograms superimposed in Figure S2), decreasing the deposition potential to negative values, from +0.4 V to −0.2 V vs. SCE, contributed to an enhancement of electrocatalytic activity. The morphology of PdNS electrodeposited under different applied potentials was investigated by scanning electron microscopy. As shown in Figure 3, under +0.4 V vs. SCE, the electrodeposited Pd particles consisted of a smooth surface like a bulk Pd electrode. Terraces formed as massive microstructures condensed and agglomerated (Figure 3c). Furthermore, the morphology of the Pd changed slightly when the applied potential was +0.0 V, and the flower-like structures became denser and thicker with almost zero porosity, as illustrated in Figure 3b. However, decreasing the applied potential to −0.2 V resulted in the complete disappearance of the massive terraces, while a 3D cauliflower-like PdNS appeared on the electrode surface, as shown in Figure 3a. The cauliflower-like morphology of electrochemically synthesized Pd nanostructures is similar to that reported by Sun et al. for dichlorination catalysis [40]. These results indicated that deposition potential has a significant effect on the shape of Pd nanostructures, which subsequently improved the electrocatalytic activity toward GOR in a neutral buffered medium.
Effect of PdCl2 Precursor Concentration
The PdCl2 precursor solution concentration has a significant influence on the catalytic activities of metallic electrocatalysts [36]. Therefore, PdNS were electrodeposited at various PdCl2 precursor concentrations. Modified nanostructured sensors (PdNS/f-CNT) were then investigated for glucose electrooxidation.
As indicated in Figure 4A, SEM characterization was used to investigate the effect of metallic precursor concentration on the morphology evolution of electrodeposited Pd nanostructures. Indeed, the morphology of nanostructured Pd changed considerably with and reduction of PdO. This method was previously reported for evaluating the RSA of Pd electrodes and was based on the assumption that the charge density required to reduce a monolayer of PdO per unit of surface area was 0.424 mC cm −2 [44][45][46]. By following the procedure described in the Supplementary Materials (Section S1.1), the RSA was calculated for PdNS obtained at different applied potentials. The RSA of PdNS electrodeposited at −0.2 V vs. SCE (7.39 cm 2 ) was higher than those at 0 V vs. SCE (4.66 cm 2 ) and +0.4 V vs. SCE (0.52 cm 2 ) (Figure 3). These values of RSA are consistent with the morphology evolution observed in the SEM images, as well as the electrocatalytic activity recorded by the LSV technique toward GOR. Therefore, −0.2 V vs. SCE was considered an optimal potential for the electrodeposition of PdNS on f-CNT/SPCE to enhance their catalytic activity by increasing the RSA in heterogeneous catalysis. This applied potential value was chosen for coating SPCE in the rest of this study.
Effect of PdCl 2 Precursor Concentration
The PdCl 2 precursor solution concentration has a significant influence on the catalytic activities of metallic electrocatalysts [36]. Therefore, PdNS were electrodeposited at various PdCl 2 precursor concentrations. Modified nanostructured sensors (PdNS/f-CNT) were then investigated for glucose electrooxidation.
As indicated in Figure 4A, SEM characterization was used to investigate the effect of metallic precursor concentration on the morphology evolution of electrodeposited Pd nanostructures. Indeed, the morphology of nanostructured Pd changed considerably with the precursor concentration during electrodeposition. Increasing the precursor concentration, the evolution of the Pd structure changed from microclusters to 3D cauliflower-like microstructured porous Pd. When the concentration of Pd 2+ was 10 mM (Figure 4(a)), the morphology of Pd was cluster-shaped microstructures until they became bigger and bigger when 20 mM (Figure 4(b)) to 50 mM (Figure 4(c)) Pd 2+ were used during the electrodeposition process. Up to 100 mM Pd 2+ concentration of the precursor resulted in dense and large cauliflower-shaped Pd with a large electrocatalytic surface area on the f-CNT/SPCE surface, as shown in Figure 4d. These results demonstrate that the precursor solution concentration has a crucial effect on controlling the morphologies of Pd structures and supplying more catalytic sites to benefit the electrocatalytic oxidation of glucose molecules.
As shown in Figure 4B-E, the electrocatalytic activity of PdNS/f-CNT/SPCE toward GOR was investigated in 0.1 M PBS (pH 7.4) with (green line) and without 20 mM glucose (black line) using LSV measurements at a scan rate of 10 mV s −1 . According to these results, the higher the concentration of the precursor, the more the enhanced electrocatalytic activity of PdNS/f-CNT/SPCE toward GOR will be. In addition, a shift of the glucose oxidation peak toward positive values of potentials from 165 to 300 mV vs. SCE was observed with the increase in precursor concentration (10 to 100 mM Pd 2+ ).
In addition, amperometric measurements on PdNS/f-CNT/SPCE were performed at +0.3 V vs. SCE. The results obtained are consistent with those obtained by LSV. This method enables a choice according to the best results in terms of detection limit and sensitivity. Figure 4F shows amperometric responses obtained from PdNS/f-CNT/SPCE at an applied potential of +0.3 V vs. SCE after a successive glucose addition in the range of 1-20 mM.
The sensitivity and detection limit (DL) were significantly improved when increasing the precursor concentration. Table 1 summarizes the analytical parameters provided by amperometry. On the basis of these results, a 100 mM precursor concentration was chosen as the optimal concentration, providing 3D porous cauliflower-like Pd nanostructures with high sensitivity and a low DL. tion peak toward positive values of potentials from 165 to 300 mV vs. SCE was observed with the increase in precursor concentration (10 to 100 mM Pd 2+ ).
In addition, amperometric measurements on PdNS/f-CNT/SPCE were performed at +0.3 V vs. SCE. The results obtained are consistent with those obtained by LSV. This method enables a choice according to the best results in terms of detection limit and sensitivity. Figure 4F shows amperometric responses obtained from PdNS/f-CNT/SPCE at an applied potential of +0.3 V vs. SCE after a successive glucose addition in the range of 1-20 mM. The sensitivity and detection limit (DL) were significantly improved when increasing the precursor concentration. Table 1 summarizes the analytical parameters provided by amperometry. On the basis of these results, a 100 mM precursor concentration was chosen as the optimal concentration, providing 3D porous cauliflower-like Pd nanostructures with high sensitivity and a low DL. Table 1. Summary of the analytical parameters of the developed non-enzymatic electrochemical glucose sensors for glucose sensing in a neutral environment (PBS, pH 7.4).
Mechanism Proposed for Glucose Oxidation on PdNS/f-CNT Biomimetic Nanocatalyst in Neutral pH
It should be noted that in all LSV voltammograms (black curves in Figure 4), two peaks, O1 and O2, appeared in the glucose-free background PBS solution. As observed, the current intensity of peak O1 increases with the concentration of the precursor up to 50 mM. This peak was only slight in the structure of the electrocatalyst obtained with 100 mM. To better understand the mechanism underlying these peaks, the Pourbaix diagram (E-pH) of the Pd-water system was used ( Figure S3). The theoretical potential values were calculated by replacing the pH with a 7.4 value in all linear equations (E-pH) of the Pourbaix diagram [47], as described in detail in the Supplementary Materials (Section S1.2). Therefore, the obtained values were compared with those corresponding to the surface oxidation peaks (O1 and O2) of the Pd electrocatalyst in the glucose-free background solution.
Based on the potential values representing the O1 and O2 peaks in the LSVs (Figure 4, black curve), it can be assumed that these two peaks correspond to oxidation equations, Equations (1) and (2), of the catalyst surface in the background solution. However, the O3 peak corresponds to Equation (3) Pd catalysts are well known for their ability to adsorb and absorb H, forming the compact and stable surfaces of Pd 2 H hydrides. Therefore, the peak (O1) that appeared in the potential range of −0.6 to −0.3 V vs. SCE (hydrogen region) can be attributed to the hydrogen desorption from the Pd catalyst, while the oxidation peak O2 (oxygen region) can be due to the electrooxidation of dehydrogenated Pd (Scheme 1A).
Equations (1) and (2), of the catalyst surface in the background solution. However, the O3 peak corresponds to Equation (3), which represents the balanced equation of glucose oxidation at the Pd catalyst, involving glucose molecules and PdO oxidizing species.
Pd catalysts are well known for their ability to adsorb and absorb H, forming the compact and stable surfaces of Pd2H hydrides. Therefore, the peak (O1) that appeared in the potential range of −0.6 to −0.3 V vs. SCE (hydrogen region) can be attributed to the hydrogen desorption from the Pd catalyst, while the oxidation peak O2 (oxygen region) can be due to the electrooxidation of dehydrogenated Pd (Scheme 1A).
In addition, the evolution of hydrogen was significantly remarked on in the catalyst structures from Figure 4(a-c). Indeed, the oxidation peak (O1) of Pd2H in the background solution increases with a precursor concentration up to 50 mM, while this peak is very weak at 100 mM, indicating that this catalytic structure (Figure 4(d)) has low proton adsorption. H2 interaction with Pd-based catalysts depends on the structure of the materials by influencing adsorbate binding energies [48]. The electrodeposition of PdNS catalysts with a precursor concentration of 100 mM probably induces structural changes. For example, Sakamoto et al. demonstrated that two phases of adsorbed hydrogen can exist in Pd-based alloys with elements that do not adsorb hydrogen under normal conditions [49,50]. In addition, the evolution of hydrogen was significantly remarked on in the catalyst structures from Figure 4(a-c). Indeed, the oxidation peak (O1) of Pd 2 H in the background solution increases with a precursor concentration up to 50 mM, while this peak is very weak at 100 mM, indicating that this catalytic structure (Figure 4(d)) has low proton adsorption. H 2 interaction with Pd-based catalysts depends on the structure of the materials by influencing adsorbate binding energies [48]. The electrodeposition of PdNS catalysts with a precursor concentration of 100 mM probably induces structural changes. For example, Sakamoto et al. demonstrated that two phases of adsorbed hydrogen can exist in Pd-based alloys with elements that do not adsorb hydrogen under normal conditions [49,50].
After adding 20 mM glucose to the supporting electrolyte, both the O1 and O2 peaks increased (Figure 4). For the O1 peak, this may be due to both the desorption of H (H coming from the medium) from the Pd catalyst and the dehydrogenation of glucose molecules via their adsorption onto the Pd catalyst [27]. This result is consistent with Pletcher's theory for concentric adsorption, which involves hydrogen extraction followed by the simultaneous adsorption of the organic species on the metallic catalyst, as shown in Scheme 1B [51]. At 100 mM Pd 2+ , the O1 peak is dramatically increased, which may be due to the strong dehydrogenation of glucose molecules by the catalyst, leading to the formation of Pd 2 H on this cauliflower-like structure. These results indicated that the PdNS structure obtained with 100 mM of the precursor was not favorable to hydrogen adsorption in neutral conditions, while it was supportive of glucose dehydrogenation. The significant enhancement of the oxidation peak of dehydrogenated Pd (O3) is probably due to the electrocatalytic oxidation of glucose molecules already adsorbed and/or accumulated on the surface of the Pd catalyst (Scheme 1B). Furthermore, the electrocatalyst response of PdNS showed satisfactory results in terms of stability in GOR, demonstrated by their regeneration after each scan. Assuming that glucose molecules can be oxidized to gluconolactone in neutral conditions by a twoelectron electrochemical reaction [27,52] and that PdO may act as an oxidizing species, a possible electrocatalytic mechanism on the PdNS/f-CNT/SPCE surface toward GOR was proposed, as indicated in Figure 5.
due to the strong dehydrogenation of glucose molecules by the catalyst, leading to the formation of Pd2H on this cauliflower-like structure. These results indicated that the PdNS structure obtained with 100 mM of the precursor was not favorable to hydrogen adsorption in neutral conditions, while it was supportive of glucose dehydrogenation. The significant enhancement of the oxidation peak of dehydrogenated Pd (O3) is probably due to the electrocatalytic oxidation of glucose molecules already adsorbed and/or accumulated on the surface of the Pd catalyst (Scheme 1B). Furthermore, the electrocatalyst response of PdNS showed satisfactory results in terms of stability in GOR, demonstrated by their regeneration after each scan. Assuming that glucose molecules can be oxidized to gluconolactone in neutral conditions by a two-electron electrochemical reaction [27,52] and that PdO may act as an oxidizing species, a possible electrocatalytic mechanism on the PdNS/f-CNT/SPCE surface toward GOR was proposed, as indicated in Figure 5. In the proposed mechanism of glucose oxidation on the PdNS catalyst in neutral media, the peak O2 corresponds to oxide formation (PdO) generated by the Pd from the hydrogen desorption reaction (O1). In addition, the peak O3 is attributed to glucose oxidation, which is induced with high-current intensity, as the current intensity of O1 increases.
Performance Comparison of PdNS/f-CNT as Non-Enzymatic Glucose Sensors in Neutral pH
Under the optimal conditions, the PdNS/f-CNT/SPCE sensor at the fixed potential of +0.3 V vs. SCE was successfully used for amperometric glucose detection at neutral pH. The response current of the non-enzymatic sensor showed a wide linear regression against a glucose concentration in the range of 1-41 mM with a DL of 95 µM (S/N = 3) and sensitivity of 9.26 µA mM −1 cm −2 . Importantly, this range covers the normal human blood In the proposed mechanism of glucose oxidation on the PdNS catalyst in neutral media, the peak O2 corresponds to oxide formation (PdO) generated by the Pd from the hydrogen desorption reaction (O1). In addition, the peak O3 is attributed to glucose oxidation, which is induced with high-current intensity, as the current intensity of O1 increases.
Performance Comparison of PdNS/f-CNT as Non-Enzymatic Glucose Sensors in Neutral pH
Under the optimal conditions, the PdNS/f-CNT/SPCE sensor at the fixed potential of +0.3 V vs. SCE was successfully used for amperometric glucose detection at neutral pH. The response current of the non-enzymatic sensor showed a wide linear regression against a glucose concentration in the range of 1-41 mM with a DL of 95 µM (S/N = 3) and sensitivity of 9.26 µA mM −1 cm −2 . Importantly, this range covers the normal human blood glucose level. Table 2 summarizes the analytical performances of the newly designed, non-enzymatic glucose sensor with those previously conducted in neutral environments. As a result, the analytical features of the proposed PdNS/f-CNT catalyst are comparable to those described in the literature. Furthermore, our designed non-enzymatic sensor can be utilized for practical sample testing with favorable accuracy and precision.
Selectivity of PdNS/f-CNT-Based Glucose Sensors
As mentioned previously, one of the main challenges in the non-enzymatic sensing of glucose is the electrochemical signals caused by some interfering substances such as ascorbic acid (AA), paracetamol (PCM), dopamine (DA), and galactose (Gal). Although the selectivity of non-enzymatic glucose sensors is often lower than their enzymatic counterparts, our developed glucose sensor still reveals a relatively good selectivity against various interfering species, such as fructose and sucrose. Amperometric measurements were conducted by applying +0.3 V vs. SCE to PdNS/f-CNT/SPCE in 0.1 M PBS (pH 7.4) containing glucose and interfering species. Table 3 summarizes the electrochemical responses of all tested molecules in the presence of glucose. No amperometric response could be identified for Suc and Fru, even at high concentrations of 1 and 10 mM. However, Gal, DA, PCM, and AA could cause interference with the glucose oxidation signal. Further discrimination of electroactive species on the PdNS/f-CNT/SPCE non-enzymatic sensor can be accomplished by: (i) using ion-exchange membranes over the sensor (e.g., if charged interference species are present), (ii) coupling the sensor with a proper sample preparation procedure to select the target analyte, and so on.
Conclusions
In conclusion, this study proposed a simple and cost-effective electrodeposition method to synthesize porous cauliflower-like Pd nanostructures (PdNS) on f-CNT/SPCE as an enzyme-free sensor for direct glucose detection in a neutral buffered solution (pH 7.4). Thanks to its high electrical conductivity, f-CNT increased the electrochemical surface area of the bare electrode, which promoted the electrodeposition of an important amount of particles of the PdNS catalyst. Furthermore, f-CNTs were used as carbon-based supporting materials for the construction of the PdNS electrocatalyst. Indeed, the electrocatalytic activity of the PdNS catalyst toward glucose oxidation was greatly enhanced by the insertion of f-CNTs that synergize with the PdNS catalyst. The PdCl 2 metallic precursor concentration and the electrodeposition conditions were studied and discussed. SEM analysis confirmed that these parameters had a significant effect on the morphological structure evolution of the deposited Pd nanostructures. Besides, a good correlation between the morphology evolution, structure, and electrocatalytic activity of the electrodeposited PdNS electrocatalyst toward glucose was observed. Furthermore, glucose electrochemical behavior and the potential-pH diagram for the palladium-water system enabled us to propose a mechanism for the glucose electrooxidation reaction. The proposed mechanism would be beneficial, as the basis, to achieve the electrocatalyst's highest activity by selecting the suitable potential range.
Nevertheless, the use of this new generation of abiotic sensors, for real applications, requires the acceptable selectivity of the f-CNT/SPCE-based PdNS, which might be obtained by: • Reducing the applied potential by inserting non-precious metals, polymers, etc.; • The use of ion-exchange membranes over the sensor (e.g., electrostatic repulsion if charged interference species are present); • Coupling the non-enzymatic sensor with a sample preparation procedure to select the target analyte.
|
2022-04-04T15:13:18.943Z
|
2022-04-01T00:00:00.000
|
{
"year": 2022,
"sha1": "26d8f16a5b0a626d1e53ca17bc3caca07501b54a",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/1424-8220/22/7/2706/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "8223e6e8de5b32fe651778a01a6da6d76aafd5b6",
"s2fieldsofstudy": [
"Materials Science"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
10991254
|
pes2o/s2orc
|
v3-fos-license
|
Plant growth-promoting activities of Streptomyces spp. in sorghum and rice
Five strains of Streptomyces (CAI-24, CAI-121, CAI-127, KAI-32 and KAI-90) were earlier reported by us as biological control agents against Fusarium wilt of chickpea caused by Fusarium oxysporum f. sp. ciceri (FOC). In the present study, the Streptomyces were characterized for enzymatic activities, physiological traits and further evaluated in greenhouse and field for their plant growth promotion (PGP) of sorghum and rice. All the Streptomyces produced lipase, β-1-3-glucanase and chitinase (except CAI-121 and CAI-127), grew in NaCl concentrations of up to 6%, at pH values between 5 and 13 and temperatures between 20 and 40°C and were highly sensitive to Thiram, Benlate, Captan, Benomyl and Radonil at field application level. When the Streptomyces were evaluated in the greenhouse on sorghum all the isolates significantly enhanced all the agronomic traits over the control. In the field, on rice, the Streptomyces significantly enhanced stover yield (up to 25%; except CAI-24), grain yield (up to 10%), total dry matter (up to 18%; except CAI-24) and root length, volume and dry weight (up to 15%, 36% and 55%, respectively, except CAI-24) over the control. In the rhizosphere soil, the Streptomyces significantly enhanced microbial biomass carbon (except CAI-24), nitrogen, dehydrogenase (except CAI-24), total N, available P and organic carbon (up to 41%, 52%, 75%, 122%, 53% and 13%, respectively) over the control. This study demonstrates that the selected Streptomyces which were antagonistic to FOC also have PGP properties.
Introduction
Plant growth-promoting (PGP) microbes are rhizosphere associated organisms that colonize the rhizosphere and rhizoplane and improve plant growth when artificially inoculated onto the seeds or into soil. PGP microbes may promote plant growth either by direct stimulation such as iron chelation, phosphate solubilization, nitrogen fixation and phytohormone production or by indirect stimulation such as suppression of plant pathogens and induction of resistance in host plants against pathogens (Basak and Biswas 2009;Hao et al. 2011;Panhwar et al. 2012). The opportunities of PGP microbes include alternating applications of PGP microbes as bio-fungicides with inorganic fungicides to manage fungicide resistance and to reduce the number of fungicide applications per year. PGP microbes also plays an important role in inte-run-off of unused fertilizers and the environment damage that results (Kloepper 2009;Gopalakrishnan et al. 2013a).
PGP and biological control of plant pathogens has been addressed using actinomycete, bacterial and fungal antagonists. For example, strains of Streptomyces, Bacillus, Pseudomonas and Trichoderma were effective not only in helping plants to mobilize and acquire nutrients but also to control plant pathogens (Postma et al. 2003;Khan et al. 2004;Perner et al. 2006;Borriss et al. Borriss et al. 2011;Gopalakrishnan et al. 2011a, b). Microorganisms with PGP and biocontrol potential were found at high incidence in compost, forest and pasture soils (Torsvik et al. 2002;Tinatin and Nurzat 2006). PGP traits of actinomycetes have been reported on pea (Tokala et al. 2002), bean (Nassar et al. 2003), tomato (El-Tarabily 2008), wheat (Sadeghi et al. 2012) and rice (Gopalakrishnan et al. 2012a).
Five strains of Streptomyces spp. , isolated from herbal vermi-compost, were earlier reported by us as having potential for biocontrol of Fusarium wilt in chickpea, caused by Fusarium oxysporum f. sp. ciceri (FOC; Gopalakrishnan et al. 2011b). Also, the selected Streptomyces strains were reported to produce siderophore, indole acetic acid (IAA; except KAI-90), hydrocyanic acid, cellulase (only KAI-32 and KAI-90) and protease (only for CAI-24 and CAI-127). The objective of this study was to further characterize the five Streptomyces strains for their enzymatic activities (chitinase, lipase and β-1-3-glucanase), physiological traits (salinity, pH, temperature, fungicide tolerance and antibiotic resistance) and to evaluate for their PGP traits under greenhouse and field conditions in sorghum and rice.
Evaluation of Streptomyces strains for their enzymatic activities Chitinase, lipase and β-1,3-glucanase production Chitinase and lipase production was conducted for all five strains (CAI-24, CAI-121, CAI-127, KAI-32 and KAI-90) of Streptomyces as described by Hirano and Nagao (1988) and Bhattacharya et al. (2009), respectively. The Streptomyces were streaked onto chitin agar (for chitinase) and Tween 80 agar (for lipase) and the Petri dishes were incubated at 28 ± 2°C for five days. At the end of incubation, the Petri dishes were observed for haloes around the colonies, indicating the production of these enzymes. All the treatments were replicated three times and the experiment was conducted three times. Chitinase production was recorded on a 0-5 rating scale as follows: 0 = no halo; 1 = halo of 1-5 mm; 2 = halo of 6-10 mm; 3 = halo of 11-15 mm; 4 = halo of 16-20 mm and 5 = halo of 21 mm and above. Observations for lipase production were recorded on a 0-5 rating scale as follows: 0 = no halo; 1 = positive; 2 = halo of 1-3 mm; 3 = halo of 4-6 mm; 4 = halo of 7-10 mm and 5 = halo of 11 mm and above.
β-1,3-glucanase was measured as per the protocols of Singh et al. (1999). All five Streptomyces strains were cultured individually in Tryptic soy broth, supplemented with 1% (weight/volume) colloidal chitin, at 28°C for four days. At the end of the incubation, the cultures were processed as per Gopalakrishnan et al. (2013b). The development of a dark red color indicated the presence of β-1, 3-glucanase. Treatments were replicated three times and the experiment was conducted three times. One unit of β-1,3-glucanase activity was defined as the amount of enzyme that liberated 1 μmol of glucose hour -1 under defined conditions.
Evaluation of Streptomyces strains for their physiological traits Salinity, pH and temperature The five Streptomyces were streaked on Bennett's agar with NaCl concentrations ranging from 0 − 8% (at intervals of 2%) and pH ranging from 5-13 (at intervals of 2 pH units); for pH 3, Bennett's broth was used and the intensity of growth was measured at 600 nm in a spectrophotometer after incubation at 28 ± 2°C for five days. For evaluating the effect of temperature, the five Streptomyces strains were streaked on Bennett's agar and incubated at 20, 30 and 40°C for five days, but for 50°C, Bennett's broth was used and the intensity of growth was measured at 600 nm in a spectrophotometer.
All the physiological traits were replicated three times and the experiment was conducted three times. Responses of the five Streptomyces to salinity, pH, temperature and fungicide tolerance were recorded as follows: 0 = no growth; 1 = slight growth; 2 = moderate growth and 3 = good growth.
Evaluation of Streptomyces strains for their PGP potential under greenhouse conditions on sorghum
The five Streptomyces strains antagonistic against FOC (CAI-24, CAI-121, CAI-127, KAI-32 and KAI-90) were evaluated in a greenhouse for their PGP potential on sorghum. A total of six treatments (five Streptomyces strains + un-inoculated control) were made with six replications. Pot mixture (1000 g) was prepared by mixing red soil, sand and farmyard manure at 3:2:2 and placed in 20 cm diam plastic pots. Sorghum seeds (variety ICSV 112) were surface sterilized with sodium hypochlorite (2.5% for 5 min) and rinsed with sterilized water (8 times) before being transferred into the respective test Streptomyces strains (10 8 cfu ml −1 ; grown in starch casein broth (SCB) separately) for an hour before being sown in the pots (six seeds/pot but thinned to three after one week). Booster doses of Streptomyces strains (5 ml per seedling, 10 8 cfu ml −1 ) were applied at 15, 30 and 45 days after sowing by the soil drench method. Growth parameters including plant height, leaf area, stem and leaf dry weight, root length, root surface area, root volume and root dry weight were determined at day 60 after sowing.
Evaluation of Streptomyces strains for their PGP potential under field conditions on rice
The field trial was performed in the 2011 Kharif (rainy) season at ICRISAT, Patancheru, Hyderabad, India. The details of the experimental site, soil and the experiment were described previously (Gopalakrishnan et al. 2012a). The experiment was conducted in a randomized complete block design with three replicates and subplot sizes of 10 × 7.5 m. Rice was grown by the system of rice intensification (SRI) method as described by Uphoff (2003). The five Streptomyces (CAI-24, CAI-121, CAI-127, KAI-32 and KAI-90) were grown on SCB at 28°C for five days before evaluated for their PGP traits. The control plots contained no Streptomyces. The 11-day-old single seedlings were uprooted from the nursery (laid next to the experimental field), their roots dipped in the respective Streptomyces broth (containing 10 7 cfu mL −1 ) for 60 min and transplanted on 4th August 2011 at a spacing of 25 × 25 cm (row-to-row and plant-to-plant spacing). Plants were inoculated with respective Streptomyces strains (1500 ml; 10 7 cfu ml −1 ) once in every two weeks until the flowering stage along with the irrigation water. Irrigation management was performed as recommended for the SRI method (alternate wetting and drying method; Uphoff et al. 2009). Weeding was performed by a Conoweeder at intervals of 15 days after transplanting until the flowering stage. The crop was not sprayed with any chemicals as no serious insect-pests/pathogens were noticed. The recommended dose of N:P:K (120: 60:40 kg ha −1 ) were applied through compost and farm yard manure. The crop was harvested manually on 23rd November 2011 and observed for plant height, number of tillers, primary and secondary panicle numbers, panicle length, stover yield, grain yield, total dry matter and 1000 seed weight. Root samples were collected from 0-15 cm and 15-30 cm soil profile after harvesting and analyzed for root length density (EPSON expression 1640×, Japan), volume and dry weight (dried in an oven at 60°C for 72 h). Soil samples (from 0-15 cm soil profile) were collected at harvesting and analyzed for soil chemistry (% organic carbon, available phosphorous and total nitrogen as per the standardized protocols of Nelson and Sommers (1982); Olsen and Sommers (1982) and Novozamsky et al. (1983), respectively) and biological analysis (dehydrogenase activity, microbial biomass nitrogen and microbial biomass carbon as per Casida (1977), Brooks et al. (1985) and Anderson and Domsch (1989), respectively).
Statistical analysis
The greenhouse and field experiment data were subjected to ANOVA (GenStat 10.1 version 2007, Lawes Agricultural Trust, Rothamsted Experimental Station) to evaluate the efficiency of the PGP agents. Significance of differences between the treatment means were tested at P = 0.01 and 0.05.
Discussion
The five Streptomyces strains (CAI-24, CAI-121, CAI-127, KAI-32 and KAI-90) used in the present investigation were earlier reported as not only having potential for biocontrol of Fusarium wilt disease in chickpea but also having PGP traits such as IAA (except KAI-90) and siderophore production (Gopalakrishnan et al. 2011b). Microbes producing IAA stimulate plant growth while siderophore producers bind Fe 3+ from the environment and make it available for plants and their growth (Patten and Glick 2002;Tokala et al. 2002;Hayat et al. 2010). In the present investigation, the five FOC antagonistic Streptomyces were further characterized for enzymatic activities, physiological traits and PGP potential on sorghum and rice under greenhouse and field conditions. In the enzymatic activity studies, all the five strains produced β-1,3-glucanase and lipase but only CAI-24, KAI-32 and KAI-90 produced chitinase. The cell walls of plant pathogens contain β-1,3-glucan, chitin and lipid, which are essential for the pathogen for disease transmission and pathogenesis, and lysis of these by β-1,3glucanase/chitinase/lipase-producing microbe leads to leakage of cell contents and collapse of the pathogenic fungi (Singh et al. 1999;Lynd et al. 2002;Macagnan et al. 2008). Hence, microorganisms having these traits can be exploited for biological control of plant pathogens, which indirectly promotes the plants. In the physiological traits studies, the five Streptomyces strains were able to grow in NaCl up to 6% and at pH values between 5 and 13 and temperatures between 20 and 40°C. The ability of Streptomyces spp. to grow in harsh pH and temperature and higher concentration of salinity is reported (Sadeghi et al. 2012). In the present study, all the Streptomyces strains were highly tolerant to the fungicide Bavistin at field application level. Therefore, it is concluded that these strains may not only have the capability to survive in harsh environments but are also compatible with the Bavistin and hence can be used in integrated disease management programs.
In the present study, under greenhouse conditions in sorghum and field conditions in rice, the Streptomyces strains significantly enhanced all the agronomic observations including root length, volume and dry weight and yield parameters over the un-inoculated control. The Streptomyces strains are reported widely in the literature for its PGP potential (Nassar et al. 2003;El-Tarabily 2008;Gopalakrishnan et al. 2012a;2013b). As hypothesized earlier, the mechanism by which the Streptomyces enhanced the morphological and yield parameters on both sorghum and rice could be their PGP traits including IAA and siderophore production (direct stimulation of PGP; Gopalakrishnan et al. 2011b) and or chitinase, lipase and β-1,3-glucanase production (indirect stimulation of PGP; as reported in the present investigation). The effect of soil microbes on PGP including root development has been reported by Birkhofer et al. (2008) and Uphoff et al. (2009).
In the SRI method of rice cultivation, irrigation was done by alternate wetting and drying method. Such a system of irrigation favors soil microbial and biological activities and enhances the availability of mineral nutrients compared to rice grown by flooded cultivation methods (Uphoff et al. 2009;Gopalakrishnan et al. 2012b). In the present investigation, soil biological activities including microbial biomass carbon, nitrogen and dehydrogenase and mineral nutrients including available P, total N and % organic carbon were found to be higher in the five Streptomyces inoculated plots over the uninoculated control plots. Similar results were found in SRI rhizospheres compared with those of the same variety of rice plants grown conventionally (Gayathry 2002). Hence, it can be concluded that the five Streptomyces strains survived in the rice rhizosphere and enhanced the soil health conditions. In the present investigation, although roots were not checked for colonization, the data on agronomical (including roots), yield, biological and mineral nutrition studies indicate that the five 6 3 C = carbon; N = nitrogen; P = phosphorous; SE = standard error; LSD = least significant difference; CV = coefficient of variance; * = statistically significant at 0.05; ** = statistically significant at p = 0.01; *** = statistically significant at p = 0.001.
Streptomyces strains had multiplied and colonized the rice roots. Hence, it can be concluded that the five Streptomyces strains used in this study were apparently well adapted to the rice and sorghum rhizosphere environments and enhanced the soil health and plant growth conditions. Significant increases in growth and yield of agronomically important crops in response to inoculation with PGP microbes have been repeatedly reported (Biswas et al. 2000;Asghar et al. 2002;Silva et al. 2006). PGP microbes are thought to stimulate plant growth by two mechanisms: (1) by altering hormone balance in the host plant and increasing mineral nutrient solubilization (direct effects); (2) by antagonism towards plant pathogens (indirect effects; Glick 1995). In the present investigation, the five Streptomyces strains were found to have both mechanisms, having PGP traits in rice and sorghum (direct effects) and biocontrol traits, against Fusarium wilt, in chickpea (indirect effects; Gopalakrishnan et al. 2011b). Broad spectrum PGP and biocontrol agents offer potentially effective novel strategies for controlling multiple pathogens and insect pests. A few of the available biocontrol agents, mostly belonging to Pseudomonas spp., have shown broad spectrum antifungal activity (Hass and Keel 2003;Viji et al. 2003;Kishore et al. 2005). Actinomycetes and other bacteria are also reported in the literature as broad spectrum PGP and biocontrol agents for soilborne fungal plant pathogens and insect-pests, such as Helicoverpa armigera and Spodoptera litura (El-Tarabily and Sivasithamparam 2006; Sadeghi et al. 2012). The five Streptomyces used in this study were well adapted to not only in the chickpea rhizosphere but also in the sorghum and rice rhizosphere where they promoted plant growth. Hence, these isolates could be used as PGP agents in addition to the biocontrol of Fusarium wilt. Though, all the five strains have been demonstrated for their PGP potential in rice KAI-32 and KAI-90 were found to have superiority over other isolates in terms of crop productivity and root development. The broad range of PGP and antifungal activities of the five Streptomyces strains demonstrates multiple mechanisms of actions including antibiosis, production of cell wall degrading enzymes and plant growth-promoting hormone (IAA). Therefore, these Streptomyces can be considered for isolation of novel secondary metabolites which may be of importance for various biocontrol and PGP applications. Use of biocontrol agents such as these broadspectrum Streptomyces will probably be one of the important tactics for plant disease management in the near future as they allow the reduced use of pesticides and fertilizers that are potential pollutants of the environment.
|
2018-04-03T00:20:39.319Z
|
2013-10-29T00:00:00.000
|
{
"year": 2013,
"sha1": "9f4e504ffe4fbc97c554313211399fcacffb3ee3",
"oa_license": "CCBY",
"oa_url": "https://springerplus.springeropen.com/track/pdf/10.1186/2193-1801-2-574",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "daf8f8f82663b31adcb3ccd2baf4d46bcc42bbf8",
"s2fieldsofstudy": [
"Agricultural and Food Sciences"
],
"extfieldsofstudy": [
"Medicine",
"Biology"
]
}
|
4231632
|
pes2o/s2orc
|
v3-fos-license
|
High Add Valued Application of Turpentine in Crop Production through Structural Modification and QSAR Analysis
Turpentine is a volatile component of resin, which is an abundant forest resource in Southern China. As one of the most important components, the integrated application of β-pinene has been studied. The broad-spectrum evaluation of β-pinene and its analogues has, therefore, been necessary. In an attempt to expand the scope of agro-activity trials, the preparation and the evaluation of the herbicidal activity of a series of β-pinene analogues against three agricultural herbs were carried out. In accordance with the overall herbicidal activity, it is noteworthy that compounds 6k, 6l, and 6m demonstrated extreme activity with IC50 values of 0.065, 0.065, and 0.052 mol active ingredients/hectare against E. crus-galli. The preliminary structure–activity relationship (SAR) was analyzed and the compounds with the appropriate volatility and substituent type that had beneficial herbicidal activity were analyzed. Simultaneously, the quantitative structure–activity relationship (QSAR) model was built and the most important structural features were indicated, which was, to a certain extent, in line with the SAR study. The study aimed to study the application of the forest resource turpentine in agriculture as a potential and alternative approach for comprehensive utilization.
Introduction
Weeds are considered worldwide as an infestation for crops. Some weed species interfere with the production of grain and vegetables and severely reduce the yield and quality [1]. Among them, Echinochloa crus-galli is one of the world's worst weeds because it can seize the available soil nitrogen from the crop [2]. Effective and integrated control of these noxious weeds is essential for modern agriculture [3]. However, the wide use of single-type herbicides causes the serious occurrence of resistance weeds and a series of environmental issues [4]. Therefore, developing more effective and low toxic herbicides has become a priority for agriculture producers, researchers, and extension personnel [5][6][7][8][9][10]. As an important alternative to replacing traditional herbicides, forest resources have attracted increasing attention from researchers due to their unique qualities, such as being widely distributed, environmentally friendly, and diverse in chemical composition [11]. The acrylopimaric acid-based botanical herbicides have been synthesized from forest resource rosin [3]. The natural herbicide activity of Satureja hortensis L., an essential oil against two important weed species, has been Molecules 2018, 23 reported [12]. Mansonone E has potential as a new natural pesticide for agricultural plant pathogen management [13]. Therefore, taking advantage of forest resources for agriculture through chemical structural modification is a viable solution for the production of new herbicides. Resin, an abundant forest resource, is obtained from pines, which use it to resist external insect bites and other damage. In the growth of plants, resin components (including pinene) play many special roles, such as plant defense, plant interspecific phytochemical roles, attracting insects for pollination, and so on. Just as plant secondary metabolites, resin is agriculturally active and the high value-added application of resin is meaningful. In the complex chemical components of resin, turpentine, a monoterpene mixture of volatile oil, has excellent performance and potential for use in many fields. Turpentine has a high content of β-pinene and the cost to isolate β-pinene is low. However, studies on the biological activity of β-pinene derivatives are not rich. Therefore, further development of derivatives with bioactive properties using β-pinene as a starting material is gradually becoming a trend in turpentine processing and utilization. For instance, β-pinene and its analogues have agricultural activities, such as antifeedant, repellent, and antimicrobial properties [14][15][16]. The preparation of most compounds has been reported by us. Based on previous research, the reaction process has been improved. In order to further study the impact of heteroatoms on the activity, other novel heterocyclic compounds (6k-6n) have been synthesized [17,18]. The herbicidal activity of β-pinene and its analogues is worthy of study in order to develop some ecofriendly herbicides that can be applied in the control of noxious weeds.
In a program to design and synthesize herbicidal active substances via derivation of volatile β-pinene, there are many factors that need to be considered. First, the primary factor is the effective contact between the herbicide and the treated soil [19]. In order to improve the permeability (the penetration of the molecule through the plant epidermis, which is influenced by amphipathic characteristics, and better the biphasic solubility) and reduce volatilization, ester and amino groups were introduced into β-pinene. Second, new technology is needed to screen potential herbicides from mickle candidates for the weed control. QSAR, an efficient means to screen biological activity and mechanisms of action, was carried out through quantification software packages [20][21][22]. After the essential structural features for the activity have been defined, a study on the mechanisms of action should be targeted. The objective of this study was to continue expanding the application of volatile β-pinene in a high value-added field and to develop its broad-spectrum biological activity in agriculture.
Synthesis of Derivatives of β-Pinene
The synthesis of two series of turpentine-based ester derivatives are shown in Schemes 1 and 2. The preparation of most of these compounds has been reported by us and, based on previous research, the reaction process has been improved. In order to further study the impact of heteroatoms on the activity, other novel heterocyclic compounds (6k-6n) were synthesized. The dehydrocumic acid was prepared from β-pinene through reactions of alkaline oxidation, dehydration, and isomerization, respectively. For the purpose of improving the reactivity of carboxylic acid, the dehydrocumic acid was activated by thionyl chloride into acid chloride. In the synthesis process of oxime esters, some of the intermediate oxime compounds have been prepared by us. In order to obtain a high yield, the synthesis process was optimized by regulating the reaction temperature and the reaction solvent and alkali. The optimal reaction conditions were ethanol as the reaction solvent, sodium carbonate as the base, and 25 • C as the reaction temperature. The structures of the title compounds have been well characterized by IR, 1 H-NMR, MS, and elemental analysis.
Herbicidal Activity and Structure-Activity Relationships (SARs)
In Supplementary Materials (Table S1), the herbicidal activities of compounds 5a-5l and 6a-6n against E. crus-galli, Amaranthus retroflexus, and Brassica campestris L. var. amplexicaulis Makino were screened primarily at 200 g active ingredients/hectare. It is shown that all of the ester and oxime eater derivatives of β-pinene had good inhibition rates against E. crus-galli compared with another two herbs. From Table 1, the research on the herbicidal activity against E. crus-galli was done and the analogues displayed good herbicidal activity, which was more significant compared with compound 3. It is noteworthy that compounds 6k, 6l, and 6m demonstrated extreme activity with IC50 values of 0.065, 0.065, and 0.052 mol active ingredients/hectare against E. crus-galli, which were similar to that for sulfentrazone, a commercialized agriculture herbicide. For different types of
Herbicidal Activity and Structure-Activity Relationships (SARs)
In Supplementary Materials (Table S1), the herbicidal activities of compounds 5a-5l and 6a-6n against E. crus-galli, Amaranthus retroflexus, and Brassica campestris L. var. amplexicaulis Makino were screened primarily at 200 g active ingredients/hectare. It is shown that all of the ester and oxime eater derivatives of β-pinene had good inhibition rates against E. crus-galli compared with another two herbs. From Table 1, the research on the herbicidal activity against E. crus-galli was done and the analogues displayed good herbicidal activity, which was more significant compared with compound 3. It is noteworthy that compounds 6k, 6l, and 6m demonstrated extreme activity with IC50 values of 0.065, 0.065, and 0.052 mol active ingredients/hectare against E. crus-galli, which were similar to that for sulfentrazone, a commercialized agriculture herbicide. For different types of Scheme 2. Structure and chemical synthesis of the compounds 6a-6n.
Herbicidal Activity and Structure-Activity Relationships (SARs)
In Supplementary Materials (Table S1), the herbicidal activities of compounds 5a-5l and 6a-6n against E. crus-galli, Amaranthus retroflexus, and Brassica campestris L. var. amplexicaulis Makino were screened primarily at 200 g active ingredients/hectare. It is shown that all of the ester and oxime eater derivatives of β-pinene had good inhibition rates against E. crus-galli compared with another two herbs. From Table 1, the research on the herbicidal activity against E. crus-galli was done and the analogues displayed good herbicidal activity, which was more significant compared with compound 3. It is noteworthy that compounds 6k, 6l, and 6m demonstrated extreme activity with IC 50 values of 0.065, 0.065, and 0.052 mol active ingredients/hectare against E. crus-galli, which were similar to that for sulfentrazone, a commercialized agriculture herbicide. For different types of β-pinene analogues, the level of herbicidal activity was oxime esters (6a-6n) > monoesters (5a-5l). Nevertheless, the environmental toxicology research on the newly synthesized compounds was very necessary. On the one hand, in order to avoid pesticide residues of herbicides in the soil, water, and air, the chemical stability and products of metabolism of the candidates must be experimentally tested under the influence of prolonged exposure to ambient temperature, sunlight, microflora or selected microorganisms, and plants. On the other hand, the impact of herbicides on non-target organisms such as bird, fish, and bees should also be assessed. Research in this area will be carried out in the follow-up work. In the process of herbicides playing a role, there are many factors that restrict the interaction with the herbs. For instance, the suitable volatility and electrostatic interaction may be mainly affected [1,23]. In order to improve the volatility, the structure of the carboxylic acid (compound 3) was modified with a hydroxyl group to obtain ester analogues. From the activity data shown in Table 1, the shorter branched chained compounds were intrinsically more toxic, for the monoesters compounds (5a-5l), and the activity sequence was 5a > 5b > 5c > 5d > 5g > 5e ≈ 5f ≈ 5h > 5i > 5j > 5k > 5l. In light of the point of view of Li et al. and Fan et al. aliphatic compounds are more volatile than aromatic compounds, and the volatility decreases with an increase in the molecular weight of the compounds [8,24]. Therefore, with regard to the same series of derivatives, along with the branched chain length increasing, the herbicidal activity showed a downward trend, and a similar trend was observed with oxime esters compounds (6a-6n). The electrostatic interaction of compounds is another important factor for herbicidal activity, and the type and position of the substituent has a great impact on the electrostatic effect of the compound. From investigations of the relationship of charge density and drug activity, the heteroatoms N, O, and S can play electron-withdrawing or electron-donating actions. Therefore, heterocyclic substituents such as thiophene, furan, and pyrrole were introduced to the derivatives. Based on the activity results, the rough order of herbicidal activity was aromatic substituted compounds was higher than aliphatic substituted compounds, and specifically heterocyclic substituted compounds (6a, 6b, 6f, 6k, 6l, 6m and 6n) was higher than benzene substituted compounds (6c, 6d, 6i and 6j). As a preliminary conclusion, during the molecules playing an effect on E. crus-galli, the ring of thiophene and furan accepts the electron, which forms an electron transfer process with a protein or enzyme to achieve the herbicidal effect. To sum up, the volatility and substituent type are two important compound properties influencing the herbicidal activity.
QSAR Study on Herbicidal Activity Against E. crus-galli
There were six groups of descriptors that reflected all of the features of the compounds, such as molecular composition, connectivity, 3D-coordinates, charge distribution, and quantum chemical data. Through calculation and selection, several significant molecular descriptors were determined during the model development process. The most important descriptors in this work were the quantum-chemical descriptors. In an attempt to establish a relationship between activity and molecular descriptors through several regression approaches, the heuristic regression was chosen to obtain a QSAR model with satisfactory values of R 2 , F, and S 2 . Through the "breaking point" rule and the number of molecular descriptors being no more than one-third of the sample number plus one, the descriptor number was determined. The "breaking point" rule results are shown in Figure 1. Finally, the four-descriptor QSAR model was determined as the best model and is shown in Table 2. The four significant descriptors and their values are listed in the Supplementary Materials (Table S2).
great impact on the electrostatic effect of the compound. From investigations of the relationship of charge density and drug activity, the heteroatoms N, O, and S can play electron-withdrawing or electron-donating actions. Therefore, heterocyclic substituents such as thiophene, furan, and pyrrole were introduced to the derivatives. Based on the activity results, the rough order of herbicidal activity was aromatic substituted compounds was higher than aliphatic substituted compounds, and specifically heterocyclic substituted compounds (6a, 6b, 6f, 6k, 6l, 6m and 6n) was higher than benzene substituted compounds (6c, 6d, 6i and 6j). As a preliminary conclusion, during the molecules playing an effect on E. crus-galli, the ring of thiophene and furan accepts the electron, which forms an electron transfer process with a protein or enzyme to achieve the herbicidal effect.
To sum up, the volatility and substituent type are two important compound properties influencing the herbicidal activity.
QSAR Study on Herbicidal Activity Against E. crus-galli
There were six groups of descriptors that reflected all of the features of the compounds, such as molecular composition, connectivity, 3D-coordinates, charge distribution, and quantum chemical data. Through calculation and selection, several significant molecular descriptors were determined during the model development process. The most important descriptors in this work were the quantum-chemical descriptors. In an attempt to establish a relationship between activity and molecular descriptors through several regression approaches, the heuristic regression was chosen to obtain a QSAR model with satisfactory values of R 2 , F, and S 2 . Through the "breaking point" rule and the number of molecular descriptors being no more than one-third of the sample number plus one, the descriptor number was determined. The "breaking point" rule results are shown in Figure 1. Finally, the four-descriptor QSAR model was determined as the best model and is shown in Table 2. The four significant descriptors and their values are listed in the Supplementary Materials (Table S2). Table 3, which were also compared in Figure 2. The following equation (1)) describes the four-descriptor QSAR model. In the model, a positive sign (+) indicates that the descriptor value had a positive correlation with log IC 50 , and that the log IC 50 value was higher as the descriptor value was larger. In contrast, a negative sign (−) indicates a negative correlation. The final four-descriptor QSAR model contained the sample number N = 26, and had the following statistical characteristics: R 2 = 0.9428, F = 86.49, S 2 = 0.0014. The experimental and predicted log IC50 values are shown in Table 3, which were also compared in Figure 2. The following equation (Equation (1)) describes the four-descriptor QSAR model. In the model, a positive sign (+) indicates that the descriptor value had a positive correlation with log IC50, and that the log IC50 value was higher as the descriptor value was larger. In contrast, a negative sign (−) indicates a negative correlation. The validation of the established QSAR model was carried out by two validation methods. The internal validation method was described as the compounds 1, 4, 7, etc. were assigned to group A; compounds 2, 5, 8, etc. were assigned to group B. The other compounds belonged to subset C. The training set was two of these three groups, and the test set was the other group. The values of the corresponding test sets were predicted by the obtained correlation equation from the training set using the identical descriptors. The results of the internal validation are listed in Supplementary Materials (Table S3). The difference between R Training 2 and R Test 2 for the three sets was tiny enough to be ignored, and the average values of R Training 2 and R Test 2 were essentially identical to the overall R 2 value, which meant the approving predictive power of the obtained model. In the "leave-one-out" method, the compounds 1, 5, 9, etc. were the external test set and the others were the training set.
The R 2 value of the training set and test set were close and the QSAR developed in this study had good predictability. With regard to the descriptors related to the QSAR model, the first important descriptor was the heat of formation of the molecule (∆H f ), which gives the energy of the molecule in the thermodynamic standard scale (elements in ideal gas state at 298.15 K and 101.325 Pa). It measures the change of enthalpy during the formation of 1 mol of the compound from its constituent elements with all substances in their standard states. This quantum-chemical descriptor indicates the level of reactivity of the molecule and can find the standard enthalpy change of any reaction [21]. The value of ∆H f signifies the difficulty and heat change of the chemical reaction. A large absolute value ∆H f implies a prone reaction and a negative value ∆H f means an exothermic reaction. The converse is also true. From Equation (1), the ∆H f value had a negative effect on log IC 50 values.
The second important descriptor is the max atomic orbital electronic population (P µµ ), and it is a simplified index to describe the nucleophilicity of a molecule. According to the molecular orbital theory and frontier orbital theory, the highest occupied molecular orbital (HOMO) and the lowest unoccupied orbital (LUMO) are influential on the activity of the molecule [21]. When a molecule reacts with a target molecule, it occurs near the molecular front-orbital; therefore, the frontier orbital energy can greatly affect the herbicidal activity of a molecule. The suitable values of HOMO and LUMO are beneficial for the pesticide activity. From Figure 3, the frontier molecular orbitals (FMOs) of compounds 6k, 6l, and 6m are shown, which indicate the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). The HOMO and LUMO of the molecule mainly distribute in heterocyclic and benzene rings, which means that these active groups accept or provide electrons during the interaction with the receptor. The third most important descriptor in the model is the tot dipole of the molecule, which measures the separation of positive and negative electrical charges within a molecule, that is, a measure of the overall polarity of the molecule [23,25,26]. The electric field strength of the dipole is proportional to the magnitude of the dipole moment. In Figure 4, the molecular electrostatic potential and contour maps of compounds 6k, 6l and 6m are illustrated. The electron withdrawing substitute of O and S atoms, demonstrated by the green parts, represents the positive molecular orbitals. The electron donating substitute of N atoms shown in red, represents the negative The third most important descriptor in the model is the tot dipole of the molecule, which measures the separation of positive and negative electrical charges within a molecule, that is, a measure of the overall polarity of the molecule [23,25,26]. The electric field strength of the dipole is proportional to the magnitude of the dipole moment. In Figure 4, the molecular electrostatic potential and contour maps of compounds 6k, 6l and 6m are illustrated. The electron withdrawing substitute of O and S atoms, demonstrated by the green parts, represents the positive molecular orbitals. The electron donating substitute of N atoms shown in red, represents the negative molecular orbitals. From Equation (1), the µ value has a positive influence on the herbicidal activity. The third most important descriptor in the model is the tot dipole of the molecule, which measures the separation of positive and negative electrical charges within a molecule, that is, a measure of the overall polarity of the molecule [23,25,26]. The electric field strength of the dipole is proportional to the magnitude of the dipole moment. In Figure 4, the molecular electrostatic potential and contour maps of compounds 6k, 6l and 6m are illustrated. The electron withdrawing substitute of O and S atoms, demonstrated by the green parts, represents the positive molecular orbitals. The electron donating substitute of N atoms shown in red, represents the negative molecular orbitals. From Equation (1), the μ value has a positive influence on the herbicidal activity. The last important descriptor in the model is the max net atomic charge for an O atom. This electrostatic descriptor reflects the characteristics of the charge distribution of the molecule and is related to the strength of the intermolecular bonding interactions and characterizes the stability of the molecules, their conformational flexibility, and the other valency-related properties [24,25]. Figure 5 shows the optimized geometries and charge distribution of compounds 6k, 6l and 6m.
Synthesis and Characterizations
β-pinene (1) was obtained from Jiangxi Jishui Hongda Natural Spices Ltd. (Jishui, China) of a commercial source. All the other chemicals used in the synthesis were of reagent grade and the purity was chemically pure (≥ 99.5%). A Nicolet IS 10 spectrophotometer (Thermo Nicolet Co., Ltd.,
Synthesis and Characterizations
β-pinene (1) was obtained from Jiangxi Jishui Hongda Natural Spices Ltd. (Jishui, China) of a commercial source. All the other chemicals used in the synthesis were of reagent grade and the purity was chemically pure (≥99.5%). A Nicolet IS 10 spectrophotometer (Thermo Nicolet Co., Ltd., Madison, WI, USA) was used for the Fourier transform infrared (FT-IR) spectra. A Bruker AV-300 nuclear magnetic resonance spectrometer (Bruker Co., Ltd., Karlsruhe, Germany) was used to measure the 1 H-NMR spectra. An Agilent-5973 spectrophotometer was (Agilent Technologies Inc., Santa Clara, CA, USA) utilized for the MS spectra. A Bruker Q-TOF mass spectrometer (Bruker Bruker Co., Ltd., Karlsruhe, Germany) equipped with an electro spray ionization source was applied for ESI mass spectral data. Thin-layer chromatography (TLC) was employed to monitor the progress of reactions and determine the reaction's end, which was carried out on Merck silica gel 60 GF254 plates (Qingdao Bangkai Hi-Tech Materials Co., Ltd., Qingdao, China) with eluent of petroleum ether/ethylacetate (v/v = 8:1). The preparation and the structures of title compounds are shown in Schemes 1 and 2.
Synthesis of 4-Isopropylcyclohexa-1,3-dienecarboxylic acid (Dehydrocumic acid, 3)
β-pinene was used as the raw material. Through alkaline oxidation, dehydration, and isomerism, the nopinic acid (2) and dehydrocumic acid (3) were prepared according to the method described previously [24]. During this process, the colorless crystal of 2 was obtained by recrystallization with organic solvent (such as ethanol or toluene) [25]. After dehydration and isomerization with concentrated sulfuric acid, the hydroxyl of the carboxylic acid was eliminated and the crude product 3 was obtained, which was used in the subsequent reaction after purification by recrystallization. Crystal Data for C 10 From the lead compound 3, the 12 resulting compounds 5a-5l were obtained by an esterification reaction between dehydrocumic acid chloride and the corresponding alcohols [26].
Herbicidal Activity Assay of β-Pinene Analogues
β-pinene analogues were evaluated for herbicidal activity of the representative monocotyledonous weed E. crus-galli and the dicotyledonous weeds A. retroflexus and B. campestris L. The herbs were grown in a greenhouse and the test method was the conventional plate method (in vitro) [25][26][27]. In the evaluation, all of the test compounds were dissolved in mixed solution of DMF and distilled water (v/v = 8:1). The inhibition percentage of seed germination was the mean value obtained through three independent experiments and the standard deviation (SD) values were limited within ±5%. On the basis of the series inhibition percentage, the IC 50 results were calculated by the SPSS statistics program version 17.0 (SPSS Inc., Chicago, IL, USA). For stringency, the commercial herbicide sulfentrazone (Shanghai Aladdin Bio-Chem Technology Co., Ltd., Shanghai, China) was chosen as the positive control, and a pure solution without test compound was tested as the negative control.
The initial concentration was set at 200 g active ingredients/hectare and the lower required concentrations were diluted with water (100, 50, 25 and 12.5 g active ingredients/hectare). For better dispersion all the solutions were prepared with Tween (final concentration 0.1%) as surfactant. In petri dishes of 9 cm diameter, 2 pieces of filter paper were placed and 2 mL test compound solution with certain concentrations was sprayed evenly. For the accuracy of the results, the negative control was the same amount of DMF mixed with Tween-80 and water. Ten just-sprouting seeds of each weed species including E. crus-galli, A. retroflexus and B. campestris L. were grown in the test plate and allowed to germinate for 72 h at 28 ± 1 • C. Finally, the average values of the height of the ground portion of the above mentioned seedlings growing in each plate were calculated. For reproducibility and credibility, all of the tests at each concentration were repeated three times. The inhibition percentage (IP) was used to describe the control efficiency of the compounds and was calculated according to Equation (2): where H 0 is the height of control test and H 1 is the height of treated test. Through regression analysis using SPSS Statistics 17.0, the IC 50 values were calculated by the inhibition percentage of E. crus-galli at five concentrations for each test compound [28].
Building and Verification of the Quantitative Structure-Activity Relationship (QSAR) Model
The QSAR model was built and validated in accordance with the common procedure. Briefly, optimization of the most stable configuration was obtained with a Gaussian 03W package of programs (Gaussian Inc., Wallingford, CT, USA,), and all the molecular descriptors were calculated by CODESSA 2.7.15 (The copyright of this version is held by the Center of Heterocyclic Chemistry, University of Florida.) [29]. The "breaking point" rule for determining the number of the descriptors is described in Figure 1. The best multilinear regression showed a significant increase in R 2 when the number of the descriptors was less than or equal to 4. However, there was negligible change in R 2 when the number of descriptors increased from 4 to 5. Descriptors with high t values were accepted and those with low t values were rejected. A "breaking point" indicates that the improvement of the regression model has become non-significant (R 2 < 0.02-0.04). For the determination of the most significant structural features of herbicidal activity against E. crus-galli, the heuristic analysis was to build the QSAR model. During the model development process, the related statistical parameters (R 2 , S 2 and F) were determined to assess the predictability of a given model. In this study, a model displaying good predictability between the compounds' structures and the log IC 50 values was developed. To ensure high quality of the final QSAR results, two verification methods, internal validation, and the "leave-one-out" cross-validation were used for determination [30].
Conclusions
In order to expand the efficient application of the forest resource β-pinene to agriculture, two series of β-pinene analogues-ester and oxime esters-were prepared with suitable volatility and a substituent type through structural modification. The herbicidal activity against E. crus-galli, A. retroflexus and B. campestris L. was evaluated. Among them, the activity against E. crus-galli was more ideal; therefore, a study on the structure-activity relationship (SAR) against E. crus-galli was focused on. The preliminary conclusion is that compounds with esterification modification and a short chain length are beneficial for herbicidal activity. In addition, the heteroatoms N, O and S can play an active role. Simultaneously, the quantitative structure-activity relationship (QSAR) model (R 2 = 0.9428, F = 86.49, S 2 = 0.0014) was built and indicates that the most important structural feature is the heat of formation of the molecule (∆H f ). In view of these results, for the further exploitation of ecofriendly herbicides and the better application of volatile β-pinene, short chain esters, and heterocyclic-substituted derivatives could be the direction of the future development of new structures. It is noteworthy that environmental toxicology should be focused on in future work.
Supplementary Materials: The following are available online. IR, 1 H-NMR, MS and elemental analysis data for the target compounds. The herbicidal activity and structure descriptors of the title compounds are also available as supplementary information.
|
2018-03-25T18:58:31.636Z
|
2018-02-01T00:00:00.000
|
{
"year": 2018,
"sha1": "0644cfdb9a102fd615b11e83b01d261e271cba46",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/1420-3049/23/2/356/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "0644cfdb9a102fd615b11e83b01d261e271cba46",
"s2fieldsofstudy": [
"Agricultural and Food Sciences",
"Chemistry",
"Environmental Science"
],
"extfieldsofstudy": [
"Mathematics",
"Medicine"
]
}
|
36891804
|
pes2o/s2orc
|
v3-fos-license
|
Evaluation of Preexisting Anti-Hemagglutinin Stalk Antibody as a Correlate of Protection in a Healthy Volunteer Challenge with Influenza A/H1N1pdm Virus
ABSTRACT Influenza virus hemagglutinin (HA) surface glycoprotein is currently the primary target of licensed influenza vaccines. Recently, broadly reactive antibodies that target the stalk region of the HA have become a major focus of current novel vaccine development. These antibodies have been observed in humans after natural infection with influenza A virus, but the data are limited. Using samples and data from the uniquely controlled setting of an influenza A/H1N1 virus human challenge study of healthy volunteers, we performed a secondary analysis that for the first time explores the role of anti-HA stalk antibody as a human correlate of protection. An anti-HA stalk antibody enzyme-linked immunosorbent assay (ELISA) was performed on samples from 65 participants challenged with a 2009 H1N1pdm virus. Pre- and postchallenge anti-HA stalk titers were then correlated with multiple outcome measures to evaluate anti-HA stalk antibody titer as a correlate of protection. Anti-HA stalk antibody titers were present before challenge and rose in response to challenge in 64% of individuals. Those individuals with higher titers at baseline were less likely to develop shedding, but not less likely to develop symptoms. Similar to the hemagglutination inhibition (HAI) titer, the baseline anti-HA stalk antibody titer did not independently predict a decrease in the severity of influenza disease, while the antineuraminidase (neuraminidase inhibition [NAI]) titer did. As a correlate of protection, the naturally occurring anti-HA stalk antibody titer is predictive of a reduction of certain aspects of disease similar to HAI titer, but the NAI titer is the only identified correlate that is an independent predictor of a reduction of all assessed influenza clinical outcome measures.
T he influenza virus hemagglutinin (HA) surface glycoprotein is currently the primary target of all licensed vaccines for influenza and considered the dominant antigen to which individuals develop a humoral immune response. Until recently, antibodies to the antigenically and genetically variable head region of the HA have been the major focus of influenza serological studies and vaccine development, specifically antibodies that sterically inhibit hemagglutinin receptor binding, utilizing the hemagglutination inhibition (HAI) assay to assess antibody titers. Over the course of the past decade, there has been more focus on discovery of antibodies that target the more conserved stalk (or stem) region of the HA (1). These antibodies can be broadly neutralizing across one of two groups of the 18 known influenza A virus (IAV) HA subtypes, group 1, which includes seasonal human H1 and highly pathogenic avian H5, and group 2, which includes seasonal H3 and the avian H7 subtype that has caused increasing numbers of human infections during the 2016 -2017 influenza season in the form of an avian H7N9 IAV throughout China (2). These broadly neutralizing anti-HA stalk antibodies have been hypothesized to have played a major role in the extinction of the previously circulating prepandemic seasonal H1N1 lineage after the 2009 H1N1pdm virus emerged (3). The possibility that these broadly neutralizing anti-stalk antibodies that bind to conserved epitopes of group 1 or group 2 HA stalk could elicit broadly protective responses has led a number of groups to investigate novel vaccines and therapeutics based on anti-stalk antibody technology (1,4,5). Broad protection in animal models with anti-stalk antibody-based vaccines/constructs has been shown (4), and some of these investigational products are in preparation for future clinical trial evaluation.
Anti-stalk antibodies that are broadly reactive and neutralizing in vitro and in vivo have been observed to occur naturally after infection in small studies of individuals infected with seasonal viruses, but these antibodies do not seem to be elicited significantly by seasonal vaccine (3,6,7). In addition, these anti-stalk antibodies have been identified in 18 patients infected with H7N9 strains in China (8), an ongoing emerging threat with pandemic potential. Insufficient serological data have been available in humans to evaluate whether anti-stalk antibodies serve as correlates of protection against influenza virus infection. Further investigation is needed to elucidate what role they play in immunity during naturally occurring infections and importantly whether novel anti-stalk-based vaccine candidates could elicit titers higher than those observed in natural infections.
The Healthy Volunteer H1N1 Influenza Challenge Model developed at the National Institutes of Health (NIH) Clinical Center (9) offers a unique opportunity to explore influenza pathophysiology and correlates of protection/disease modification in a controlled setting. Recently, it was demonstrated in the H1N1 challenge model that HAI titer and neuraminidase inhibition (NAI) titer, a measure of anti-NA antibody, predicted protection from different aspects of influenza disease (10). Using serum samples as well as clinical and virological data from this previous study, we performed a secondary analysis to explore the role of anti-HA stalk antibody in protection from and modification of various aspects of influenza disease and compared this to the previous measures of anti-HA head antibody as well as anti-NA antibody as correlates of protection and disease modification.
RESULTS
Validation of the anti-HA stalk enzyme-linked immunosorbent assay (ELISA) used in this study demonstrated that the assay was capable of measuring relevant anti-stalk antibodies that bind to important conformational epitopes on group 1 HA stalk. Antigen coating conditions for ELISA used in this study maintained important conformational epitopes of group 1 HA stalk as shown by ELISA using three well-known potent neutralizing monoclonal antibodies: CR6261, C179, and 70-1F02 ( Fig. 1A to D). Inhibition ELISAs were performed that measured inhibition levels of serum samples to CR6261, C179, and 70-1F02 antibodies and resulted in strong positive correlations (P Ͻ 0.0001* [statistically significant P values are indicated with an asterisk throughout]) observed between the stalk antibody titers and the level of inhibition to these monoclonal antibodies (Fig. 1E to G). These validation results confirm that the ELISA used on the clinical samples allows assessment of the level of antibodies binding to key epitopes of the influenza HA stalk.
The median age of clinical study volunteers was 27 years, and 48.6% were female and 51.4% were male. Of the volunteers, 43.2% were white, 48.6% were black, 6.8% were Asian, and 1.4% were American Indian. Of all subjects, 9.5% were Hispanic. No significant correlations were observed between prechallenge or postchallenge anti-HA stalk titers in relation to age, sex, or race.
The mean anti-HA stalk antibody titer for the 65 participants prior to influenza challenge was observed to be 64,843 (median, 60,221), with a minimum titer of 3,562 and maximum titer of 356,879. The lowest quartile of participants had a titer below 24,209, and the highest quartile had a titer above 83,585 (Fig. 2). Anti-HA stalk antibody titers increased after challenge in 64% of subjects. A statistically significant increase in mean titer was observed overall to a mean titer of 72,637 by week 8 (P ϭ 0.012*) ( Fig. 2A). Individuals with a prechallenge titer below the median titer demonstrated a larger significant increase from 28,010 to 57,116 (P Ͻ 0.0001*) (Fig. 2B), while those with prechallenge titers above the median titer demonstrated no significant change in mean titer (P ϭ 0.092) (Fig. 2C). Participants who developed mild to moderate influenza disease (MMID) (n ϭ 35) were found to have a significantly lower mean anti-HA stalk antibody titer prechallenge than those who did not develop MMID (n ϭ 30) (P ϭ 0.002*) (Fig. 3A). This was primarily driven by a difference in mean prechallenge anti-HA stalk antibody between those who developed detectable shedding (n ϭ 38) and those who did not (n ϭ 27) (P ϭ 0.002*) (Fig. 3B). There was no significant difference in anti-HA stalk antibody titers between those who developed influenza symptoms (n ϭ 55) and those who did not develop influenza symptoms (n ϭ 10) (P ϭ 0.58) (Fig. 3B). This difference in mean prechallenge titer in those who developed MMID versus those who did not develop MMID was similar to differences observed in HAI and NAI titers as well (Fig. 4).
Two variable correlation analyses demonstrated that anti-HA stalk antibody titer was negatively correlated with shedding duration (P Ͻ 0.001*) (Fig. 5A), but not to duration of symptoms (P ϭ 0.16) (Fig. 5B). Although there was a small but significant negative correlation to reduction in the number of symptoms (P ϭ 0.02*) (Fig. 5C), there was no correlation to reduction in symptom severity (P ϭ 0.22) (Fig. 5D). Previous multiple regression analysis to measure the independent effects of HAI and NAI on all four of these disease severity measures demonstrated that only increasing NAI titers had a statistically significant independent effect on decreasing severity of disease by all four of these clinical disease outcome measures (10). When this type of analysis was repeated in the current study with anti-HA stalk antibody titers included, a similar result was observed, demonstrating that only NAI titer was shown to be an independent predictor of a reduction in all four disease outcome measures assessed, while no statistically independent effect of either HAI or anti-HA stalk antibody titers was observed (Table 1). In our previous report, we identified individuals with a prechallenge HAI titer of Ͻ1:40 who had no increase in HAI titer after challenge with influenza (10). These individuals had baseline prechallenge anti-HA stalk antibody titers that varied greatly and included some in the top 25%, bottom 25%, and closer to the median (Fig. 6).
FIG 2 Legend (Continued)
confidence intervals for all participants who demonstrated a prechallenge titer less than the median prechallenge titer of 60,221. They had a larger increase postchallenge (P Ͻ 0.0001*). (C) Mean titer with 95% confidence intervals for all participants that demonstrated a prechallenge titer above the median titer (Ͼ60,221). These individuals demonstrated no significant increase in the overall titer (P ϭ 0.092). The dotted lines represent the bottom 25%, median, and top 25% of all prechallenge anti-HA stalk antibody titers. Statistically significant P values are indicated with an asterisk. Thirty-four percent (12 of 35) of the participants who experienced MMID had titers above the median titer. (B) Mean prechallenge anti-HA stalk titer in those participants who experienced influenza symptoms versus those who did not (P ϭ 0.58) and in those who experienced detectable viral shedding versus those who did not (P ϭ 0.002*). Black bars represent 95% confidence intervals. Statistically significant P values are indicated with an asterisk.
Anti-HA Stalk as an Influenza Correlate of Protection ® Compared to those with low HAI titer who did have HAI increases after challenge, there was no significant difference in mean anti-HA stalk titer prechallenge (P ϭ 0.629) (Fig. 6). A statistically significant increase in mean anti-HA stalk titer by week 8 was noted in those individuals who had HAI responses (P Ͻ 0.001*), while no significant increase was noted in the nonresponder group (P ϭ 0.063). In those who did not have increases in HAI antibody after challenge and had prechallenge anti-HA stalk titers below the median, many did not have an increased titer and their titer remained below the median titer or in the bottom 25% of titers (Fig. 6).
DISCUSSION
A healthy volunteer influenza challenge study afforded an ideal opportunity to perform an evaluation of naturally occurring anti-HA stalk antibodies in healthy individuals in a controlled setting. The ability to measure both pre-and postexposure polyclonal anti-HA stalk antibody titer and outcomes in subjects with a wide range of prechallenge antibody titers allowed us, for the first time, to measure anti-HA stalk antibody responses after infection/exposure to IAV. This enabled us to evaluate whether preexposure levels of these antibodies could predict protection from disease or modification of illness after intranasal challenge. This type of analysis is of critical importance to better understand how new influenza vaccines that may induce anti-HA stalk antibodies as their primary mechanism of action may perform and what pitfalls and hurdles may exist with this strategy.
The data generated here clearly demonstrate that many, if not all, healthy individuals likely have detectable levels of circulating anti-HA stalk antibody in their serum and that increases in titer after intranasal challenge and infection vary from person to person, similar to titers of anti-HA head antibody as assessed by HAI (Fig. 2). After challenge with an H1N1 virus, 64% of participants developed increases in their anti-HA stalk titer from their prechallenge baseline, demonstrating that natural generation of these antibodies in response to an IAV infection/exposure occurs in the majority of individuals after infection/exposure (Fig. 2). However, we observed variability not only in baseline circulating titers of anti-HA stalk antibody but also in the quantity of the postinfection/exposure change in anti-HA stalk antibody. Those individuals with the highest levels of anti-HA stalk antibodies before exposure to influenza developed smaller increases in titer, suggesting there may be a limitation to how high a titer of anti-HA stalk antibody could be achieved naturally. This type of antibody ceiling phenomenon is consistent with what has been observed previously, as there does seem to be a limit to how high anti-HA head antibodies will increase after traditional vaccination (11,12). In addition, variation between individuals was observed in terms of response and included a subset of individuals with a low prechallenge HAI titer and anti-HA stalk titer who did not have much increase in anti-HA stalk titer despite exposure/infection with IAV (Fig. 6). This is quite similar to the subset of individuals who demonstrated no increase in HAI titer after exposure to influenza reported previously (10).
These data suggest that antibody responses to HA stalk antigenic epitopes may be analogous to antibody responses to the HA head epitopes measured by HAI, namely, that there may be limited achievable levels and that those individuals who do not respond well by HAI to HA head antigens after infection/exposure/vaccination may also in some cases not respond well to HA stalk antigens. This would indicate that the limitations, inconsistencies, and pitfalls identified in response to current vaccines that primarily target antigenic sites on the HA head region may also still be problematic with new generations of anti-stalk vaccines. The fact that 10 to 30% of healthy individuals have poor HAI responses after vaccination, and that this response rate is even worse in the elderly (13), may mean that vaccines that solely target HA head or stalk may not be adequate to fully protect many of the individuals who are at risk of severe consequences of influenza. As a predictor of protection from or modification of disease, naturally occurring anti-HA stalk antibody titers displayed similar predictive qualities as anti-HA head antibody titers. As with baseline HAI and NAI titers, the mean anti-HA stalk antibody titer at baseline was higher in those individuals who did not exhibit MMID than in those who did (Fig. 3A). This difference in MMID was completely driven by a reduction in shedding, with a significant difference seen in mean baseline anti-HA stalk titer between those with detectable shedding and those without detectable shedding (Fig. 3B). This observation mimics anti-HA head antibody as measured by HAI as a correlate to reduction in MMID and detection of shedding as reported previously (10). In addition, there was no significant difference in the mean anti-HA stalk titer in terms of the presence of symptoms (Fig. 3B). Some individuals with very high baseline anti-HA stalk titers demonstrated clinically symptomatic influenza, with 12 participants demonstrating MMID despite having anti-HA stalk titers above the median titer and in the same titer range as those who did not experience MMID (Fig. 3A). This again is very similar to what was observed previously when evaluating baseline HAI titers in those with or without clinically symptomatic influenza disease (10).
FIG 6
Mean anti-HA stalk titer in those individuals with a prechallenge HAI titer of Ͻ1:40 who did not develop HAI increases after challenge and those who did. Of those individuals with low prechallenge HAI titers who did or did not have increases in HAI titer after challenge, we observed variable levels of prechallenge anti-HA stalk antibody titer and no statistical difference in prechallenge anti-HA stalk antibody titer (P ϭ 0.629). Those who responded after challenge by HAI demonstrated a significant increase in mean anti-HA stalk titer (P Ͻ 0.001*), while those who did not demonstrate a good HAI response did not (P ϭ 0.063). The dotted lines represent the bottom 25%, median, and top 25% of all prechallenge anti-HA stalk antibody titers. Black bars represent 95% confidence intervals. Statistically significant P values are indicated with an asterisk.
Anti-HA Stalk as an Influenza Correlate of Protection ® Baseline anti-HA stalk antibody titer was negatively correlated with shedding duration similar to baseline HAI and NAI titers (10) and also demonstrated a small but significant negative correlation with the number of symptoms (Fig. 5). No significant correlation was seen with duration of symptoms or severity of symptoms (Fig. 5). This suggests that similar to HAI titer, a higher baseline anti-HA stalk antibody titer can be used to predict that an individual will have a reduced duration of shedding of influenza virus, but not necessarily a significant reduction in duration or severity of illness, whereas NAI was previously demonstrated to predict a reduction of shedding duration, symptom duration, and a reduction in symptom severity (10). In addition, multiple regression with all three baseline antibody titers (HAI, NAI, and anti-HA stalk antibody) once again demonstrated that NAI titer was the only independent correlate that predicted a reduction in all four disease measures assessed ( Table 1), suggesting that NAI may be the best currently available predictor of a reduction of all aspects of influenza A virus clinical disease.
The naturally occurring anti-HA stalk antibodies measured in this study in response to infection likely reflected a polyclonal response, and what portion of these anti-stalk antibodies may be neutralizing and/or broadly reactive against other group 1 HA subtypes is not yet known, although inhibition ELISAs showed significant correlation to binding mapped epitopes recognized by well-studied monoclonal antibodies (Fig. 1). These naturally generated anti-HA stalk responses might be somewhat different from responses from some of the proposed anti-HA stalk-based vaccine strategies currently being evaluated, especially those that might seek to induce immunity to conserved peptide motifs rather than the intact HA stalk. Therefore, it is possible that these future vaccine strategies could potentially be more broadly reactive or induce a higher level of neutralizing anti-stalk antibody titers than those produced by natural infection. However, the current data do suggest that vaccine strategies designed to generate broadly neutralizing anti-HA stalk antibodies may have some limitations, many of which seem to be similar to those seen with current seasonal vaccines that primarily target anti-HA head antibodies and must be considered.
Conclusion. Anti-stalk-based vaccine strategies may be an improvement over current seasonal vaccines, as there are significant data to suggest that these vaccines could offer broader protection (5)(6)(7)(8)(14)(15)(16), but the data for humans is limited, and it is yet to be determined if these vaccines will be more effective in controlling annual influenza epidemics or able to prevent the emergence of pandemics. This study is the first study of humans to evaluate naturally occurring anti-HA stalk antibody as a correlate of protection and disease modification of IAV infection. In this study, we observed that individuals with some of the highest levels of preexisting anti-HA stalk immunity may still develop significant IAV illness and that anti-HA stalk antibody was similar to anti-HA head antibody in that it predicted a reduction in shedding but was not an independent predictor of modification of human IAV disease.
It is likely that including targets that induce anti-HA stalk antibody could play an important role in future broadly protective or universal vaccine strategies, but this analysis demonstrates that antibody responses to influenza virus vary and that protection is likely to be complex and multifactorial. Our belief is that future universal vaccine strategies should ideally focus on more than one target and/or aspect of immunity. In terms of antibodies elicited by vaccines, these data suggest that targeting the generation of not only anti-HA stalk antibody but also anti-NA antibody, observed to be the only independent predictor of a reduction of all aspects of IAV disease in this study, may augment protective efficacy. Careful consideration of the complexity of influenza immune protection and evaluation of all aspects of the anti-influenza virus immune responses will ultimately be necessary in the development of a successful broadly protective or universal influenza vaccine.
MATERIALS AND METHODS
Clinical study. A healthy volunteer challenge study was performed at the NIH Clinical Center, and healthy volunteers between the ages of 18 to 50 years were enrolled and intranasally inoculated with Park et al.
wild-type influenza A/H1N1pdm virus. This clinical study and the primary results were described in detail in a previous report (10). In this study, multiple clinical endpoints were measured including the presence or absence of mild to moderate influenza disease (MMID) defined as a positive molecular clinical test for influenza plus symptoms, symptom severity based on the inFLUenza patient-reported outcome (FLU-PRO) symptom assessment tool (17,18), presence or absence of symptoms or shedding alone, duration of symptoms, and duration of shedding. This study (clinicaltrials.gov identifier NCT01971255) was approved by the NIAID Institutional Review Board and was conducted in accordance with the provisions of the Declaration of Helsinki and good clinical practice guidelines.
Immunologic assays. Anti-hemagglutinin (HA) stalk antibody titers were determined for the 65 participants challenged with A/H1N1pdm using an enzyme-linked immunosorbent assay (ELISA) method as described below. Measurements of hemagglutination inhibition (HAI), neuraminidase inhibition (NAI), and viral shedding used for this analysis were detailed previously (10).
Production of ELISA stalk antigen. A group 1 influenza stalk construct without the HA head was produced based on a previously described method (4) with minor modifications. The group 1 HA stalk-only construct, designated construct #4900 by Impagliazzo et al. (4) was cloned into the pFastBac1 vector (catalog no. 10360014; ThermoFisher Scientific, USA) with the Strep-Tag II (Trp-Ser-His-Pro-Gln-Phe-Glu-Lys) sequence at the 3= end instead of the hexahistidine tag sequence. Recombinant baculovirus containing construct #4900 with Strep-Tag II (rBV_#4900_StrepII) was generated using the Bac-to-Bac baculovirus expression system (catalog no. 10359016; ThermoFisher). Sf-9 insect cells maintained in Sf-900 III medium (catalog no. 12658027; ThermoFisher) were infected with the rBV_#4900_StrepII at a multiplicity of infection (MOI) of approximately 10. Three days after infection, Sf-9 cell culture supernatant was harvested and clarified by centrifugation (3,000 ϫ g, 20°C, 10 min). The group 1 influenza stalk construct was purified using Strep-Tactin Sepharose (IBA GmbH, Germany). Concentration and buffer exchange to phosphate-buffered saline (PBS) was done using Vivaspin 20 (10,000 [10K]-molecular-weight cutoff [MWCO]) (catalog no. VS2001; Sartorius, Germany). A bicinchoninic acid (BCA) protein assay kit (catalog no. 23225; ThermoFisher) was used to measure the concentration of the purified stalk construct.
Anti-stalk antibody ELISA. A purified group 1 influenza stalk construct was diluted in PBS (1 g/ml) and added to a 96-well ELISA plate (50 l/well) (catalog no. 456537; ThermoFisher). The plates were incubated overnight at 4°C, and 100 l of blocking buffer (1% BSA in PBS) was added to each well. After incubation with the blocking buffer (room temperature [RT], 30 min), the plates were washed three times with wash buffer (0.05% Tween 20 in PBS), followed by blotting. Serum samples from patients were diluted 1:100 with the antibody diluent (1% BSA and 0.05% Tween 20 in PBS) and added to the washed plates (100 l/well). The dilution factor of 100 was determined by pretesting every serum sample to avoid saturation of the ELISA reaction. After incubation (37°C, 2 h), the plates were washed three times, followed by blotting. To make measurements of IgG bound to the group 1 influenza stalk construct, goat anti-human IgG antibody (catalog no. 62-8400; ThermoFisher) was labeled with horseradish peroxidase (HRP) using an HRP conjugation kit (catalog no. ab102890; Abcam, USA). The estimated concentration was 0.83 mg/ml. The HRP-conjugated anti-human IgG antibody was diluted 1:10,000 with the antibody diluent and added to the washed plates (100 l/well). After incubation (37°C, 1 h), the plates were washed six times and blotted, and the HRP substrate solution was added (100 l/well). The substrate solution was prepared by adding a 10-mg o-phenylenediamine dihydrochloride (OPD) tablet (catalog no. P8287; Sigma-Aldrich, USA) to 20 ml of a phosphate-citrate buffer preparation (catalog no. P4922; Sigma-Aldrich, USA). After incubation with the substrate (20°C, 30 min), 100 l of 1 M sulfuric acid was added to each well to stop the reaction, and the optical density at 490 nm (OD 490 ) was measured.
Calculation of anti-HA stalk antibody titers. Antibody titers were calculated using an extrapolation method. Four serum samples with the highest reactivity to the group 1 influenza stalk construct were preselected using the ELISA described. Equal volumes of these four sera were pooled, designated standard serum, and used to generate a standard curve for each plate. To determine the titer, the standard serum was 3-fold diluted from the initial 100-fold dilution. The OD 490 of each dilution was measured using the ELISA described, and the cutoff value was set as the mean OD plus 3 standard deviations (SD) of the wells containing secondary antibody only. The titer was defined as the highest dilution factor to produce an OD value above the cutoff value. The titer of this standard serum was 218,700. Serial dilutions of standard serum were added to each plate to generate a standard curve. The anti-HA stalk antibody titer of each sample was calculated from the sample OD 490 using this standard curve. The anti-HA stalk antibody titers were measured in triplicate, and means of the replicates were used for further analysis.
Validation of the anti-stalk antibody ELISA. To ensure the conformational integrity of the group 1 stalk antigen after hydrophobic binding to the ELISA plate, we performed a set of ELISAs using three well-known monoclonal antibodies binding to conformational epitopes on the stalk: CR6261 (Janssen, Netherlands), C179 (Clontech Laboratories, USA), and 70-1F02 (19). Monoclonal EM-4C04 antibody (19) binding to a globular head region of H1 HA was used as a negative control (70-1F02 and EM-4C04 were a kind gift from Rafi Ahmed at the Emory Vaccine Center). The ELISA plate was prepared as described above. Serial 10-fold dilutions of each monoclonal antibody were made with the antibody diluent and added to the washed plates (50 l/well). After incubation (37°C, 2 h), the plates were washed three times, followed by blotting. Antibodies bound to the stalk antigen were detected using HRP-conjugated anti-human IgG secondary antibody (catalog no. A18811; ThermoFisher) for antibodies CR6261, 70-1F02, and EM-4C04. HRP-conjugated anti-mouse IgG secondary antibody (catalog no. A28177; ThermoFisher) was used for C179. Each secondary antibody was diluted 1:10,000 in antibody diluent and added to the plates (100 l/well). After 1 h at 37°C, the plates were washed six times, followed by blotting, and the colorimetric signal from each well at OD 490 was measured as described above.
To analyze a correlation between the anti-HA stalk antibody titer measured and the inhibition level to stalk-binding monoclonal antibodies, we performed a set of inhibition ELISAs. ELISA plates were prepared as described above, and the prechallenge serum samples were diluted 1:25 and added to the plates (50 l/well) to block epitopes of the stalk antigen. Serum diluent, without serum, was added to 10 control wells (50 l/well) in each plate to serve as a baseline of monoclonal antibody binding level without inhibition. After incubation (37°C, 2 h), the plates were washed three times, followed by blotting. CR6261, C179, and 70-1F02 monoclonal antibodies were labeled with HRP conjugation kit (ab102890; Abcam), diluted in serum diluent, and added to the plates (50 l/well). The estimated concentrations of HRP-conjugated CR6261, C179, and 70-1F02 after dilution were 0.042, 1.042, and 0.042 g/ml, respectively. After incubation (1 h, 37°C), the plates were washed six times and blotted, and the levels of monoclonal antibody binding were colorimetrically measured at OD 490 as described above. The percent inhibition was calculated as 100 Ϫ (OD sample /OD control ϫ 100) where OD sample is the optical density of the sample and OD control is the optical density of the control. For the correlation analysis, samples were separately ranked by anti-HA stalk antibody titer and the percent inhibition to each monoclonal antibody. The samples with the highest values were given a ranking of 1, and the samples with the lowest values were given a ranking of 65.
Statistical analysis. Two variable correlations between the anti-HA stalk titer and the percent inhibition to each monoclonal antibody were performed using the nonparametric Spearman's correlation coefficient during the validation process of the anti-HA stalk antibody assay. Changes in anti-HA stalk antibody titer in human volunteers prechallenge to postchallenge were evaluated for significance using the Wilcoxon signed-rank test. Differences in mean anti-HA stalk antibody titers based on binary outcome measures used to evaluate influenza disease were evaluated using the Wilcoxon rank sum test. Two variable correlations of all four nonbinary clinical outcome measures to anti-HA stalk titer were performed using the nonparametric Spearman's correlation coefficient, and multiple regression analysis was performed using a linear model to examine independent effects of anti-HA stalk antibody and HAI and NAI titer on disease severity outcomes. All tests were two sided and at the 0.05 significance level. Statistical analyses were performed using R (version 3.0.1; R Development Core Team, Vienna, Austria) and GraphPad Prism software (version 7.0h; La Jolla, CA).
|
2018-04-03T02:02:59.934Z
|
2018-01-23T00:00:00.000
|
{
"year": 2018,
"sha1": "08a8be613b7e55f5a9590c36b11ab657879c5f41",
"oa_license": "CCBY",
"oa_url": "https://mbio.asm.org/content/mbio/9/1/e02284-17.full.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "08a8be613b7e55f5a9590c36b11ab657879c5f41",
"s2fieldsofstudy": [
"Biology",
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
220923701
|
pes2o/s2orc
|
v3-fos-license
|
Lean Cushing ’ s
This is a case report of 59 years old male with adrenocorticotropic hormone ACTH‐independent macronodular adrenal hyperplasia (AIMAH), which is a rare cause of Cushing’s syndrome. He was detected to have adrenal lesions incidentally on imaging. The biochemical evaluation was suggestive of endogenous ACTH‐independent Cushing’s syndrome. Contrast Enhanced CT of the abdomen done to characterize the lesion was suggestive of bilateral macro‐nodular adrenal hyperplasia. There was no evidence of cushingoid features except for type 2 diabetes mellitus suggestive of subclinical Cushing’s. He was not planned for any immediate medical/surgical management since metabolic control was achieved with oral anti‐diabetics (OADs) and anti‐hypertensive medications. He is on regular follow up with well‐controlled sugars and blood pressure.
Introduction
Adrenocorticotropin-independent macronodular adrenal hyperplasia (AIMAH) is a rare cause of endogenous Cushing's syndrome. [1] Most of the patients have only mild cortisol production and usually present as bilateral incidentalomas. [2] Multiple bilateral non pigmented enlarged macro nodules (greater than 1 cm) are seen in adrenal glands (each adrenal weighing from 25 to 500 mg) that produce excess cortisol independent of adrenocorticotropic hormone (ACTH) that leads to the development of Cushing's syndrome. [2] In routine practice, diagnosis is based on the clinical presentation, biochemical evaluation and imaging. However, it is confirmed by histopathology of the adrenal tissue. [3] Depending on the amount of cortisol excess management can vary from regular monitoring to surgical resection of the adrenal gland. Recently, it has been attributed secondary to aberrant hormone receptor thereby providing the potential for pharmacological therapies.
Case Report
59 year old male presented with complaints of nocturia from the past 3 months. There was no history of fever with chills, dysuria, dribbling after micturition, increased urgency or hematuria. He was known to have type 2 diabetes mellitus and hypertension that was well controlled on oral anti-diabetics (OADs) and anti-hypertensive agents. He had a history of diabetic ischemic mononeuropathy of the left lateral rectus which was managed conservatively. There was no history of allergy and substance abuse. Historically there was no complaint of osteoporotic fracture, depression, weight gain, irritability, or fatigue.
On physical examination, patient's weight was 66 kg, BMI 25.14 kg/m 2 , pulse rate of 80 beats per minute, blood pressure of 140/90 mmHg. Systemic examination was inconclusive.
On evaluation, his blood reports showed HbA1C of 5.7%, Na-134 meq/L, K-4.83 meq/L, Complete blood count, fasting lipid profile, and urine analysis were otherwise normal. Ultrasonography of the abdomen revealed bilateral grade-I renal parenchymal changes, prostatomegaly (45 g, postvoid residual volume 15 mL), fatty liver, and enlarged right adrenal gland (3 cm × 2.2 cm). In view of adrenal incidentaloma, the patient was further subjected to contrast-enhanced computed tomography (CECT) of the abdomen. As described in the image below CECT of the abdomen showed bilaterally enlarged adrenals with multiple nodules-likely macronodular adrenal hyperplasia possibility of AIMAH [ Figures 1 and 2].
Further biochemical evaluation was done to confirm the diagnosis of AIMAH. Morning serum cortisol was elevated, which ruled out exogenous consumption of glucocorticoids[ Table 1]. [4] After exogenous glucocorticoid consumption was ruled out, diagnostic tests were done to confirm hypercortisolism. The following three tests are commonly used. [4][5][6] The patient had elevated midnight cortisol, 24-h urine free cortisol (UFC) and lack of cortisol suppression after 1 mg dexamethasone suppression test (DST) [ Table 2].
Once endogenous Cushing's was confirmed, morning and midnight ACTH levels were done to look for ACTH-independent Cushing's syndrome. [6] As both the values were low, a diagnosis of ACTH-independent Cushing's was made [ Table 3].
Plasma metanephrines level was normal, which ruled out pheochromocytoma/paragangliomas.
Discussion
Endogenous Cushing's syndrome secondary to AIMAH is extremely rare, comprising less than 1% of all the cases. [2] This clinical entity has a bimodal age distribution. Few patients present during the 1 st decade of life, associated with McCune-Albright syndrome but most of the cases present during the 5 th and 6 th decade. [2] It is more prevalent in middle-aged women. Usually, patients present with Cushingoid features like obesity, facial plethora, moon face, purple striae, proximal muscle atrophy and skin pigmentation. But our patient is a rare case of subclinical Cushing's syndrome in which there was endogenous hypercortisolism independent of ACTH levels without any overt signs or symptoms of hypercortisolism. Hypertension and diabetes can be attributed secondary to hypercortisolism. The diagnosis is made using biochemical tests and imaging; which is confirmed by biopsy. [7] The treatment of AIMAH depends on the severity of hypercortisolism. In patients with mild cortisol excess, yearly biochemical assessment and CT scans are sufficient, in the case of disease progression steroid synthesis inhibitors like ketoconazole or even adrenalectomy can be considered. For patients with moderate to severe life-threatening hypercortisolism surgical resection of one or both the adrenals with postoperative hormonal supplementation is the treatment of choice. [1,2,8,9] Recently, it has been revealed that most of the patients with AIMAH and Cushing's syndrome have aberrant cortisol response to various endogenous hormones like vasopressin, gastric inhibitory polypeptide, serotonin, luteinizing hormone/ beta-HCG, etc., suggesting the presence of aberrant ectopic receptors. These patients can be treated with a pharmacological antagonist with regular monitoring of UFC levels. [2,[10][11][12][13] Our patient with subclinical Cushing's syndrome is on regular follow up and biochemical monitoring.
Conclusion
Patients presenting with common illness may have underlying rare disorders, which can be diagnosed if evaluated carefully.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient has given his consent for his clinical information to be reported in the journal. The patient understand that their names and initials will not be published and due efforts will be made to conceal his identity.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
|
2020-07-30T02:08:23.288Z
|
2020-07-01T00:00:00.000
|
{
"year": 2020,
"sha1": "6625a943ccd9bb574d29281aff0078860d7bb847",
"oa_license": "CCBYNCSA",
"oa_url": "https://doi.org/10.4103/jfmpc.jfmpc_230_20",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "12664f464c9655598e3a171848be1baa12a5c7d5",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
5807146
|
pes2o/s2orc
|
v3-fos-license
|
Enpp1: A Potential Facilitator of Breast Cancer Bone Metastasis
Bone is the most common site of breast cancer metastasis and once established, it is frequently incurable. Critical to our ability to prevent and treat bone metastasis is the identification of the key factors mediating its establishment and understanding their biological function. To address this issue we previously carried out an in vivo selection process to isolate murine mammary tumor sublines possessing an enhanced ability to colonize the bone. A comparison of gene expression between parental cells and sublines by genome-wide cDNA microarray analysis revealed several potential mediators of bone metastasis, including the pyrophosphate-generating ectoenzyme Enpp1. By qRT-PCR and Western analysis we found that expression of Enpp1 was elevated in human breast cancer cell lines known to produce bone metastasis in animal models compared to non-metastatic and normal mammary epithelial cell lines. Further, in clinical specimens, levels of Enpp1 were significantly elevated in human primary breast tumors relative to normal mammary epithelium, with highest levels observed in breast-bone metastasis as determined by qRT-PCR and immunohistochemical analysis. To examine the potential role of Enpp1 in the development of bone metastasis, Enpp1 expression was stably increased in the breast cancer cell line MDA-MB-231 and the ability to colonize the bone following intracardiac and direct intratibial injection of athymic nude mice was determined. By both routes of administration, increased expression of Enpp1 enhanced the ability of MDA-MB-231 cells to form tumors in the bone relative to cells expressing vector alone, as determined by digital radiography and histological analysis. Taken together, these data suggest a potential role for Enpp1 in the development of breast cancer bone metastasis.
Introduction
Metastasis is the ultimate cause of mortality in breast cancer patients, developing in the bone more frequently than any other site. Bone metastasis occurs in approximately 80% of patients with advanced breast cancer and causes considerable morbidity in the form of bone pain, pathological fractures, nerve compression, and life-threatening hypercalcemia. Sadly, bone metastasis is frequently incurable and the median survival time following diagnosis is just 2 years. Despite ongoing research efforts, the molecular and cellular mechanisms that regulate the development of bone metastasis and resultant osteolysis remain poorly understood. Identification of factors regulating this process may reveal novel targets for preventative and therapeutic interventions against this devastating disease.
In this study, we identified the ectoenzyme ectonucleotide pyrophosphatase/phosphodiesterase I (Enpp1) as being overexpressed in human primary breast cancer relative to normal mammary epithelium and provide the first evidence of its potential to foster the development of bone metastasis. Enpp1 is a type II transmembrane glycoprotein with pyrophosphatase and phosphodiesterase activity, expressed highly in bone and cartilage [1]. Enpp1 is located at 6q22-q23, a region reportedly amplified in breast cancer [2,3], and its mutation has been associated with several disorders including infantile arterial calcification [4,5], ossification of the posterior longitudinal ligament of the spine [6], and insulin resistance [7]. Biologically, Enpp1 is well known for its role in regulating bone mineralization, serving as the principal ectoenzyme responsible for the generation of extracellular inorganic pyrophosphate (PP i ) [1], a potent inhibitor of hydroxyapatite formation [8,9]. In addition, Enpp1 has been shown to modulate insulin signaling, recently reported to regulate bone acquisition and energy metabolism through effects on osteoblasts [10], while also functioning in purinergic signaling [11].
Association of Enpp1 with Breast Cancer Bone Metastasis
Previously, we utilized the murine mammary tumor cell line NT2.5 in an in vivo selection process to generate sublines having an increased propensity to develop metastasis to bone as compared to other sites [12]. To identify genes potentially important for the development of bone metastasis, microarray analysis was conducted comparing gene expression between parental NT2.5 cells and sublines having a moderately (BO3) and highly (BO6) increased preference for the colonization of bone following intracardiac injection. These microarray data have been deposited in the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) under accession number GSE47714. While many candidates were identified in this analysis [12], of particular interest was the pyrophosphate-generating ectoenzyme Enpp1, a central regulator of extracellular PP i levels, which have a substantial impact on bone physiology [13]. By microarray analysis, Enpp1 mRNA levels were incrementally increased from parental NT2.5 cells to BO3 and finally BO6 sublines (p = 0.008), which was confirmed in a separate experiment by qRT-PCR (Fig. 1A).
Expression of Enpp1 in Human Breast Cancer Cell Lines and Tissues
To determine whether the expression pattern of Enpp1 observed using the NT2.5 murine mammary tumor model translated to human breast cancer, we examined the expression of Enpp1 in a panel of human breast cancer cell lines with and without the ability to establish bone metastasis in animal models (our unpublished observations). Similar to our observation in NT2.5 cells and bone metastasis sublines, Enpp1 expression was elevated in human breast cancer cell lines that are capable of forming bone metastasis in animal models (MDA-MB-231/MDA-MB-468) as compared to those incapable of colonizing the bone (MCF-7/SKBR3), which displayed expression levels similar to that in immortalized normal mammary epithelial cell lines (MCF-10A/ HBL-100) (Fig. 1B, C). Extending our investigation to clinical samples, we found that Enpp1 mRNA expression was significantly elevated in primary breast tumors relative to normal mammary epithelium (p,0.05), with the highest levels observed in breastbone metastasis ( Fig. 2A). Correspondingly, immunohistochemical (IHC) analysis revealed strong Enpp1 protein expression in 32% (n = 37) of primary breast tumors while little to no Enpp1 expression was detected in adjacent normal mammary epithelium ( Fig. 2B, C). Further, Enpp1 was strongly expressed in 100% (n = 6) of breast-bone metastases (Fig. 2D). Importantly, the results of our IHC analysis not only correlated with our mRNA findings, but confirmed that Enpp1 was expressed in tumor cells as opposed to other cell types within these tissues.
Effect of Enpp1 on the Establishment of Bone Metastasis
To investigate the potential importance of Enpp1 in the establishment of bone metastasis, we utilized the MDA-MB-231 breast cancer cell line, which is capable of forming osteolytic bone metastasis in animal models and displayed moderate expression of Enpp1 by Western analysis (Fig. 1C). MDA-MB-231 cells were stably infected with a retroviral expression vector containing fulllength Enpp1 cDNA or with vector alone (EV). Levels of biologically active Enpp1 were increased by greater than twofold as determined by Western analysis (Fig. 3A) and nucleotide phosphodiesterase activity assay (Fig. 3B). Further, the rate of cell proliferation in MDA-MB-231-Enpp1 cells was equivalent to that in MDA-MB-231-EV cells as determined by MTS assay (data not shown). It should be noted that in an attempt to compliment this gain-of-function approach with a loss-of-function approach employing RNAi, we observed that all control shRNA tested, whether scrambled Enpp1 sequence or unrelated sequence, resulted in reduced Enpp1 expression, preventing use of this strategy and indicating that Enpp1 may be subject to off-target RNAi effects.
To determine whether increased levels of Enpp1 influence the formation of bone metastasis in vivo, both intratibial and intracardiac xenograft model systems were used. As expected, both the MDA-MB-231-Enpp1 and MDA-MB-231-EV group displayed a 100% overall incidence of bone metastasis. However, following both intratibial and intracardiac tumor cell administration, animals that received MDA-MB-231-Enpp1 cells demonstrated more rapid disease progression, evidenced by a significant increase in both osteolysis and tumor area as measured on radiographic images (Fig. 4A, C) and histological sections ( Fig. 4 B, D), respectively. Taken together, these data indicate that increased expression of Enpp1 can enhance the development of bone metastasis.
Although the mechanism through which Enpp1 enhances the development of osteolytic bone metastasis remains unknown, given its function as a nucleotide pyrophosphatase/phosphodiesterase, it may be linked to the production of extracellular inorganic pyrophosphate. The balance of PP i and inorganic phosphate (P i ) is critical to bone physiology, as PP i acts as an inhibitor of mineral deposition [14,15]. In addition, PP i has been shown to increase production of osteopontin in osteoblasts, another key mineralization inhibitor [16]. In the case of bone metastasis, it is possible that levels of PP i within the bone environment become elevated by the presence of tumor cells expressing Enpp1. This imbalance in PP i could lead to reduced mineralization, potentially creating an environment favoring bone resorption, which has been shown to be critical to the development of bone metastasis as it liberates growth factors from the bone, fueling tumor growth [17]. Beyond its effects on PP i /P i balance, the involvement of Enpp1 in insulin and purinergic signaling represent additional avenues worthy of investigation, all having the potential to yield novel insights regarding the interaction between tumor cells and the bone environment.
In summary, we have provided evidence that Enpp1 expression is elevated in human primary breast tumors relative to normal mammary epithelium, with highest levels observed in breast-bone metastasis. Further, consistent with the expression patterns observed in both murine and human breast cancer cell lines, increased expression of Enpp1 enhanced the development of bone metastasis in animal models. These data not only support a potential role for Enpp1 in breast-bone metastasis, but suggest that Enpp1 may be useful as a prognostic indicator for breast cancer.
Ethics Statement
All experiments were carried out in accordance with the National Research Council's ''Guide to the Care and Use of Laboratory Animals''. Animal use was approved by the Johns Hopkins Animal Care and Use Committee, animal welfare assurance #A3272-01, protocol #MO10M450.
Cell lines and tissues
The murine mammary carcinoma cell line NT2.5 and sublines, BO3 and BO6, were obtained as previously described [12]. NT2.5 parental cells and sublines were maintained in RPMI supplemented with 20% fetal bovine serum (FBS), 10 mM HEPES, 1% nonessential amino acids, 1 mM sodium pyruvate and 10 lg/ml insulin. All other cell lines were obtained from American Type Culture Collection (Rockville, MD) and cultured according to conditions specified. Mammary organoid samples, kindly provided by Dr. Saraswati Sukumar (Johns Hopkins University School of Medicine, Baltimore, MD), were prepared from reduction mammoplasty specimens of women with no breast abnormalities. Breast tumor and bone metastasis tissues were obtained from the Surgical Pathology Division of the Johns Hopkins Hospital following the approval of the institutional review board (IRB) of
Western analysis
Total protein was extracted from cell lines using lysis buffer consisting of 15% glycerol, 5% SDS and 250 mM Tris-HCl, pH 6.7. Equal amounts of protein were resolved using 10% SDS-PAGE. Protein was transferred to ECL nitrocellulose membranes (Amersham) and probed with anti-Enpp1 (Everest Biotech) and bactin (Sigma-Aldrich) antibody. Membranes were then incubated with horseradish peroxidase-conjugated antibody against rabbit IgG (Amersham) and binding was revealed by chemiluminescence (Amersham).
Immunohistochemistry
Paraffin-embedded sections were deparaffinized in xylene and rehydrated through graded ethanol. Antigen retrieval was achieved by immersing sections in 0.01 mol/L sodium citrate (pH 6.0) and heating in a steamer for 20 min. Sections were cooled to room temperature (RT), and endogenous peroxidase activity was quenched by immersing in 0.3% hydrogen peroxide. Blocking was carried out by incubation in diluted normal rabbit serum (Vector Laboratories) as per the manufacturer's instructions. Sections were then incubated with goat polyclonal anti-Enpp1 (Everest Biotech) at a 1:750 dilution for 16 hours at 4uC. Diluted biotinylated anti-goat IgG (Vector Laboratories) was added to the sections and incubated for 30 min. at RT followed by Vectastain ABC reagent. Enpp1 proteins were visualized using 3, 39-diaminobenzidine (DAB) as per the manufacturer's instructions (Vector Laboratories). Sections were subsequently washed in water and counterstained in hematoxylin (Richard-Allan Scientific). Images were acquired by light microscopy.
Nucleotide Phosphodiesterase activity assay
Cells were plated in triplicate in a 96-well plate at 3000 cells/ well and cultured in complete media for 72 hours. 1 mg/ml pnitrophenylthymidine monophosphate was then added to each well as a substrate for phosphodiesterase and the reaction was allowed to proceed for 1 hour at 37uC. Formed p-nitrophenol was then quantified by measuring the absorbance at a wavelength of 415 nm in a microplate reader.
In vivo assessment of bone metastasis and osteolysis
To determine the effect of Enpp1 on the establishment of bone metastasis, Enpp1 cDNA was cloned into the retroviral expression vector, pBabe-puro (Addgene plasmid 1764), and MDA-MB-231 cells were infected with pBabe-puro or pBabe-puro-Enpp1. Stable pools were selected in the presence of 1 mg/ml puromycin (Sigma-Aldrich) for one week. In separate experiments, tumor cells were injected into the right tibia (2.5610 4 cells) and left cardiac ventricle (1610 5 cells) of 5week-old female athymic nude mice (NCI-Frederick Cancer Research Facility). To monitor osteolysis, digital radiographic images were obtained once per week using a Faxitron MX-20 Xray unit (Faxitron X-ray Corp.) and osteolytic area was measured using MetaMorph image analysis software (Meta Imag-ing Series version 6.1, Universal Imaging Corp.). The experiment was terminated at 4 weeks when there was marked tibial osteolysis with or without soft tissue extension of the tumor in any group. To assess tumor area, tibiae were harvested, fixed in formalin for 24 hours, and decalcified in 10% EDTA, pH 7.4 for 48 hours. Paraffin-embedded sections (5 mm) were generated laterally throughout the tibiae at 100 mm intervals. A minimum of five sections was stained with hematoxylin and eosin, and light microscopic images were acquired. The total area occupied by intraosseous and extraosseous tumor was then measured using MetaMorph image analysis software.
|
2016-05-12T22:15:10.714Z
|
2013-07-05T00:00:00.000
|
{
"year": 2013,
"sha1": "8af9a4d5e1ea0632b45e3465518f54ba7474d032",
"oa_license": "CCBY",
"oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0066752&type=printable",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "8af9a4d5e1ea0632b45e3465518f54ba7474d032",
"s2fieldsofstudy": [
"Biology",
"Medicine"
],
"extfieldsofstudy": [
"Biology",
"Medicine"
]
}
|
250452795
|
pes2o/s2orc
|
v3-fos-license
|
Effects of High Gamma Doses on the Structural Stability of Metal–Organic Frameworks
Four different MOFs were exposed to γ rays by a cobalt-60 source reaching a maximum dose of 5 MGy. The results showed that the MIL-100 (Cr) and MIL-100 (Fe) did not exhibit obvious structural damage, suggesting their excellent radiation stability. MIL-101 (Cr) showed good radiation stability up to 4 MGy, but its structure started degrading with increasing radiation dose. Furthermore, the results showed that the structure of AlFu MOFs started to decompose at a gamma dose of 1 MGy, exhibiting a much lower tolerance to γ radiation. At this radiation energy, the dominant interaction of the gamma-ray with MOFs is the Compton effect and the radiation stability of MOFs can be improved by prolific aromatic linkers, high linker connectivity, and good crystallinity. The results of this study indicate that MIL-100 and MIL-101 MOFs have a good potential to be employed in nuclear applications, where relatively high radiation doses play a role, for example, nuclear waste treatment and radionuclides production.
■ INTRODUCTION
In recent years, increasing energy demand has renewed interest in nuclear energy, especially since it is an effective way to achieve carbon dioxide neutral energy production. 1 Nevertheless, nuclear waste is still a major concern and has resulted in studies aiming at better waste management in which separation or sequestration of the different radionuclides is essential, especially in the long-run. On top of that, the production of radionuclides of medical interest is also a major topic in the nuclear field. 2 Metal−organic frameworks (MOFs) materials have been investigated extensively as adsorbents for nuclear waste treatment as well as for radionuclide production, due to their high specific surface area, tunable functional groups allowing high selectivity, and good chemical stability. 3−6 For instance, several MOFs have shown outstanding performance in the field of radioactive gas separation ( 85 Kr, 129 I, 135 Xe, and 222 Rn), 7−11 seawater purification, 12−14 radionuclide adsorption for wastewater remediation ( 59 Fe, 65 Zn, 137 Cs, and 235 U), 15−24 and radionuclide production. 25 Although these MOFs have excellent chemical stability and have shown great potential in waste treatment application, their resistance to ionizing radiation has hardly been reported. To fully exploit the potential of MOFs in these fields, it is imperative to determine their radiation resistance. So far, the stability of only a few MOFs has been investigated under γ radiation. For example, A series of SIFSIX-3 MOFs have been studied under beta and gamma irradiation by Elsadi et al., who reported that SIFSIX-3-Cu had the best radiation resistance to gamma and beta radiation up to a dose of 50 kGy and 25 MGy, respectively. 22 The radiation stability of several MOFs with different metals (Al, Zr, Cu, Zn) have been studied under different gamma doses and the results have demonstrated that MIL-100 (Al) shows the best radiation tolerance, that is, 2 MGy. 23 Gilson et al. have developed a thorium-based MOF, which has survived γ radiation up to a dose of 4 MGy and a dose of α particles up to 25.5 MGy. 26 Furthermore, Nenoff et al. have investigated the influence of gamma dose rates (0.78 Gy/min, 3days and 423.3 Gy/min, 23 min) on the stability of NU-1000 and UiO-66, and found that NU-1000 exhibits better stability because of its high linker connectivity and lower node density. 27 However, the radiation resistance of many other promising MOFs and especially at higher γ radiation dose remains unknown.
MIL-101 (Cr), possessing high surface area and excellent chemical stability, has shown good potential to separate radionuclides. 28 In addition, this MOF can also be applied to produce the 51 Cr radionuclide, which is widely used to label platelets and red blood cells, and to evaluate their lifespan in clinical application and is desired radionuclide, requiring high specific activity (high activity per unit mass). Achieving high specific activity with nuclear reactors is hard but would be possible by the combination of Cr-based MOFs and hot atom approaches (Szilard-Chalmers effect). 29 In order to realize the application of MIL-101 (Cr) in a highly radioactive environment, the effects of γ radiation on its structural evolution need to be explored. At the same time, it is very interesting to determine the influence of organic linkers and metal clusters on the radiation stability of MOFs in a more systematic fashion allowing rational choice of a MOF according to its application. Therefore, MIL-100 (Fe), MIL-100 (Cr), and aluminum fumarate MOFs (AlFu MOFs) have been studied and compared. The structure of these MOFs and corresponding organic linkers are shown in Figure 1. All MOFs in this study have been irradiated by gamma-ray from 0 Gy to 5 MGy, and their structural changes were monitored by XRD, SEM, FT-IR and nitrogen adsorption. Powder X-ray diffraction patterns of the synthesized MOFs were obtained by a PANalytical X'Pert Pro pw3040/60 diffractometer with Cu Kα radiation operating at 45 kV and 40 mA. Brunauer−Emmett− Teller (BET) surface area of the samples was collected on a Micromeritics Tristar II at 77 K, and all samples were pretreated at 200 C for 15 h before measurement. Fourier transform infrared spectra (FTIR) of powder samples was directly measured by a PerkinElmer Spotlight 400 FT-IR spectrometer with a range of 650− 2500 cm −1 . The morphology and particle size of the samples were determined by scanning electron microscopy (SEM, JEOL, JSM-IT100). X-ray photoelectron spectra (XPS) was collected using a ThermoFisher Al K-alpha apparatus and scans were performed by a 400 μm spot size with an energy step size of 0.2 eV. The thermal stability of four MOFs was investigated by thermogravimetric analysis (TGA) using a Mettler-Toledo/STDA 851e apparatus with a heating rate of 5 C/min. Gamma Irradiation. The gamma source (GC220) is an irradiation cell using the radionuclide 60 Co (as shown in SI Figure S1), which was used to study the effects of gamma irradiation on the selected MOFs. The cobalt-60 (1.17, 1.33 MeV) source has a half-life of 5.272 years. We packed ∼0.2 g of powder of each sample in Posthumus plastic capsules in air and irradiated for different times to achieve different doses. The gamma dose was calculated according to eq 1-1: Where D and ḊḊare dose and dose rate, respectively. The initial dose rate was 0.65 kGy/h and λ
■ RESULTS AND DISCUSSION
To evaluate the radiation stability, the structure changes or the loss of crystallinity of the synthesized MOFs were monitored through XRD, SEM, FT-IR, and nitrogen adsorption after exposure to a cobalt-60 source at different doses of γ radiation. Figure 2 shows the XRD patterns of MIL-100 (Fe), MIL-100 (Cr), MIL-101 (Cr), and AlFu MOFs before and after the irradiation at different γ radiation doses. Their diffraction peaks are consistent with the simulated patterns for each material. After exposure to different γ radiation doses, the diffraction peaks of MIL-100 (Fe) (Figure 2(a)), MIL-100 (Cr) ( Figure 2(b)) and MIL-101 (Cr) (Figure 2(c)) maintained the same diffraction patterns. The full-width half-maximum (fwhm) of the most intense peak in XRD patterns was analyzed and shown in SI Figure S2. There is a small variation in the values of fwhm for MIL-100 (Fe) and MIL-100 (Cr). The values of fwhm for MIL-101 (Cr) increased slightly at relatively low radiation dose (from 0 to 4 MGy) and the value increased significantly (22%) when exposed to 5 MGy, showing severe loss of crystallinity. The diffraction peaks of AlFu MOF became broader as the gamma dose increased (Figure 2(d)). When the γ radiation dose was lower than 1 MGy, the fwhm of AlFu MOF remained stable. The value of fwhm increased from 0.4 to 0.51 after receiving 2 MGy of gamma irradiation, resulting in broader diffraction peaks.
The surface area of the four materials was determined by N 2 adsorption at 77 K. Figure 3(a) shows the N 2 adsorption− desorption isotherms of MIL-100 (Fe) after exposure to the different gamma doses. MIL-100 (Fe) possessed type I adsorption isotherm and had a surface area of 1574 m 2 /g. After exposure of 1 and 2 MGy, its surface area was 1527 and 1528 m 2 /g (having a 3% drop), respectively. When MIL-100 (Fe) was exposed to higher irradiation doses (between 3 and 5 MGy) its surface area decreased to 1513 m 2 /g, 1498 m 2 /g, and 1507 m 2 /g (as shown in SI Table S1).
The SEM images in SI Figure S3 show that MIL-100 (Fe) exhibited a rod-like shape with inhomogeneous size. There Langmuir pubs.acs.org/Langmuir Article were no detectable changes observed to the particle morphology and size after the different radiation exposures. The FT-IR spectrum of MIL-100 (Fe) can be found in SI Figure S4. The spectrum displays two peaks at 1625 and 1382 cm −1 that are attributed to asymmetric and symmetric vibrations of carboxyl groups, 34 respectively. The peak at around 1445 cm −1 is assigned to stretching vibrations of the O−C−O group. 35 No obvious changes can be observed in the FT-IR spectra, which is consistent with the XRD analysis. Figure 3(b) shows the N 2 adsorption isotherms of MIL-100 (Cr) after different gamma irradiation doses. The BET surface area of MIL-100 (Cr) calculated from nitrogen adsorption isotherm was 1862 m 2 /g before gamma irradiation. The shape of the adsorption isotherms was the same as that of fresh MIL-100 (Cr), but they were found to decrease gradually in adsorption capacity. Correspondingly, their surface area decreased by 3.2% (1802 m 2 /g), 6.3% (1744 m 2 /g,) 10.8% (1660 m 2 /g), 14.1% (1600 m 2 /g), and 18.9% (1510 m 2 /g) after receiving gamma doses of 1 MGy, 2 MGy, 3 MGy, 4 MGy and 5 MGy, respectively, where there was an approximately 4% decrease in surface area for each gamma dose. SEM images (SI Figure S5) showed that the morphology of MIL-100 (Cr) was unchanged. But the crystal size decreased a little after high γ radiation exposure (4 MGy and 5 MGy). The FT-IR spectrum of MIL-100 (Cr) was measured (as shown in SI Figure S6) and no obvious changes could be observed with γ radiation up to 5 MGy.
Nitrogen adsorption isotherms of MIL-101 (Cr) after exposure to the different gamma doses are shown in Figure 3(c). The BET surface area of the fresh sample was 2203 m 2 /g. After receiving gamma doses of 1 MGy, 2 MGy, 3 MGy, and 4 MGy, the BET surface area of MIL-101 (Cr) decreased from 2203 m 2 /g to 2159 m 2 /g, 2101 m 2 /g, 2072 m 2 /g, and 2099 m 2 /g, respectively (SI Table S1). The BET surface area of MIL-101 (Cr) was reduced by only 4.7% when exposed to a gamma dose of 4 MGy. When the gamma dose reached 5 MGy, MIL-101 (Cr) decreased by 20.0% in surface area (1762 m 2 /g). Its micropore volume decreased from 1.0715 cm 3 /g to 0.7508 cm 3 /g (30% reduction, as shown in SI Table S2). Additionally, the microstructure and morphology of MIL-101 (Cr) particles were examined by SEM, as shown in SI Figure S7. The images show that the MIL-101 (Cr) particles are irregular spheres with a relatively uniform distribution. The morphology and size of the particles did not show any obvious changes after exposure to different gamma doses. Figure 3(d) shows the N 2 adsorption isotherms of AlFu MOFs. The surface area of AlFu MOFs at 2 MGy had a significant decrease (63 m 2 /g) when compared to the original surface area of 1070 m 2 /g. SEM images in SI Figure S9 showed that the morphology of MIL-100 (Cr) was unchanged. The damage to the structure of AlFu MOFs was further confirmed by FT-IR spectra (see SI Figure S10). The C�C vibrations at 1600 cm −1 and O−H bending vibrations of the aluminum clusters 36,37 at 998 cm −1 could not be observed after 1 MGy radiation.
The radiation stabilities of four MOFs were explored by determining their most important characteristics (e.g., crystallinity). The small fluctuation value of fwhm and the lack of significant reduction in BET surface area (Figure 4a) after receiving γ radiation of 5 MGy indicate that MIL-100 (Fe) has excellent radiation stability toward γ radiation. After exposure to a γ radiation dose of 5MGy, the XRD pattern and fwhm of MIL-100 (Cr) did not have obvious changes, suggesting that it kept good crystallinity. Although a small decrease in surface area and micropore volume (Figure 4b) could be observed with increasing gamma doses, MIL-100 (Cr) showed tenacious resistance to gamma irradiation, exhibiting as good radiation stability as MIL-100 (Fe). The TGA curves also demonstrated that they still kept good thermal stability after receiving 5 MGy of gamma irradiation (see SI Figure S11). Furthermore, MIL-101 (Cr) showed the same performance as MIL-100 (Cr) when it was irradiated by γ rays at doses of 1, 2, 3, and 4 MGy. Its surface area and micropore volume had a slight decrease, but it also demonstrated that this MOF is highly resistant up to 4 MGy of a gamma dose. Subsequently, its surface area decreased significantly (20%) and the fwhm value also increased visibly, suggesting that the structure of MIL-101 (Cr) started decomposing at a gamma dose of 5 MGy, causing lower decomposing temperature (see the SI Figure S11(c)). In addition, the crystallinity of AlFu MOF began to degenerate after receiving 1 MGy of γ radiation and its pore structure characteristics completely disappeared at a gamma dose of 2 MGy, exhibiting much lower radiation stability, which could also be proved by the TGA curve (SI Figure S11(d)). Why the different MOFs react differently when exposed to γ radiation is not clear but there seem to be certain clues that can explain why one material is more stable than another. Three main processes rule the interaction of gamma-ray with matter, namely the photoelectric effect, the Compton effect, and the pair production. The probability of these interactions strongly depends on the atomic number (Z) and the energy of gamma-ray. 38 The Compton effect is expected to be the most dominant based on the gamma energy (Eγ = 1.17, 1.33 MeV) and the Z of the elements comprising the MOFs. The Compton process consists in a partial transfer of energy to an electron in the MOF resulting in the energy loss of the γ radiation, which can then further interact with other electrons, accompanied by the second Compton effect or photoelectric effect, and thus generating recoil electrons (Figure 5(a)). Some of the Compton electrons could travel to air without collision and the other electrons can collide with the orbital electrons of surrounding atoms in the MOFs by incoherent scattering, causing ionizations or excitations ( Figure 5(b)). Finally, the energy of the excited atoms can be dissipated through the emission of fluorescence or Auger electrons, and, in the last instance, through vibration (heat dissipation).
As discussed above, the structure of the MOF seems to affect its stability toward radiation. First, the mass-energy absorption coefficient is an effective index to measure the average fraction of photon energy absorbed by materials. Metal clusters of MOFs in this research consist of metal and oxygen atoms, which have a much higher total attenuation coefficient than organic linkers, suggesting that metal clusters can act as radiation antennas.
As shown in Table 1, Al metal atom had the lowest photon cross-section, but AlFu MOF had the worse radiation stability compared with the other three MOFs, which could be attributed to the lack of aromaticity of the organic linkers. The aromatic linkers can promote delocalization and migration of excitations based on high energy delocalization. 39,40 Therefore, the aromaticity of the linker has a significant impact on the radiation stability of MOFs under gamma-ray environment. Second, MIL-100 (Fe) and MIL-100 (Cr), which have the same linker and crystal structure, showed good irradiation stability. The surface area of MIL-100 (Cr) reduced a little bit more (16%), although Cr has a lower photon crosssection. The larger fwhm value of MIL-100 (Cr) indicated that it had a worse crystallinity compared with MIL-100 (Fe). The formed defects could be not beneficial for the energy transfer and dissipation, 41 resulting in low tolerance ability for γ radiation, which could be the reason for the better stability of MIL-100 (Fe). Figure 6 shows the XPS spectra of MIL-100 (Cr) before and after a γ radiation dose of 5 MGy. It can be seen that the sample contains Cr, O, and C elements according to the XPS survey ( Figure 6a). The Cr 2p spectrum is shown in Figure 6(b) and two peaks at 577.6 and 587.2 eV are ascribed to Cr 2p 3/2 and Cr 2p 1/2 , respectively. 42 The binding energy shifts cannot be observed suggesting that the metal clusters in MIL-100 (Cr) maintain integrity after exposure to a gamma dose of 5 MGy. The C 1s XPS (Figure 6(c)) spectrum of MIL-100 (Cr) before irradiation could be deconvoluted into four peaks at binding energies of 284.8 eV, 286.2 eV, 288.5 eV, and 290.6 eV, which were attributed to C−C/C−H, C−O, C�O, and O�C�O/π−π, respectively. 43 Apparently, the peaks ascribed to O�C�O/π−π and C−O positively shift to the binding energy of 291.6 and 286.3 eV, implying the decrease of electron density of the carboxyl groups. 44 The O 1s spectrum of MIL-100 (Cr) (Figure 6(d)) could be deconvoluted into three peaks at 531.9 eV, 534.2 eV, and 536.1 eV, which were ascribed to Cr−O−Cr, Cr−O−,C and O�C�O, respectively. 45 After gamma irradiation, the two peaks of Cr−O−C and O�C�O shift to higher binding energies by 0.2 and 0.1 eV, which were attributed to decreased electron density of the junction of metal clusters and organic linkers, indicating that the bonds between Cr−O clusters and the carboxylate of H 3 BTC linkers could be partly broken during irradiation, resulting in the formation of defects. Third, MIL-100 (Cr) and MIL-101 (Cr), which have the same metal clusters but Langmuir pubs.acs.org/Langmuir Article different connecting linkers, had good radiation stability under γ radiation of 4 MGy. However, MIL-101 (Cr) started to decompose with increasing gamma dose, which is related to the linker connectivity. Since each linker in MIL-100 (Cr) is connected with three metal clusters and the structure can still be kept when one or two connection sites are broken, causing this material to exhibit better stability. Another possible explanation could be that the metal nodes concentration per volume unit of MIL-100 (Cr) is slightly higher than that of MIL-101 (Cr), indicating that closer metallic atoms might attenuate more the generated electrons, which is instrumental in stabilizing the structure of MOFs.
To better understand the relationship between structural characteristics and radiation stability of the MOFs, more systematic experiments need to be carried out to explore other possible factors that determine the stability of MOFs. In addition, more studies should be carried out to determine the maximum radiation dose that the MOFs can tolerate without loss of their characteristic properties.
■ CONCLUSIONS
In conclusion, MIL-100 (Fe), MIL-100 (Cr), MIL-101 (Cr), and AlFu MOFs were prepared and irradiated using γ rays at doses ranging from 0 Gy to 5 MGy. The structure of all materials was characterized by XRD, BET, SEM, and FT-IR. The MIL-100 (Fe) and MIL-100 (Cr) presented outstanding stability when exposed to radiation of high doses (5 MGy). MIL-101 (Cr) exhibited good radiation stability when the material was subjected to gamma doses within a range of 0−4 MGy. A sudden decrease in the surface area demonstrated that MIL-101 (Cr) started to be damaged with increasing gamma dose. Meanwhile, the XRD results of AlFu MOF proved that the crystallinity of AlFu MOFs suffered a severe loss after receiving 1 MGy gamma dose. The BET and FI-IR results indicated that the structure of AlFu MOFs collapsed after exposure to high radiation dose. According to the structural analysis, the linker aromaticity plays an important role in the radiation stability of the MOFs. Additionally, high linker connectivity and good crystallinity of MOFs can also strengthen their radiation stability. To fully examine the potential of MOFs in nuclear applications their resistance to even higher doses should be assessed in the future.
Detailed information on materials, synthetic procedures, Cobalt-60 irradiation setup, SEM images, FT-IR spectra, TGA curves, (Tables S1 and S2) summary of the BET surface area and micropore volume (PDF)
|
2022-07-13T06:16:22.419Z
|
2022-07-11T00:00:00.000
|
{
"year": 2022,
"sha1": "99c2f4970ca82aedf14a9405153c50bc594a6a32",
"oa_license": "CCBY",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "4b118d395911b70b3ef89d5f6004d0709cbdebfa",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
52251549
|
pes2o/s2orc
|
v3-fos-license
|
Genetic Diversity of Cassava ( Manihot esculenta Crantz ) Genotypes in Ethiopia
Article No.: 061013658 DOI: 10.15580/GJAS.2013.9.061013658 The experiment was conducted at Jimma Agricultural Research Centre. The objective of the experiment was to evaluate the genetic diversity of cassava using agronomic traits so as to characterize and cluster with in collected cassava genotypes. Thirty five cassava genotypes were sampled from the collection. Relatively high broad sense heritability was observed for plant height (42.08) and root diameter (31.88) indicating the existence of possibility for selection of genotypes for high fresh tuber weight. The clustering of genotypes based on 11 quantitative traits revealed the existence of five distinct groups. The maximum inter cluster distance was observed between genotypes under cluster II and V and I and V, hence, the genotypes grouped in these clusters could be used for crossing if high fresh tuber yield genotypes are planned in breeding program. Submitted: 10/06/2013 Accepted: 22/09/2013 Published: 29/09/2013
INTRODUCTION
Cassava (Manihot esculenta Crantz) is one of staple food crop for millions of people who live in the world [1].It is particularly important for food security since many tropical areas often experience unfavourable environmental conditions [2].It is originated in the tropical Latin America [3] and was first introduced to Ethiopia by the British.After introduction, the crop has been found to have an excellent adaptation and growth performance in different agro-ecologies with productivity variation [4] which is by far greater than the global average tuber yield of 46t/ha [5].
Although reliable statistical information on the distribution and production of cassava in Ethiopia is lacking, the crop has been cultivated, particularly, in the South, South West, and Western parts.Its use as a potential food crop in Ethiopia has increased during and after the 1984 famine [6 and 4].The most important feature of cassava is its adaptability and produce reasonable amounts of yield in different areas especially on tropical environments and it often grows where most other crops fail [7].In south and south-western parts, there are huge amounts cassava genotypes are distributed across diverse agro-geographical areas that have not been properly evaluated before [3] and their attribute remains unknown by breeders [8].So, detailed descriptions of genotypes based on agronomical characters have tremendous impact on the conservation and genetic diversity of the crop.The present study, therefore, intended to evaluate the genetic diversity of cassava using agronomic traits so as to characterize and cluster with in collected genotypes for further breeding works.
Description of the Study Area
The experiment was conducted at Jimma Agricultural Research Center located 366 km south west of Addis Ababa situated at latitude 7 o 46' N and longitude 36 o E having an altitude of 1753 m.a.s.l.The soil of the study area is Eutric Nitosole with a pH of 5.3 that receives mean annual rainfall of 1432 mm with maximum and minimum temperature of 29.2 0 C and of 8.90 0 C, in 2010/11 growing seasons.These environmental conditions are conducive for production of cassava [1].
The Genotypes Evaluated
Thirty five genotypes of cassava were considered in this study.The genotypes were collected from south, west and south-western parts of Ethiopia, during 2004-2008 cropping seasons.The collections covered diverse agroecologies with an altitude range of 1170-1940 m. a.s.l, representing one of the major cassava production areas in the country.
Experimental Materials and Design
The experiment was laid in RCBD with three replications on spaced 1m x 1m and provided all agronomic practices as needed on growing period.Cuttings with 25-30 cm from 12 months aged cassava mother plants were used as a planting material on a ridge during onset of rainy season (early April).One month after planting, after the crop was well established, the plants were earthed up.Cultivation and weeding were carried out when necessary.The crop was harvested at 18 months after planting.
Data Collection
All the data were collected 18 months after planting as it have been suggested by [9].Accordingly, data on plant height(cm), number of main stem/plant, number of branch/plant, average canopy diameter/plant(m), average stem girth(cm), average number of roots/plant, average length of roots/plant(cm), average diameter of roots/plant(cm), root fresh weight (kg/plot), above ground biomass weight (kg/plant) and root dry weight (kg/plot) were recorded.Five plants were (400g each) randomly taken from the plot and were floured to get the dry matter yield of the product.After measuring the yield, the amounts are converted in to kg per plot.
Statistical analysis
Heritability in broad sense (h 2 B) and genetic advance as percent of means were calculated for all characters according to the method described by [10].Phenotypic coefficient of variation (PCV) and genotypic coefficient of variation (GCV) were estimated according to [11] while the Mahalanobi's generalized distance (D 2 ) statistics [12] was used for clustering of genotypes by assessing the divergence between genotypes for the traits measured by using SAS software statistical package [13].
Genetic Variance
The estimates of genotypic variance, phenotypic variance, broad sense heritability, genetic advance, genetic advance in percent of means and phenotypic and genotypic coefficients of variations are presented in Table 1.
The magnitude of phenotypic variance was higher than genotypic variance as the latter is a component of the former.The phenotypic variance (σ 2 p) is the sum of environmental variance (σ 2 e), genetic variance(σ 2 g) and their interaction(σ 2 ge).However, the phenotypic and genotypic variance values can not be used for comparing degrees of variability since different traits have different means across environments.For this reason the genotypic and phenotypic coefficients of variations were used.Comparatively, wider differences between genotypic and phenotypic coefficients of variations were observed for traits root dry weight kg/plot, weight of above ground plant parts kg/plant, canopy diameter/plant and number of vertical stem/plant while relatively narrow differences were observed were observed for traits plant height cm/plant, number of primary branch/plant and stem girth (cm).
This implies that, traits showed narrow differences between genotypic coefficients of variation and respective phenotypic coefficients of variations had somewhat low sensitivity to the environmental effects while those with wider differences were affected by environmental factors.
Heritability
Heritability estimates ranged from 1.78 for weight of above ground plant parts to 42.08 for plant height (Table 1).Maximum heritability was obtained for plant height, followed by root fresh weight, root diameter and number of branches per plant.Although yield is a complex character liable to have more environmental influence, heritability of root fresh weights per plot is the maximum under this study.For example, [14] found heritability of only 18.22% for tuber yield per plant in potato, which is very low as compared to the heritability obtained in this study even if the crop is different.So, further investigation could be undertaken in order to make sure such result.Heritability indicates the ease in which a trait can be improved through selection and could vary with materials studied and environments.Therefore, varieties that are early with high fresh root yield can be developed through selection.Heritability estimates of weight of above ground plant parts, canopy diameter, root dry weight and number of vertical stem per plant were low (Table 1).Low estimate of heritability for number of tuber /hill, tuber dry weight and tuber diameter was reported in previous studies [4, 15 and 16].
High heritability coupled with high genetic advance is an important factor for predicting the resultant effect for selecting the best individuals.In the present investigation, high heritability along with high genetic advance as percent of mean was obtained for root fresh weight per plot and plant height per plant.High GCV along with high heritability and high genetic advance will provide better information than single parameters alone [17] Hence, in this study, root fresh weight per plot and plant height per plant exhibited high genotypic coefficients of variation, high heritability together with high genetic advance as percent of means.This indicates that these characters would be very useful as a base for selection in Manihot esculenta improvement programs.
Genotype Clustering
In this study, genotypes were grouped into five distinct clusters with different sizes.The clustering patterns of the genotypes based on quantitative characters are presented in Table 2 The clustering pattern indicated that the number of genotypes in each cluster varied from one in cluster IV to 15 in cluster II.Cluster I, which comprised nine genotypes (25.71%).Genotypes in this cluster were predominantly characterized by highest number of vertical stem/plant and higher number of roots/hill (Table 2).Cluster II consisted of the maximum number of genotypes, accounting for about 42.85% of the total genotypes.
Genotypes in this cluster were predominantly characterized by largest number of branch/ plant and number of roots/hill (Table 3).
Figure 1: Dendrogram showing hierarchical clusters of 35 cassava Genotypes (UPGMA) based on quantitative characters
Cluster III had only eight genotypes (22.85%) originally from southwest Ethiopia; it was medium performance in all parameters of quantitative traits to other clusters.For example, it has an average of 150.43cm canopy diameter per plant whereas the cluster with the highest mean next to it has only 158.7cm in cluster V.The other clusters have much less number of canopy diameter (Table 12).
Cluster V had of two (5.71%) genotypes originally from Jimma; it was superior in almost all parameters of quantitative traits to other clusters.For example, it has an average of 14,040.0kg fresh root yield per plot whereas the cluster with the highest mean next to it has 8139.9kg fresh root yield/plot in cluster IV.All other clusters have much less than cluster V (Table 3).The cluster mean of all characters is presented in (Table 3) showed that genotypes falling in cluster IV and V showed highest mean performance for all characters of interest.For example, as tuber fresh weight, tuber dry weight, weight of above ground plant parts, canopy diameter and root length.In line with this, cluster II, which consisted of 15 genotypes was the higher mean performance for the majority of quantitative characters studied (Table 3).For example, the genotypes grouped under this cluster gave the higher average number of branches/plant, stem girth, number of roots/plant, root diameter and weight of above ground plant parts.This indicated that different clusters have different breeding values that enable breeders to improve different traits and parental selection should be made based on the relative merits of each cluster for each trait depending on the objective of the breeding program.[18] quoted by [15 and 16] further showed that while selecting genotypes from a particular cluster, the inter cluster distance; cluster mean performance should be taken into consideration.The pair wise generalized square distances (D 2 ) between the clusters (Table 4) showed that the distance between most of the clusters were highly significant (P < 0.01) suggesting diversity among Genotypes in different clusters.The maximum inter-cluster distance (D 2 = 13809.60)was noticed between cluster I and V followed by II and V (D 2 = 11666.10)and II and IV (D 2 = 4884.73)suggesting diversity between these groups.Hence, inter-mating between genotypes included in these clusters may give high heterotic response and thereby better sergeants in view of the genetic diversity.
Cluste
For instance, in this study, based on their tuber yield, selecting and crossing genotypes AAGT 108 (serial number1) from cluster I with genotypes 5632-8 (serial number 24) from cluster V and AAGT 150 (serial number 5) from cluster II with genotypes 30 from cluster V (Table 2) may produce desirable recombinants for high tuber fresh weight.However, in this study, there is no cluster that have low inter cluster distance (Table 2).Intensive selection for agronomically important characters and similarity in parentage might be the cause of narrow genetic diversity and uniformity between these clusters.However, [19] quoted by [4] and [14] indicated that, selection of parents should also consider the special merits of each cluster and each genotype within a cluster depending on the specific objective of hybridizations.In this study there was no clear grouping of the genotypes according to regions of collection.Genotypes of one region were classified into different clusters although some of them also belonged to the same cluster.Genotypes collected from different regions were also grouped into the same cluster.Therefore, it may be of considerable importance to enlarge the genetic base of Mainhot esculenta by sustainable and continual collection of genotypes throughout the growing areas of the country, conduct genetic improvement work to exploit desirable traits of the genotypes.
CONCLUSIONS
In this study, root fresh weight showed moderately heritability associated with higher genetic advance and higher genetic advance as means of percentage indicates the better chance of selection for varieties with high fresh root yield than those traits with moderate heritability but having low genetic advance such as weight of above ground plant parts, number of vertical stem/plant and diameter of root.The higher genetic distance observed in this study indicates the potential of improving root fresh yield of cassava by crossing those parents shows higher genetic distance which can yield hetrosis.
Cluster
Cluster
Table 1 : Estimates of Components of Variance, PCV, GCV, Heritability and Genetic Advance for 11 Quantitative Traits of Cassava Grown at Jimma. 2010/2011 PH=
Plant height, NS= Number of vertical stem/plant, NB= Number of branches/plant, CD= Canopy diameter, GR= Stem girth, NoRo= number of roots/plant, LR= Length of root, DR= Diameter of roots, RFW= root fresh weight, WAGP= Weight of above plant biomass and RDW= Root dry weight.
. Cluster means of 11 quantitative traits used for clustering are presented in Table 3.A dendrogram summarizing similarity among 35 genotypes of Manihot esculenta is given in Figure 1.
|
2018-09-12T17:20:11.754Z
|
2013-09-20T00:00:00.000
|
{
"year": 2013,
"sha1": "c71d2d825ea6bd5cde70bbea1d17848d30e3ea48",
"oa_license": "CCBY",
"oa_url": "https://zenodo.org/record/3377505/files/061013658%20Tewodros.pdf",
"oa_status": "GREEN",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "7e2eaaf98f18b0b4be2c8209d75e9d0f3013cb75",
"s2fieldsofstudy": [
"Agricultural and Food Sciences"
],
"extfieldsofstudy": [
"Biology"
]
}
|
92439586
|
pes2o/s2orc
|
v3-fos-license
|
GCIB-SEM: A path to 10 nm isotropic imaging of cubic millimeter volumes
Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) generates 3D datasets optimally suited for segmentation of cell ultrastructure and automated connectome tracing but is limited to small fields of view and is therefore incompatible with the new generation of ultrafast multibeam SEMs. In contrast, section-based techniques are multibeam-compatible but are limited in z-resolution making automatic segmentation of cellular ultrastructure difficult. Here we demonstrate a novel 3D electron microscopy technique, Gas Cluster Ion Beam SEM (GCIB-SEM), in which top-down, wide-area ion milling is performed on a series of thick sections, acquiring < 10 nm isotropic datasets of each which are then stitched together to span the full sectioned volume. Based on our results, incorporating GCIB-SEM into existing single beam and multibeam SEM workflows should be straightforward and should dramatically increase reliability while simultaneously improving z-resolution by a factor of 3 or more.
sectioning reliability while simultaneously improving z-resolution and maintaining compatibility with MultiSEM imaging.
In FIB-SEM a tightly focused (~1 µm) beam of high-energy (30 kV) gallium ions is directed at an almost parallel angle (< 1 o ) to the surface of a tissue block, ablating material while avoiding deeper damage. Repeated FIB/SEM cycles produce a 3D stack, but milling artifacts accumulate in the direction of the FIB beam due to the shallow milling angle used, and these limit the area which can be imaged 13 . We reasoned that replacing FIB with a more 'top-down' approach might overcome this, but the energy of the ions would have to be lowered considerably. We therefore decided to try GCIB milling, a lowenergy surface polishing method used in semiconductor fabrication and mass spectroscopy 15 . In GCIB, clusters of argon atoms are ionized and accelerated while a magnetic field selects clusters containing a particular number of atoms. In this way the average energy per atom can be tuned to just a few electron volts. Molecular dynamics simulations 16 show that such clusters 'splat' against the surface, scattering atoms at glancing angles thereby removing asperities.
Our prototype system consists of a GCIB gun (GCIB-10s from Ionoptika) mounted to a single beam SEM (Ultra SEM from Zeiss) (Supplementary Fig. 1). We first tested milling a variety of ultrathin tissue sections, exploring a range of angles, cluster sizes and energies, and embedding resins. We SEM imaged milled surfaces with 1.2 kV landing energy (chosen to provide adequate z-resolution) using energy selective backscatter (ESB) and in-lens secondary electron (InLens-SE) detectors. ESB (500 V filtering, see Methods) produced high quality images under virtually all milling conditions, but InLens-SE showed rough surfaces in some. InLens-SE detection is much more sensitive to surface topography and charging than ESB detection but approximates MultiSEM imaging 11 . Epon samples showed rough InLens images under all but the shallowest milling angles. Durcupan and Sprurr's samples gave high-quality InLens-SE images at glancing angles up to 30 o (Supplementary Fig. 2).
Sections thicker than 100 nm charged during imaging, and GCIB milling did not correct this (unlike FIB-SEM). Charging lessened with extended imaging and we reasoned that this was due to the slow conversion of the polymer embedding into a more conductive amorphous carbon 17 . We subsequently demonstrated that such electron irradiation could be used to make sections at least 10 µm thick sufficiently conductive to allow quality InLens-SE imaging (Supplementary Fig. 3,4). We found that a dose of ~0.5x10 27 eV/cm 3 was needed to completely eliminate charging. GCIB milling of irradiated sections remained sufficiently smooth to produce 3D volume images but the sputter removal rate of irradiated regions dropped by up to a factor of six (Supplementary Fig. 5). Our current 'optimized' protocol recommends irradiating to a dose of 0.5x10 27 eV/cm 3 , then milling with 10 kV Ar2000 clusters at a glancing angle of 30 o while the sample is rotating. Fig. 1a outlines this process.
To test, we sectioned a Durcupan-embedded Drosophila brain and collected three sequential 1 µm sections onto gold-coated silicon. We performed ~250 GCIB/SEM cycles (using InLens-SE detection) completely milling through all three sections (6x6x4 nm voxels). Fig. 1b shows images of one section at 200 nm intervals. Fig. 1c shows a Z-reslice through the image stacks of all three sections showing an unevenness at the bottoms of each section, the result of milling rate variance (~10%). We choose to collect on gold-coated silicon because the InLens-SE and ESB signals for gold are beyond the maximum signals produced by heavy-metal stain. Remarkably this 10% milling variance is sufficiently small and sufficiently spatially uniform that these variances could be corrected by software. We wrote algorithms to find and 'flatten' this unevenness (see Methods). Once flattened (Fig. 1d), we stitched the sections into a single volume (Fig. 1e). Membranes and synapses are clear and resemble those we typically see using FIB-SEM. A close inspection (Supplementary Video 1) shows some slight artifacts relative to FIB-SEM, particularly a light texture of milling streaks is visible in the Z-reslice.
This sample was embedded in Durcupan resin which appears to be insufficiently resilient to be sectioned this thick (diamond cut surface showed some damage), so we have now switched to Spurr's embedding which can be room-temperature sectioned with high quality surfaces at up to 1 µm, and can be hot knife sectioned at up to 25 µm. Fig. 2a shows a Z-reslice of a GCIB-SEM volume (using InLens-SE detection) covering three sequential 500 nm thick sections of Spurr's-embedded mouse cortex (8x8x6 nm voxels). Ultrastructural features like post synaptic densities and vesicles are clearly defined. Fig. 2b shows the simultaneously-acquired ESB volume which looks almost identical under these milling conditions. Cut surfaces were smooth and were easily flattened and stitched with an estimated loss between sections of ~30 nm (see Methods), and visual tracing of all neuronal processes appeared straightforward (Fig. 2c, Supplementary Video 2). These datasets demonstrate that wide-area GCIB milling and InLens-SE detection can be performed over multiple sequential thick sections and can produce datasets approaching the quality of FIB-SEM.
To address the applicability of GCIB-SEM to mammalian connectomics we sectioned Spurr'sembedded mouse cortex at 1 µm and collected ten sequential sections, then performed ~170 GCIB/SEM cycles (8x8x6 nm voxels) completely milling through all ten sections. We computationally flattened each section and stitched them together into the single volume shown in Fig. 2d, Supplementary Fig. 6-11, and Supplementary Video 3. The cut surfaces of these Spurr's-embedded sections were of high quality and we estimated the loss between sections at ~30 nm. A test segmentation of the ESB stack was performed using a flood-filling neural network 14 that had previously been trained on a quite different SBEM dataset and then 'bootstrap-trained' on this dataset (see Methods). Fig. 2e shows an example spiny dendrite spanning the ten section GCIB-SEM volume along with synapsing axons. Fig. 2f shows views through a single spiny synapse.
Cutting at 1 µm appears to be approaching the limit for room temperature sectioning of Spurr's since we saw significant surface damage when we attempted sectioning at 2 µm. Cutting much thinner (e.g. 150 nm) resulted in less loss and higher quality surfaces (see Methods), which may be preferred despite the fact that it results in more stitch boundaries. An alternative is hot knife ultrathick sectioning which uses a heated diamond knife and oil lubrication to minimize stresses on the tissue 18 .
To test this, we 'hot knife' sectioned Spurr's-embedded mouse cortex at 10 µm thickness and flat embedded two sequential sections against gold-coated silicon. The resulting GCIB-SEM volume (10x10x12 nm voxels) spanning the two hot knife sections is shown in Fig. 3a and a test segmentation based on the InLens-SE stack using a flood-filling network is shown in Fig. 3b. Supplementary Figs. 12-16 show Z-reslice views of the GCIB-SEM stack spanning the two 10 µm thick sections before and after flattening and stitching as well as additional segmentation examples. The cut surfaces were of high quality but the flat embedding procedure resulted in some contamination. We estimated the loss between sections at ~30 nm (see Methods). Supplementary Video 4 shows a side-by-side comparison of the ESB vs. InLens-SE volumes (also Fig. 2c,d).
We noticed that GCIB milling produces a slight drop in the InLens-SE membrane contrast which progresses over the first ~500 nm of milling. We speculate this is the result of a few nm thick layer of disrupted surface material persisting in equilibrium with milling which changes SE yield. Initial electron irradiation also produces a slight drop in InLens-SE contrast. These effects are quantified in Supplementary Fig. 17. We also verified that GCIB-SEM imaging was generally compatible with ATUM tape collection 9 by imaging two 1 µm thick sections collected on copper coated tape ( Supplementary Fig. 18).
Finally, we verified that the InLens-SE detection used in our prototype was an adequate proxy for MultiSEM imaging. We GCIB-SEM milled and imaged the first 250 nm of a 500 nm section of mouse cortex then shipped the half-milled section to Zeiss for imaging on their MultiSEM system. MultiSEM images of the GCIB-milled surface (acquired with similar electron dose) were essentially identical to our InLens-SE images of the same surface (Supplementary Figs. 19,20).
We see Serial Thick-Section GCIB-SEM as a potential route to 'industrial-scale' 3D EM and connectomics as it addresses key limitations of existing techniques: GCIB-SEM resolution is not limited by section thickness, and imaging dose is not limited by interactions with sectioning 19 , so resolution and dose can be adjusted as needed to achieve fully automated segmentation. GCIB-SEM utilizes thick sectioning which should be reliable even over arbitrarily large volumes. GCIB is dramatically less sensitive to beam position and focus than FIB, allowing lossless restarts. The gas cluster source, unlike FIB, does not require periodic reheating and is comparatively simple and reliable. GCIB-milling is widearea and fast (up to 450 µm 3 /s, 1 mm 3 /month, see Methods), and, if necessary, can be performed across multiple GCIBs in parallel with imaging when sections are spread across multiple wafers. This should allow close to 100% utilization of MultiSEM imaging time and should allow projected 20 MultiSEM improvements to be taken full advantage of. We envision a day when acquiring automatically segmented EM volumes of large tissue samples (e.g. whole invertebrate and small vertebrate brains) is routine; where such volumes are reliably thick sectioned and spread across multiple wafers which are then robotically shuttled between multiple GCIB-milling and MultiSEM-imaging stations in a setting reminiscent of today's semiconductor fabrication plants.
Online Methods: Sample preparation:
All experimental protocols were conducted according to US National Institutes of Health guidelines for animal research and were approved by the Institutional Animal Care and Use Committee at Janelia Research Campus. A three-month-old adult C57/bl6 mouse was euthanized with an overdose of isoflurane and immediately perfused via the heart with a buffered solution of 1.25 % glutaraldehyde and 4 % paraformaldehyde (0.1M PB pH 7.4). Two hours after perfusion the brain was removed and left in the same fix overnight at 4°C. Coronal sections of the brain were then cut with a vibratome (Leica VT1200). These were stained with 1.5% potassium ferrocyanide, on ice, followed by 2% osmium, each diluted in phosphate buffer. Sections were stained in 1% thiocarbohydrazide for 20 minutes before transferring them into 2% osmium tetroxide (Agar Scientific) for a further 30 minutes. They were then placed in 1% uranyl acetate at 4 o C overnight then in lead aspartate solution at 50 o C for 2 hours. The sections were finally dehydrated in increasing concentrations of alcohol, for 5 minutes each change and then transferred to increasing concentrations of epoxy resin (Spurs, EMS) until 100%. These were placed between two glass slides, coated with mold separating agent (Glorex, Switzerland), and hardened at 60 ° C for 24 h.
A six-day-old male adult Canton S G1 x w1118 fly of the Drosophila was GCIB-SEM imaged producing the micrographs shown in Fig. 1b-e. The preparation method used was a heavy metal contrast enhancement on C-PLT, an optimized preparation protocol for Drosophila brains with optimal contrast and morphological preservation for FIB-SEM. Chemical/ Progressive Lowering of Temperature (C-PLT) fixation/dehydration with en bloc staining is a modified conventional chemical fixation method 13,18 . Isolated whole brains were fixed in 2.5% formaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffer at pH 7.4 for 2 hours at 22 o C. After washing, the tissues were post-fixed in 0.5% osmium tetroxide in 0.05M sodium cacodylate buffer for 40 min and then treated with 0.8% potassium ferricyanide in buffer for 2 hours at 4 °C. After washing, tissue was incubated with 0.5% aqueous uranyl acetate for 30 min at 4 o C then followed by lead aspartate en bloc staining at 4 o C for overnight. A PLT procedure started from 1 o C when the tissues were transferred into 10% ethanol after secondary osmication with 0.8% osmium tetroxide for 20 min at 4 o C. The temperature was progressively decreased to −25 o C while the ethanol concentration was gradually increased to 97%. The tissue was incubated in 1% osmium tetroxide and 0.2% uranyl acetate in ethanol for 32 hours at −25 °C. After PLT and low temperature incubation, the temperature was increased to 22 o C, and tissues were rinsed in pure ethanol following by acetone, then infiltrated and embedded in Durcupan (ACM Fluka).
Sectioning, Irradiation, Milling and Imaging:
Drosophila brain (Fig. 1b-e): We sectioned a Durcupan-embedded Drosophila brain at 1 µm thickness on an ultramicrotome (Leica UC7) using a room temperature diamond knife (Diatome, 35 o clearance angle) and collected three sequential sections from the water boat onto a gold-coated silicon wafer. Silicon wafers used for this and other runs were pre-diced 5 x 7 mm p-type silicon support chips (Ted Pella). We deposited 5 nm of chromium (to promote gold adhesion) followed by 50 nm of gold onto the silicon surface using a Precision Etching and Coating System (Gatan). We then glow-discharged the gold surface just prior to collection to promote wetting. Electron irradiation parameters used: 10,000 μm 2 area, 2 hrs at 6 kV, 4 hrs at 10 kV, 1x10 27 eV/cm 3 total dose. Electron irradiation for this and other samples was performed using the SEM's beam set to its highest aperture, defocused and scanned across the area to be irradiated. GCIB parameters used: 22nA of Ar2000 at 10 kV spread over a 10 mm 2 area, 21 o glancing angle, 360 o rotation (3 steps), 900 s per mill cycle giving 4 nm of removal per image (44 µm 3 /s milling rate). GCIB milling for this and other samples was performed by slightly defocusing the GCIB beam (~0.5 mm) and scanning the area to be milled with a 50 x 50 position grid. SEM parameters used: 1.2 kV, 2 nA electron beam, InLens-SE detection, 6 nm pixels, 2 MHz acquisition rate.
Mouse cortex (Fig 2d-f) : We sectioned Spurr's embedded mouse cortex at 1 µm thickness on an ultramicrotome (Leica UC7) using a room temperature diamond knife (Diatome, 35 o clearance angle) and collected ten sequential sections onto a gold-coated silicon wafer. Electron irradiation parameters used: 10,000 μm 2 area, 1.6 hrs at 8kV, 2.4x10 26 eV/cm 3 total dose. GCIB parameters used: 22nA of Ar2000 at 10kV spread over a 18 mm 2 area, 30 o glancing angle, 360 o rotation (3 steps), 240s per mill cycle giving 6nm of removal per image (450 µm 3 /s milling rate). SEM parameters used: 1.2kV, 2nA, simultaneous ESB (500 V filtering) and InLens-SE detection, 8nm pixels, 1.25MHz acquisition rate. Fig 3): We 'hot knife' sectioned Spurr's embedded mouse cortex at 10 µm thickness using the method described in (Hayworth et al. 2015) 18 . Hot knife parameters used: Diamond knife (Diatome cryo, 25 o clearance angle) lubricated with 0.22 µm filtered tapping oil (Master Plumber), 65 o C knife temperature, no oscillation, 0.1 mm/s cutting speed. Two sequential sections were collected, washed of oil by dipping repeatedly in uncured Durcupan resin, and flat embedded against a gold coated silicon wafer (while still covered in uncured Durcupan) by sandwiching sections between the wafer and a glass optical flat that had a Kapton release film taped over it. This sandwich was put in a 65 o C oven to cure under a weight. This flat embedding procedure resulted in the two thick sections having a thin covering of Durcupan that had to be GCIB-milled to reveal their tissue surfaces (see text). A scalpel and laser ablation were used to remove large regions of the surrounding Durcupan 'flash' to prevent peripheral charging, then matching regions in each thick section were electron irradiated to make them conductive. Electron irradiation parameters used: 110,000 μm 2 area, 117 hrs at 30 kV, 0.9x10 27 eV/cm 3 total dose. GCIB parameters used: 22nA of Ar2000 at 10kV spread over a 13 mm 2 area, 30 o glancing angle, 360 o rotation (6 steps), 360 s per mill cycle giving 12 nm of removal per image (400 µm 3 /s milling rate). SEM parameters used: 1.2 kV, 2 nA electron beam, simultaneous ESB (500 V filtering) and InLens-SE detection, 10 nm pixels, 1.25 MHz acquisition rate. (MultiSEM test sample, Supplementary Figs. 19, 20): We sectioned Spurr's embedded mouse cortex at 500 nm thickness on an ultramicrotome (Leica UC7) using a room temperature diamond knife (Diatome, 35 o clearance angle) and collected a single section from the water boat onto a gold-coated silicon wafer. A region was electron irradiated to make it conductive, then GCIB-SEM imaged to a depth of ~250 nm before shipping to Zeiss for MultiSEM imaging tests. Electron irradiation parameters used: 188,000 μm 2 area, 89 hrs at 6 kV, 0.9x10 27 eV/cm 3 total dose. GCIB parameters used: 32 nA of Ar2000 at 10kV spread over a 8 mm 2 area, 30 o glancing angle, 360 o rotation (3 steps), 360 s per mill cycle giving ~12 nm of removal per image. SEM parameters used for GCIB-SEM imaging on Janelia Ultra SEM: 1.2 kV, 2 nA electron beam, InLens-SE detection, 8 nm pixels, 1.25 MHz acquisition rate.
Computational flattening and stitching of sections:
Processing of all GCIB-SEM runs proceeded in three steps (overviewed in Fig. 1): 1.) Align the SEM images of each thick section individually to generate a set of 3D volumes, one for each thick section. 2.) Computationally flatten the bottoms of each section individually. 3.) Align and stitch these 3D volumes together to form a final spanning 3D volume. We used the SIFT and BUnwarpJ alignment tools in the Fiji software 21 to perform steps #1 and #3. For step #2 we wrote a custom Matlab script to perform flattening.
To perform flattening we first identified the grayscale threshold that best separated tissue pixels from substrate pixels, which could be reliably distinguished by grayscale as long as the thick sections were collected on a gold substrate. Our flattening script used this grayscale threshold to determine, for each x,y pixel position, the image index (z-position) where the tissue broke through to the substrate, creating a "bottom index map". This "bottom index map" was lightly smoothed spatially and then used to stretch (with linear interpolation) each column of pixels in the z-direction so that all substrate breakthroughs would occur in a single z-plane.
We have provided this custom Matlab flattening script in the Supplementary Software along with an example dataset and a brief manual describing its operation.
Section-to-section loss estimation:
We estimated section-to-section tissue losses for each of the GCIB-SEM runs described in the text as well as for a test run performed with a series of 150 nm thick sections. Supplementary Figs. 21-25 summarize these results.
Our method for estimating section-to-section loss was as follows: First the z-thickness of each GCIB-SEM mill step was estimated by dividing the microtome-set cutting thickness by the number of mill steps spanning a thick section. Then each image in the final stitched GCIB-SEM stack covering multiple serial thick sections was first spatially filtered (using FIJI) with a 2D Gaussian (r = 4 pixels), and a matrix of mean pixel-wise image differences (absolute valued) spanning the full GCIB-SEM volume was computed using a custom Matlab script. We used this to plot the expected image difference vs. mill steps (shown in each supplementary figure). The images right next to the cut surfaces display differences in contrast and features relative to images in the interior of a stack. For example, the first few images of each thick section often have somewhat higher contrast which diminishes after milling (see Supplementary Fig. 17), and the last few images of a flattened stack show contrast and texture artifacts from the flattening algorithm where pixel grayscale values start to blend with the gold substrate. Looking at these 'boundary' images makes clear that they contain information crucial for tracing but directly computing the image difference between such 'top' and 'bottom' images would lead to an erroneously large estimated loss dominated by such contrast and flattening-artifact effects. Instead we picked images slightly removed from the boundary to compare and computed the loss based on these.
For example, the GCIB-SEM dataset shown in Fig. 2a-c spanned three 500 nm microtome cut sections. The flattened stack for the first thick section from this run contained 81 images giving an estimated mill step size of 500 nm / 81 = 6.2 nm. The computed image difference matrix is shown in SOM Fig. 22 along with a plot of the expected image difference vs. mill steps derived from looking at only 'interior' images. The boundary between the second and third thick section occurred between image #162 and #163 in this stack, but image #162 was not sufficiently clean for direct comparison due to the contrast and flattening artifacts described above. Instead we backed up to image #159 (which was clean) and compared it to image #163. The difference between these images (according to the computed mean difference matrix) was 14.5 units which corresponded to 8.7 mill steps according to the plot. This implies that there is an equivalent of 8.7 -3 = 5.7 mill steps between #162 and #163. But if there was no loss of tissue during sectioning then there should only be 1 mill step between #162 and #163. So the loss is equivalent to 5.7 -1 = 4.7 mill steps or 4.7 * 6.2 nm = 29 nm.
As described, the flattening procedure results in boundary images containing artifacts and, depending circumstances (for example, how sensitive one's segmentation algorithm is to noise), one may decide to discard one or more of the boundary images. To account for this Supplementary Figs. 21-25 display both the boundary image we based our loss estimate on and the next one that would be used if that one was discarded along with loss estimates for both.
Flood-filling network segmentation:
As described in the text, test segmentations were performed on an ESB-detected GCIB-SEM volume spanning ten 1 µm sections of mouse cortex which was acquired with 8x8x6 nm voxels ( Fig. 2d-f, Supplementary Figs. 6-11) and on an InLens-SE-detected GCIB-SEM volume spanning two 10 µm 'hot knife' sections of mouse cortex which was acquired with 10x10x12 nm voxels (Fig. 3, Supplementary Figs. 12-16). The two GCIB-SEM stacks were segmented with flood-filling networks (FFN) 14 . Instead of producing de novo dedicated volumetric training data for these GCIB-SEM volumes, an existing model trained on 10x10x25 nm SBEM data of zebra finch Area X tissue (provided by Jorgen Kornfeld) was used to bootstrap the segmentation, and object-based proofreading was utilized to correct errors as described next.
Using the ESB-detected GCIB-SEM volume spanning ten 1 µm sections, a prototype variant of the CycleGAN-based transfer procedure described in (Januszewski & Jain 2019) 22 was first applied between (a) 20x20x25 nm SBEM and 16x16x24 nm GCIB-SEM data (downsampled from its native 8x8x6 nm acquisition resolution), as well as 10x10x25 nm SBEM and (b) 16x16x12 nm, and (c) 8x8x18 nm GCIB-SEM data. Reduced-resolution downsampled variants of the GCIB-SEM volume were obtained via areaaveraged resampling of the full resolution data.
Segmentations obtained for multiple CycleGAN checkpoints from the (a) and (b) variants were screened for merge errors, and combined with oversegmentation consensus after upsampling to a common resolution. 12 neurite fragments were manually assembled from correct fragments in the base segmentation. These fragments were then used to train an FFN model for 16x16x12 nm GCIB-SEM data, with loss masking in 2 images before and after every 1 μm thick section seam. The segmentation from this new model correctly resolved the larger structures in the volume, but the model could not track finer processes due to insufficient downsample resolution of input data and a lack of representative training examples. To compensate for that, oversegmentation consensus between the FFN-generated reduced resolution segmentation, and the full resolution segmentation from SBEM-transfer (see (c) above) was computed, an additional 13 neurites were reconstructed by manual fragment assembly to form the training set for a full-resolution GCIB FFN model. Once trained, a segmentation was generated, 450 objects were proofread and selected for another round of training. All objects larger than 100k voxels were inspected in the final segmentation volume and split errors were manually corrected where necessary.
A similar procedure was applied to segment the second GCIB-SEM volume (InLens-SE-detected GCIB-SEM volume spanning two 10 µm 'hot knife' sections, acquired at 10x10x12 nm voxel resolution). First, the network trained on the first GCIB-SEM volume at 8x8x6 nm resolution was transferred to the InLens-SE 10x10x12 nm stack with the help of a CycleGAN. No attempt was made to correct the mismatched voxel resolution.
Candidate segmentations were generated for multiple checkpoints of the CycleGAN using forward-backward oversegmentation consensus 14 . The following iterative procedure was then applied to identify the best checkpoint. Starting with a random checkpoint, 3D meshes of objects in the segmentation were screened for mergers in descending order of object voxel count until 10 merge locations were found and added to a list of known mergers. The segmentations were then evaluated using all recorded merge locations, and the segmentation with the fewest mergers was screened again. Screening was terminated once a single segmentation without known errors remained, at which point 61 distinct merge errors were recorded. FFN agglomeration with standard settings 14 was then applied to the selected segmentation, with all merge pairs from agglomeration subjected to manual review. All objects larger than 100k voxels after agglomeration and touching 0 or 1 faces of the volume were then manually inspected, and if necessary connected to other objects within the volume. For the final segmentation, an FFN model was retrained on this proofread segmentation with the voxelwise loss computed only over voxels belonging to objects larger than 100k voxels.
The FFN network architecture and training procedures described in (Januszewski et al. 2018) 14 were used everywhere, but the field of view was extended to (33, 33, 33) voxels to account for the more isotropic resolution of the GCIB-SEM data, and the depth of the network for the second GCIB-SEM volume was extended to 12 residual modules. The manual inspections performed did not reveal any objects that could not be unambiguously traced. No manual voxel-level corrections were performed at any step of the reconstruction. Ar2000, 10kV). This is a time-of-flight plot made by pulsing the GCIB source while the beam is hitting a sample mounted at a known distance. The GCIB was tuned to give clusters containing an average of 2000 argon atoms each, but the plot shows that this beam actually contains a range of cluster sizes from mainly within the range of Ar1000 to Ar3000.
|
2019-04-03T13:11:29.379Z
|
2019-02-28T00:00:00.000
|
{
"year": 2019,
"sha1": "9dfd9d41d2ac614d9b23e40cb7992002ea055b6a",
"oa_license": "CCBY",
"oa_url": "https://www.biorxiv.org/content/biorxiv/early/2019/02/28/563239.full.pdf",
"oa_status": "GREEN",
"pdf_src": "BioRxiv",
"pdf_hash": "ed96b0abc4196a802272af266cc0b1992c872aed",
"s2fieldsofstudy": [
"Materials Science",
"Engineering",
"Biology"
],
"extfieldsofstudy": [
"Biology",
"Materials Science"
]
}
|
119061343
|
pes2o/s2orc
|
v3-fos-license
|
Stochastic De Giorgi Iteration and Regularity of Stochastic Partial Differential Equation
Under general conditions we show that the solution of a stochastic parabolic partial differential equation of the form \[ \partial_t u = \mathrm{div} (A \nabla u) + f(t,x, u) + g_i (t,x,u) \dot{w}^i_t \] is almost surely H\"older continuous in both space and time variables.
INTRODUCTION
Stochastic partial differential equations (SPDEs) arise in many pure and applied sciences. Regularity of solutions is of central importance for theoretical development as well as for numerical simulations. For linear equations, W k,2 -theory has been well developed (see Pardoux [13] and Rozovskii [15]). A more general W 2,p -theory has been established by Krylov [9]. Such equations can also be studied from a semigroup point of view (Brzèzniak, van Neerven, Veraar and Weis [1] and Da Prato and Zabczyk [14]). Results concerning nonlinear equations can be found in Debussche, De Moor and Hofmanova [5] and Pardoux [12]. In particular, many examples of quasilinear SPDEs with measurable coefficients can be found in the survey monograph edited by Carmona and Rozovskii [4]. Although an obviously important question in applications, regularity of solutions of quasilinear SPDEs does not seem to have been adequately addressed in the literature.
In this paper we consider the following type of quasilinear SPDEs on R n : (1.1) ∂ t u = div (A∇u) + f (t, x, u) + g i (t, x, u)ẇ i t , where {w i } is a sequence of independent standard Brownian motions on a filtered probability space (Ω, F * , P) and g = {g i } is an ℓ 2 -valued function such that for each fixed x and an F * = {F t }-progressively measurable process h, the process g(t, x, h t ) is also progressively measurable. We will show that almost surely a (stochastically strong) solution with L 2 -initial data is Hölder continuous in both space and time variables. The basic assumptions on the SPDE (1.1) are as follows: (1) uniform ellipticity: A(t, x; ω) is F * -progressively measurable and uniformly elliptic, i.e., there is a positive constant λ such that (2) linear growth: there exist a nonnegative function K ∈ L 2 (R n ) ∩ L ∞ (R n ) and a positive constant Λ such that We emphasize that no further conditions concerning the continuity A, f or g are imposed. A function u = u(t, x; ω) is said to be a (stochastically strong) solution of (1.1) if u ∈ C(R + ; W 1,2 (R n )) almost surely and satisfies the corresponding partial differential equation (PDE) in the sense that Here ·, · denotes the standard inner product on L 2 (R n ). The main result of the current work is the following regularity theorem.
Theorem. Let u be a solution of the SPDE (1.1) with a (deterministic) initial condition u(0) = u 0 ∈ L 2 (R n ). Then there exists a positive exponent α = α(n, λ, Λ) such that for all Remark 1.1. In general one needs to enlarge the underlying probability space in order for it to support a strong solution; see Carmona and Rozovskii [4] or Viot [16] for a detailed exposition.
The novelty of our result is that we do not impose any assumptions on the smoothness of A, f or g. Indeed, if A and g have some continuity, for example Dini continuity, then the above result follows directly from Krylov [8,9]. The approach we adopted in this work is quite different from the usual ones in the study of SPDEs. Largely motivated by the recent work of Glatt-Holtz,Šverák and Vicol [7], rather than relying on abstract or explicit estimates of the solution kernel, we analyze the energy of the solution by a combination of PDE techniques and stochastic analysis. Indeed, our work can be viewed as a stochastic version of De Giorgi-Nash-Moser theory. As such our flexible method is potentially applicable to other types of nonlinear SPDEs.
The paper is orgainzed as follows. In SECTION 2 we present a stochastic modification of De Giorgi's iteration. In SECTION 3 we prove the decay of the tail probability of the solution. The main theorem stated above is proved in the last section.
ACKNOWLEDGMENT. The authors are very grateful to Professor Nicolai Krylov for the method used in SECTION 4 of passing from the L ∞ -bound to Hölder regularity. The authors would like to thank Professors James Norris andÉtienne Pardoux for the electronic communications during the preparation of this work. The second author would like to thank Professors Luis Silvestre and Benjamin Gess for very helpful comments.
STOCHASTIC DE GIORGI ITERATION
De Giorgi's iteration is a classical method for studying heat equations with measurable coefficients. In this section we develop a stochastic extension of this method appopriate for the type of SPDEs under investigation. See Cafarelli and Vasseur [2,3] for an exposition of the classical theory without random perturbation.
Throughout the paper, an L p -norm without specifying a domain is implicitly assumed to be taken on R n ; thus K p = K L p (R n ) . For a time interval I ⊂ R + , we define the norm The norm most relevant for this paper is · 4,2,I .
, a sequence of time intervals shrinking from [0, 2] to [1,2]. For each a ≥ 1, write u k,a = (u − a(1 − 2 −k )) + and let U k,a := u k,a For simplicity we will denote f (t, x, u) and g(t, x, u) by f (u) and g(u), respectively. We have the following iterative inequality.
Proof. Hölder's inequality with the conjugate exponents (n + 1)/n and n + 1 gives Squaring (2.3) and integrating with respect to t on I k we have Applying Hölder's inequality again with the same conjugate exponents, we obtain The third factor on the right side is exactly U We claim that To prove this inequality we use the L p t L q x interpolation inequality Using this inequality with the parameters with p = 2(n + 1)/(n + 2) and q = 2(n + 1)/n we obtain (2.5) immediately. Applying the Sobolev inequality on R n to the second term on the right side of (2.5) and then substituting the result in (2.4), we obtain We now come to the key step of the proof, namely using Itô's formula to bound the terms involving the supremum over I k and the gradient of u. The function h r (u) = |(u − r) + | 2 is piecewise smooth with a single point of discontinuity and the process u(t) has the martingale part whose quadratic variation process is absolutely continuous with respect to the Lebesgue measure on R + . Thus Itô's formula can be applied to the composition h a k (u(t)) = |u k,a (t)| 2 and we have The uniform ellipticity assumption can be applied to the first term on the right side. For the third term, we observe that if u k,a ≥ 0, then 1 ≤ a ≤ 2 k u k−1,a and 0 < u ≤ u k−1,a + a ≤ (1 + 2 k )u k−1,a . By the linear growth assumption (2) on f and g and K ∞ ≤ 1, the third term is bounded by C k u k,a 2 2 dt for some C. Now, integrating (2.7) from t 0 to t with t 0 ∈ I k−1 \ I k and t ∈ I k gives Taking supremum over t ∈ I k , we have with X * k−1,α as defined in (2.2). By the mean value theorem, we can find a t 0 ∈ I k−1 \ I k such that Combining (2.6), (2.8) and (2.9), we obtain the desired iterative inequality (2.1).
Remark 2.2.
In the case n = 2, the proof in this section shows that for any µ ∈ (0, 1/3), there is a constant C = C(n, λ, Λ, µ) such that U k,a ≤ C k a 2µ U k−1,a + X * k−1,a U µ k−1,a . This is sufficient for estimating the tail probability of u ∞ in the next section, for all we need is that the factor U k−1,a carries an exponent strictly greater than 1.
ESTIMATE OF THE TAIL PROBABILITY
In the context of the stochastic De Giorgi iteration, controlling the size of u + ∞,[T,2T]×R n means estimating the decay of the tail probability P u ∞,[T,2T]×R n ≥ a . In order to use the iterative inequality in PROPOSITION 2.1 for this purpose we need to show that X * k−1,a is comparable with U k−1,a . This is accomplished in LEMMA 3.2 below, whose proof depends on the following simple result from stochastic analysis (see Norris [11, page 123]).
Lemma 3.1. Suppose that {M t } is a continuous local martingale. Then we have
Proof. There is a Brownian motion B such that M t = B M t , hence the event in the statement implies the event sup 0≤t≤b B t ≥ a/2 . Since sup 0≤t≤b B t has the same distribution as √ b|B 1 |, we obtain the inequality from the explicit density function of a standard Gaussian random variable.
Consider the continuous martingale X t :=ˆt 0 g i (u(s)), u k+1,a (s) dw i s and recall from (2.2) that X * k,a = sup 1−2 −k ≤s≤t≤2 (X t − X s ). Lemma 3.2. Assume that K ∞ ≤ 1. There exists a constant C = C(n, λ, Λ) such that for all positive α and β, and the desired estimate follows immediately from LEMMA 3.1. To prove (3.1), we start with which follows from the definition of X t . We observe that if u k+1,a ≥ 0, then 1 ≤ a ≤ 2 k+1 u k,a and 0 < u ≤ u k,a + a ≤ (1 + 2 k+1 )u k,a . By Minkowski's inequality (integral form) and the linear growth assuption (2) on f and g we have Integrating over the interval I k we obtain the desired inequality (3.1).
Armed with the iterative inequality (2.1) and the comparison result LEMMA 3.2 we are in a position to control the size of u + ∞, [1,2]×R n by estimating its tail probability. It is important that the constant M 0 in the following proposition is independent of a. Proposition 3.3. Assume that K ∞ ≤ 1. There exists a constant M 0 = M 0 (n, λ, Λ) such that for all a ≥ 1 and M > M 0 , Proof. As in the classical theory, we start with the observation that u + ∞, [1,2]×R n > a ⊂ G c a , where G a = {lim k→∞ U k,a = 0}. Consider the events E k = {U k,a ≤ (a/M) 2 γ k } for a constant γ < 1 to be determined later. Since u 4,2,[0,2] = U 0,a , it suffices to prove We estimate the probability P E c k ∩ E k−1 . For simplicity let δ = 1/(n + 1). Let α = (2C) k/2 M δ and β = a 2 γ k−1 /M 2 in LEMMA 3.2. If X * k−1,a ≤ αβ and U k−1,a ≤ β, then by the iterative inequality (2.1) in PROPOSITION 2.1 we have (after canceling a 2δ !) The last inequality holds if we choose γ sufficiently small such that (C 1 γ δ ) k (1 + (2C) k/2 M δ ) ≤ M δ for all k ≥ 1 and M ≥ 1 and then M sufficiently large such that γ 1+δ M δ ≥ 1. Now the above inequality implies that E c k ∩ E k−1 ⊂ {X * k−1,a > αβ, U k−1,a ≤ β}. Its probability is estimated by LEMMA 3.2 and we have Using this in (3.2) we obtain, again for sufficiently large M, This completes the proof of PROPOSITION 3.3.
HÖLDER CONTINUITY
In this section, we prove our main result, namely the almost surely Hölder continuity of the solution of the SPDE (1.1) subject to the conditions stated in SECTION 1. We start with the following moment estimate.
Proof. By scaling it suffices to consider the case T = 1, K 2 + K ∞ ≤ 1, and u 0 2 ≤ 1. We need to show that there exists a constant C (depending on p of course) such that As P is a probability measure, we may assume p ≥ 4. We start with the first inequality. Let where The solution of SDE (4.2) is explicitly given by By the assumptions we have F(t) ≤ 4Λ 2 and G t ≤ 2Λ 2 for all t ≤ 2, hence This implies the first inequality in (4.1). Next we show the second inequality in (4.1). Let for all a ≥ 1 and M ≥ M 0 , hence for a ≥ M 2 0 , assuming that M 0 ≥ 1. By the first inequality in (4.1), we have Hence, The second term is bounded by EY 2p , and the third term is finite by (4.3). This proves the second inequality in (4.1).
|
2016-01-10T23:33:37.000Z
|
2013-12-11T00:00:00.000
|
{
"year": 2013,
"sha1": "1e89a7561d50a76736278b864b90deaabf7e0b3c",
"oa_license": "implied-oa",
"oa_url": "https://doi.org/10.1214/16-aop1126",
"oa_status": "HYBRID",
"pdf_src": "Arxiv",
"pdf_hash": "1e89a7561d50a76736278b864b90deaabf7e0b3c",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
14305296
|
pes2o/s2orc
|
v3-fos-license
|
accepted for publication in the Astronomical Journal The 1000 Brightest HIPASS Galaxies: Newly Cataloged Galaxies
The HI Parkes All-Sky Survey (HIPASS) is a blind 21-cm survey for extragalactic neutral hydrogen, covering the whole southern sky. The HIPASS Bright Galaxy Catalog (BGC; Koribalski et al. 2002) is a subset of HIPASS and contains the 1000 HI-brightest (peak flux density) galaxies. Here we present the 138 HIPASS BGC galaxies, which had no redshift measured prior to the Parkes multibeam HI surveys. Of the 138 galaxies, 87 are newly cataloged. Newly cataloged is defined as no optical (or infrared) counterpart in the NASA/IPAC Extragalactic Database. Using the Digitized Sky Survey we identify optical counterparts for almost half of the newly cataloged galaxies, which are typically of irregular or magellanic morphological type. Several HI sources appear to be associated with compact groups or pairs of galaxies rather than an individual galaxy. The majority (57) of the newly cataloged galaxies lie within ten degrees of the Galactic Plane and are missing from optical surveys due to confusion with stars or dust extinction. This sample also includes newly cataloged galaxies first discovered in the HI shallow survey of the Zone-of-Avoidance (Henning et al. 2000). The other 30 newly cataloged galaxies escaped detection due to their low surface brightness or optical compactness. Only one of these, HIPASS J0546-68, has no obvious optical counterpart as it is obscured by the Large Magellanic Cloud. We find that the newly cataloged galaxies with |b|>10 are generally lower in HI mass and narrower in velocity width compared with the total HIPASS BGC. In contrast, newly cataloged galaxies behind the Milky Way are found to be statistically similar to the entire HIPASS BGC. In addition to these galaxies, the HIPASS BGC contains four previously unknown HI clouds.
Introduction
The 21-cm line of neutral hydrogen (H i) is unique as it can probe regions of the sky where no stars have (yet) formed (see Schneider 1996). Within individual galaxies, H i is frequently found well outside the optical radius (e.g. Meurer et al. 1996;Salpeter & -3 -Hoffman 1996), and many tidal tails or bridges between galaxies are only detected in H i (see e.g. Koribalski 1996; Ryder et al. 2001). Until now, the majority of H i observations were made of objects that had first been identified in the optical (or lately the infrared), thus imposing H i selection effects on top of already existing optical selection effects. Important H i structures like the Leo ring (Schneider 1989) and the Virgo cloud (Giovanelli & Haynes 1989) were discovered by accident and indicate the enormous potential for discovery in an untargeted H i survey.
The sky has been extensively surveyed for galaxies at optical wavelengths (e.g. Lauberts 1982) but severe limitations remain, mainly due to the foreground extinction of the Milky Way (which affects ∼25% of the sky, see e.g. Kraan-Korteweg & Lahav 2000). In many optical catalogs, including the input catalogs for optical redshift surveys, low-surface brightness (LSB) galaxies are easily missed and galaxies with diameters less than ∼1 ′ are often misclassified as stars. For example, all objects with brightness less than 1.15 times the sky and objects classified as stars were excluded from the input catalog of the Las Campanas Redshift Survey (Shectman et al. 1996). To supplement the galaxy catalogs, targeted searches for LSB galaxies (Schneider et al. 1990(Schneider et al. , 1992Impey et al. 1996Impey et al. , 2001Morshidi-Esslinger et al. 1999;Cabanela & Dickey 2000) and dwarf galaxies (Karachentseva & Karachentsev 1998;Drinkwater et al. 1999;Drinkwater et al. 2000) as well as deep optical searches for galaxies behind the southern Milky Way (Woudt & Kraan-Korteweg 2001) are being carried out. In the infrared less than 10% of the sky is affected by foreground extinction, surveys like the Two Micron All Sky Survey (2MASS, Jarrett et al. 2000) and the Deep Near-Infrared Survey of the southern sky (DENIS, Epchtein et al. 1997) are now cataloging large numbers of galaxies.
In contrast H i emission is not affected by extinction and enables us to identify many previously hidden galaxies. In addition, H i can easily be detected in LSB and late-type dwarf galaxies which are generally gas-rich (Impey & Bothun 1997). H i surveys complement optical galaxy catalogs and substantially improve the census of galaxies and measurement of the H i content of the local Universe. H i surveys also clarify voids by placing reliable upper limits on the mass of objects in these regions. Furthermore, there are some components of galaxies that have so far only been discovered in H i, e.g. high-velocity clouds (Putman et al. 2002), tidal H i clouds (e.g., HIPASS J0731-69, Ryder et al. 2001) and other nearby H i clouds (see e.g. Kilborn et al. 2000). Elliptical galaxies, which are typically H i-poor, are the main component missing from H i surveys (see e.g. Sanders, 1980;Knapp, Turner & Cunniffe, 1985) The current view of the large-scale structure in the Local Universe, with its filaments and voids, is based almost entirely on optical observations of high-luminosity galaxies. This view -4is highly selective and it will be interesting to see how large-scale surveys for extragalactic neutral hydrogen affect the current picture. Until recently, the Arecibo Dual-Beam Survey (ADBS) and the deeper Arecibo H i Strip Survey (AHISS) were the largest blind 21-cm surveys, covering areas of 430 and 65 deg 2 respectively. Rosenberg & Schneider (ADBS, 2000) detected 265 galaxies, of which 81 were uncataloged, whereas Zwaan et al. (AHISS, 1997; see also Zwaan 2000) detected 66 galaxies, half of which were uncataloged. With the advent of a 21-cm multibeam system at the 64-m Parkes telescope (see Staveley-Smith et al. 1996) as well as new observing and data reduction software (Barnes et al. 2001), much larger and deeper surveys are now possible. The H i Parkes All-Sky Survey (HIPASS, see e.g. Koribalski 2002) is the largest 21-cm survey for neutral hydrogen to date, covering the whole southern sky. With these surveys, extragalactic H i astronomy no longer depends entirely on observations at other wavelengths.
There is a large potential for detecting invisible H i clouds and uncataloged galaxies with unusual properties in HIPASS. We expect to find many uncataloged galaxies, either hidden behind the Milky Way with H i properties similar to the overall galaxy population or missed due to optical/infrared selection criteria. The former are important for the completion of our picture of the local large-scale structure as they bridge previously known structures that are optically intercepted by the Galactic Plane (e.g. Henning et al. 2000;Sharpe et al. 2001). The latter are equally important to enhance the completeness of galaxy catalogs across all morphological types.
Subsets of HIPASS within particular regions of the sky have already been analysed. In each of these regions uncataloged galaxies were discovered, many of which are also in our sample. Five uncataloged galaxies have been found in the Centaurus A group (Banks et al. 1999). The South Celestial Cap (SCC) region of the sky (δ < −62 • ) has been studied extensively by Kilborn et al. (2002;see also Kilborn 2001), who found 114 uncataloged galaxies (out of 536 galaxies in total). Banks et al. (1999) and Kilborn et al. (2002) searched the HIPASS data to full sensitivity, so only some of their galaxies will appear in the HIPASS Bright Galaxy Catalog (Koribalski et al. 2002). On-going analysis of the full-sensitivity HIPASS data over the total survey area is expected to reveal many more uncataloged galaxies. Henning et al. (2000) searched the H i Zone-of-Avoidance shallow survey (HIZSS; 212 • ≤ l ≤ 36 • , |b| ≤ 5 • ) and found 110 H i sources, 67 of which had no published optical counterpart (see also Staveley-Smith et al. 1998). An H i survey of the Great Attractor region (l = 300 • to 332 • , |b| < 5 • ) by Staveley-Smith et al. (2000) has so far revealed 305 galaxies, most of which were previously unknown (see also Juraszek et al. 2000). Complete analysis of deep Zone-of-Avoidance H i data is under way (Staveley-Smith et al., in prep.).
The H i properties of all 1000 galaxies in the HIPASS Bright Galaxy Catalog are presented by Koribalski et al. (2002). The optical properties of all previously catalogued galaxies in the HIPASS BGC are analysed by Jerjen et al. (2002). And the H i mass function for the HIPASS BGC will be discussed by Zwaan et al. (2002). Here we present the H i properties of 138 galaxies from the HIPASS BGC without velocity measurements prior to the Parkes multibeam surveys; 87 of these galaxies are newly cataloged -that is they do not have a cataloged optical (or infrared) counterpart listed in the NASA/IPAC Extragalactic Database (NED). The number distribution of the BGC galaxies presented in this paper are summarised in Table 1. In Section 2 we briefly describe the observations and the HIPASS BGC selection criteria as well as the method for optical (and infrared) identifications. In Section 3 we compare the H i properties of newly cataloged galaxies with low and high absolute Galactic latitudes. Section 4 contains the conclusions. A short description of all the newly cataloged galaxies with identified optical counterparts is given in the Appendix.
Observations & Selection Criteria
The H i Parkes All-Sky Survey (HIPASS) was conducted from 1997 to 2000 with the multibeam system on the 64-m Parkes radio telescope; it covers the whole southern sky in the velocity range from -1200 to 12700 km s −1 . For an overview of the survey parameters as well as data calibration and imaging techniques see Staveley-Smith et al. (1996) and Barnes et al. (2001). The HIPASS Bright Galaxy Catalog (BGC, Koribalski et al. 2002) is a subset of HIPASS and contains the 1000 H i-brightest sources in the southern sky (δ < 0 • ) based on their peak flux density (S peak 116 mJy). Although the total flux density of a galaxy (F HI ), which relates directly to its H i mass, is a more useful physical measurement, the peak flux density cutoff was applied as the observations were made in the spectral domain. Consequently this is not a total flux density limited sample.
The HIPASS BGC selection criteria are briefly described below. The following velocity ranges were searched for H i signals: −1200 < v < −350 km s −1 and 350 < v < 8000 km s −1 , i.e. omitting the range |v| < 350 km s −1 where confusion with high-velocity clouds (see Putman et al. 2002) makes it difficult to find galaxies. 2 Known galaxies with |v| < 350 km s −1 as well as HIZSS galaxies were added to the sample. The resulting cutoff, after fitting the H i parameters and selecting the 1000 brightest H i sources, is S peak 116 mJy. This corresponds to a typical minimum detection level of 9σ.
-6 -The galaxy finding process, selection criteria, fitting and analysis of the H i parameters and identification of cataloged optical counterparts are described by Koribalski et al. (2002). The FWHM of the gridded HIPASS beam is 15. ′ 5 and the velocity resolution is 18 km s −1 . The search for optical (and infrared) counterparts was conducted using NED. We define newly cataloged as any galaxy without an optical (or infrared) counterpart in NED. For all newly cataloged galaxies, images from the Digitized Sky Survey (DSS I & II) were searched for optical counterparts within an area of 15 ′ ×15 ′ centered on the H i position. Fig. 1 shows the separations between H i and optical positions. Galaxy positions from HIPASS are accurate to within a few arcminutes, depending on the H i peak flux density and source extent (Barnes et al. 2001). In addition, offsets from the optical position can occur intrinsicly due to multiple optical counterparts, asymmetries or warping of the H i. After completion of the identifications with NED late in the year 2000, references to a small number of the newly cataloged galaxies appeared in the literature, these are noted in Table 2. Since NED is a dynamic compilation, the counterparts presented here are valid for NED 2002, March 15th. Some galaxies may also be present in other catalogs not included in NED (e.g. The APM Sky Catalogue), these have not been searched.
Results and Discussion
There are 138 galaxies without velocity measurements prior to the Parkes multibeam H i surveys in the HIPASS BGC. Their distribution on the sky (Fig. 2) compared to all known HIPASS BGC galaxies, reveals -not surprisingly -most lie near the Galactic Plane. This is emphasised in the Galactic latitude histogram (Fig. 3). In the following we concentrate our study on the 87 newly cataloged galaxies listed in Table 2. For the analysis we divide the newly cataloged galaxies into two samples: there are 57 galaxies with |b| < 10 • (see § 3.1) and 30 galaxies with |b| > 10 • (see § 3.2). We discuss the H i properties of the two samples and derive optical properties where possible. In § 3.3 we briefly discuss the remaining 51 known galaxies without previous velocity measurements. A short description of the newly cataloged galaxies for which we have identified optical counterparts is given in the Appendix. Although H i parameters of the same HIPASS galaxies may vary slightly between catalogs, depending on the chosen fitting parameters, the original HIPASS name of each source is maintained for consistency and cross-identification purposes. Table 2 lists the H i properties of the 87 newly cataloged galaxies. Optical properties are given for those H i sources with one or more counterparts in the DSS . The columns are as follows: Column 1 -HIPASS Name; We adopt a uniform percentage error of 10% on all integrated H i flux densities. This is an empirical estimate based on a comparison of integrated H i flux densities of 620 galaxies in the LEDA database (see Koribalski et al. 2002). To calculate the H i mass, recessional velocities were corrected for the motion of the Sun around the Galaxy and the motion of the Galaxy in the Local Group. The correction used is the IAU convention, v LG = v sys + 300 sin(l) cos(b). The H i mass of each galaxy is then calculated using M HI = 2.356 × 10 5 D 2 F HI (Giovanelli & Haynes 1988), where F HI is the integrated H i flux density in Jy beam −1 km s −1 and D = v LG / H o is the distance in Mpc. We adopt a Hubble constant of H o = 75 km s −1 Mpc −1 .
The HIPASS BGC also contains four H i sources, which are most likely H i clouds; no obvious optical counterparts have been identified for these sources, and investigations as to their nature are under way. Three H i clouds are possibly Magellanic debris (Koribalski et al., in prep.) and lie within ∼10 • of each other, all with heliocentric velocities ∼400 km s −1 : HIZOA J1616-55 (Staveley-Smith et al. 1998), HIPASS J1712-64 (Kilborn et al. 2000, and HIPASS J1718-59 (Koribalski 2001). The fourth cloud, HIPASS J0731-69 (Ryder et al. 2001), is believed to be a tidal H i cloud associated with the NGC 2442 galaxy group.
3.1. Newly Cataloged galaxies with low absolute Galactic latitudes (|b| < 10 • ) There are 57 newly cataloged galaxies with absolute Galactic latitudes smaller than ten degrees. As expected, very few (14) of these have counterparts in the Digitized Sky Survey; their optical morphologies range from (Sc) spirals to magellanic irregular (Im) galaxies. These galaxies are mainly absent from optical catalogs because of Galactic foreground extinction. As expected, their H i properties are similar to the known galaxies in the HIPASS BGC. We expect a small number (∼3% -since we find 30 newly cataloged BGC galaxies with |b| > 10 • out of 1000) of these to suffer from both intrinsic low surface brightness and Galactic extinction. Indeed, the morphological type of some newly cataloged galaxies with |b| < 10 • is indicative of this (see Appendix and Table 2). Spectra of these 57 galaxies are given in Figure 14.
Newly Cataloged galaxies at high absolute Galactic latitudes (|b| > 10 • )
There are 30 newly cataloged galaxies with absolute Galactic latitudes larger than ten degrees. All but one of these galaxies have a potential optical counterpart. Twenty-five have a single optical counterpart and four have two or more possible counterparts. The one galaxy without a possible optical counterpart is HIPASS J0546-68, which lies behind the Large Magellanic Cloud (LMC). The field is too obscured to identify an optical counterpart in this case (see Dutra et al. 2001).
Galaxies have been morphologically classified within the extended Hubble system set out for giants by Hubble (1926;1927) and for dwarfs by Sandage & Binggeli (1984). The optical -9morphology of these galaxies (see Table 2) is dominated by magellanic spiral and irregular galaxies as well as Blue Compact Dwarf (BCD) galaxies. In terms of surface brightness we find most galaxies in two distinct groups: very compact sources of high surface brightness (e.g., HIPASS J0617-17 and J2200-56) and extended sources of low surface brightness (e.g., HIPASS J1106-14 and J1255-03). There are also a few galaxies that have both signatures, a bright core and a low surface brightness disk (e.g. HIPASS J1415-04A and J1424-16B). We conclude that the newly cataloged galaxies with |b| > 10 • are mostly absent from optical catalogs because of their small optical diameters or low surface brightness. The velocity distribution ( Fig. 10) shows that newly cataloged galaxies at high absolute Galactic latitudes follow the same general trend as all the newly cataloged galaxies, which is similar to that of the whole HIPASS BGC (Koribalski et al. 2002). But their H i mass distribution (see Fig. 11) is significantly shifted towards lower values. The median of the mass distribution shifts from log (M HI /M ⊙ ) = 9.4 for newly cataloged galaxies with |b| < 10 • to log (M HI /M ⊙ ) = 8.7 for newly cataloged galaxies with |b| > 10 • . A Kolmogorov-Smirnov test was performed and the distribution of H i masses from the two data sets very found to differ at the 99.6% level. The median H i mass of the entire HIPASS BGC is log (M HI /M ⊙ ) = 9.5, similar to that of the newly cataloged galaxies with |b| < 10 • .
In Fig. 12 we explore the H i profile shapes of the newly cataloged galaxies by comparing the measured 50% and 20% velocity widths. We find that most of the newly cataloged galaxies at high absolute Galactic latitudes have narrow H i profiles (mean w 50 = 64 km s −1 ). This value stands in sharp contrast to the equivalent parameter derived for newly cataloged galaxies with |b| < 10 • (135 km s −1 ). A recent survey of local dwarf galaxies (Huchtmeier, Karachentsev, & Karachentseva 2001) found a mean H i line width of w 50 = 66 km s −1 (from 98 H i detected dwarf galaxies, velocity resolution for most galaxies = 6.2 km s −1 ), which is similar to our value for newly cataloged galaxies with |b| > 10 • . Likewise, the standard deviation of w 50 for our sample is 38 km s −1 and Huchtmeier et al. (2001) find 49 km s −1 . Correcting for velocity resolution does not alter these results. Interestingly they find some galaxies with very narrow profiles (w 50 < 20 km s −1 , which would not be found by HIPASS. Narrow H i velocity profiles are indicative of either face-on spiral galaxies or low-luminosity galaxies, such as dwarfs. Given that this catalog is peak flux density selected, future HIPASS catalogs can be selected by integrated flux density and can contain newly cataloged galaxies with a different distribution of profiles, including those from highly inclined spiral galaxies. There are three galaxies, HIPASS J0403-01, J0605-14 and J1415-04A with |b| > 10 • , for which we measure w 20 200 km s −1 . The large 20% velocity width of HIPASS J0403-01 (v sys = 910 km s −1 ) is probably due to confusion with H i in and around NGC 1507 (= HIPASS J0404-02, v sys = 863 km s −1 , see Koribalski et al. 2002). HIPASS J0605-14 is potentially associated with a small group of galaxies. And HIPASS J1415-04A is an edge-on -10spiral galaxy, close to HIPASS J1415-04B.
Known galaxies with newly cataloged velocity measurements
There are 51 cataloged galaxies, in addition to the newly cataloged galaxies, with no velocity measurement prior to the Parkes multibeam H i surveys listed in Table 3. Of these, only 17 lie within 10 • of the Galactic Plane.
Four of the galaxies in Table 3 were previously only classified as infrared sources: • HIPASS J0747-26 (= HIZSS 022) is a very faint galaxy associated with IRAS 07451-2610. VLA H i snap-shot observations have been obtained.
Follow-up Observations
Follow-up H i observations of the newly cataloged galaxies as well as galaxy pairs/groups in the HIPASS BGC are under way with the Australia Telescope Compact Array (ATCA). The aim is to obtain accurate H i positions which will be used to check the candidate optical -11 -(and infrared) counterparts. Some examples are shown in Fig. 13 and Table 4. The table gives the ATCA H i position, the offset from the HIPASS position, the position angle of the detection and the total integrated ATCA H i flux density. The offset between the HIPASS and ATCA position is less than 2 ′ for all galaxies. In most cases the integrated ATCA H i flux density is ∼10-20% lower than the HIPASS flux density. This is typical for this type of observation since an interferometer filters out the more extended, diffuse H i emission. Such observations are particularly necessary for confused galaxies where it is not clear which of the optical counterparts are associated with the H i detection. Numerous H i follow-up observations have also been obtained by Kilborn (2001) and Kilborn et al. (2002). For example, SCC detection HIPASS J1004-73, also in our sample, has been observed with the ATCA (Kilborn 2001). It has a large (∼8 kpc) symmetric disk of H i surrounding the optical counterpart (∼2 kpc). Follow-up observations with the VLA have also taken place for some galaxies, e.g. HIPASS J0700-04 (Rivers 2000). Optical spectroscopy is planned to obtain redshifts for some of the newly cataloged galaxies.
Conclusions
A blind H i survey such as HIPASS provides a view of the local Universe free from optical selection effects. Although the potential for detecting previously unknown H i structures is high, we do not find any invisible H i clouds not gravitationally bound to any stellar system in the HIPASS Bright Galaxy Catalog (BGC). The four indentified H i clouds are most likely associated with Magellanic debris or other visible galaxies (The NGC 2442 group in the case of J0731-69). This can place important upper limits on the contribution of H i gas, not associated with galaxies, to the local baryon density. The HIPASS BGC has improved the census of galaxies in the local Universe by detecting galaxies behind the Milky Way. Furthermore, over the whole sky, we have easily detected galaxies missed in traditional optical surveys due to low surface brightness or misclassification as stars.
There are 87 newly cataloged galaxies in the HIPASS BGC, an additional 51 galaxies had no redshift measurement prior to the Parkes H i multibeam surveys. The majority (57) of the newly cataloged galaxies lie behind the Milky Way (|b| < 10 • ) and are missing from optical catalogs due to confusion or dust extinction. Optical counterparts are found in the Digitized Sky Survey for only 14 of these galaxies. Statistically, these 57 galaxies are found to have a similar H i mass distribution and velocity widths to the entire HIPASS BGC.
All the newly cataloged galaxies with high absolute Galactic latitudes (30) have a candidate optical counterpart(s) with morphologies ranging from late-type spiral to irregular, including four with multiple optical counterparts. The exception is HIPASS J0546-68, which -12lies behind the LMC and has no visible optical counterpart. The characteristic surface brightness of these galaxies is extreme, either diffuse low surface brightness or compact high surface brightness. Although these galaxies are H i-rich, they are not high in H i mass. The newly cataloged galaxies with |b| > 10 • on average have a lower H i mass (median log(M HI /M ⊙ ) = 8.7) and narrower velocity width (mean w 50 = 64 km s −1 ) than H i selected galaxies with optically cataloged counterparts.
Acknowledgements
• We are grateful to the staff at the ATNF Parkes and Narrabri observatories for assistance with HIPASS and follow-up observations.
• This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.
• SuperCOMOS Sky Surveys material is also acknowledged.
Appendix
Here we provide a short description of the newly cataloged galaxies for which optical counterparts have been identified. In addition to DSS I & II we also inspected 2MASS images, where available. Morphologically classifications have been assigned within the extended Hubble system set out for giants by Hubble (1926;1927) and for dwarfs by Sandage & Binggeli (1984). The BCDs classified below are only candidates and will need optical spectroscopy to confirm their morphology.
Newly Cataloged galaxies with low absolute Galactic latitudes (|b| < 10 • ) The LSB appearance of galaxies in this section is mostly likely due to foreground extinction. The Galactic foreground extinction in the photometric B-band, A B , is estimated from the IRAS DIRBE maps of Schlegel et al. (1998). Note that the extinction values from the DIRBE maps are uncalibrated at |b| < 5 • and may be unreliable.
HIPASS J0718-09 (HIZSS 006) must be a low surface brightness galaxy, as it is not easily discernible at the relatively low extinction level of A B = 1.9 mag. There are 2 extended patches of LSB emission visible on the DSS I image which are confirmed on the DSS II(R) image; one patch of ∼2. ′ 0×1. ′ 5 centered on 07 h 18 m 20.8 s , -09 • 03 ′ 20.2 ′′ , and a slightly smaller one of ∼1. ′ 0×1. ′ 0 centered on 07 h 18 m 14.5 s , -09 • 02 ′ 59.6 ′′ . Both together might define one very extended, face-on LSB source of up to 4 ′ . In either case the morphology is hard to determine. Type = Sd/Sm. HIPASS J0833-37 (HIZSS 045) is a galaxy with an angular extent of ∼25 ′′ ×20 ′′ with a bright bulge/nucleus and a small LSB envelope. It seems a bit small for the velocity (v sys = 958 km s −1 ) and the extinction (A B = 3.78 mag) but could have an obscured LSB halo. Type = Sm? or bulge/nucleus of earlier type galaxy.
HIPASS J0957-48 (HIZSS 060) is a spiral galaxy with an angular extent of ∼40 ′′ ×30 ′′ (A B = 2.5 mag). It consists mainly of a bulge with some LSB halo around it (one star very close to the center). The morphology is difficult to classify. Type = middle to late-type spiral.
HIPASS J1430-54 is an extremely LSB face-on spiral disk with a very small possible nucleus, visible but even fainter on DSS II(R) (A B = 2.9 mag). See also the ATCA image in Fig. 13. Type = Sc.
HIPASS J1451-50 has a small bright nucleus with a symmetric outer envelope (A B = 1.4). See also the ATCA image in Fig. 13. Type = Sm.
HIPASS J1522-49 (WKK4860) is a galaxy with LSB extended features, a bit clumpy on SRC-J film (A B = 2.6 mag). It is not visible on DSS I and very weak on DSS II(R). Its angular extent is 67 ′′ ×20 ′′ , see also Woudt & Kraan-Korteweg (2001). B = 16.6 mag. Type = Im. HIPASS J0447-57 is another LSB galaxy just to the North-West of the bright star HD 30804. It is possibly confused. Type = Im.
HIPASS J0532-67 is an early-type spiral galaxy which lies within the boundaries of the Large Magellanic Cloud (LMC). One can recognize a prominent bulge and a low surface brightness disk component. The light distribution is too regular for a BCD (see also the 2MASS image). This galaxy was also cataloged by Kilborn et al. (2002). The H i position coincides with the infrared sources 2MASXi J0531491-672133 and IRAS 05319-6723. Type = Sa or Sb.
HIPASS J0605-14 is associated with a group of galaxies (see Fig. 6) including two possible LSB optical counterparts. The positions and types of three optical galaxies are given in -15 - Table 2. The H i spectrum of HIPASS J0605-14 peaks quite sharply between 3000 and 3100 km s −1 . Additional low level emission is seen between 3100 and 3200 km s −1 . By integrating separately over the two velocity ranges we can associate the bright H i emission with the Imtype galaxy at the center, whereas the other two late-type galaxies are probably contained within the lower intensity H i gas envelope to the East.
HIPASS J0617-17 is a bright dwarf irregular with one bright H ii region; it is not visible in the 2MASS data. Type = Im/BCD. (Fouqué et al. 1990). Its H i flux density is F HI = 4.5 Jy km s −1 with a peak flux of ∼30 mJy, slightly too faint for a HIPASS detection. Interestingly, the Nancay H i spectrum of ESO375-G003 also includes HIPASS J1015-34. Both sources are part of the IC 2558 galaxy group (Hopp & Materne 1985). ATCA H i observations have been obtained. There is a small high surface brightness galaxy 2 ′ SW of the H i center. Type = BCD.
HIPASS J1118-17 is a compact, high surface brightness galaxy of irregular shape. Type = BCD.
HIPASS J1225-06 is possibly associated with two galaxies (see Fig. 7); a high surface brightness dwarf galaxy (LCRS B122316.1-061244) and another similar galaxy at 12 h 25 m 39 s , -06 • 33 ′ 08 ′′ . The H i profile is very narrow, suggesting a single galaxy, but there could be additional low level H i emission. Type = Im/BCD. HIPASS J1244-08 could be associated with several galaxies (see Fig. 8), although we note that the H i spectrum shows a typical double-horn spectrum indicative of a single gas-rich galaxy. The integrated H i distribution shows a point source. The narrow profile either indicates a face-on galaxy or a slowly rotating dwarf galaxy as the main component. The full HIPASS spectrum reveals no other H i sources at this position. There are at least four galaxies visible in the surroundings of the HIPASS position: 1) a spectacular, nearly face-on Sm galaxy at 12 h 45 m 13 s , -08 • 21 ′ 31 ′′ , 2) a small, but bright edge-on galaxy, possibly in the background, 3) an edge-on Sm galaxy at 12 h 45 m 08 s , -08 • 23 ′ 05 ′′ (the infrared counterpart is 2MASXi J1245078-082305), and 4) a Im/BCD galaxy at 12 h 45 m 04 s , -08 • 23 ′ 46 ′′ (NPM1G-08.0394). The latter two show some signs of interaction. Numerous small and faint galaxies are visible to the North of this group. H i synthesis imaging is needed to study these galaxies in more detail.
HIPASS J1247-77 is a nearby irregular, LSB dwarf galaxy. An ATCA H i image has been published by Kilborn et al. (2002;their Fig. 15). HIPASS J1247-77 has the lowest H i mass (∼ 5 × 10 6 M ⊙ ) among the newly cataloged galaxies in both the HIPASS BGC and the SCC sample (Kilborn et al. 2002). Type = Im.
HIPASS J1248-08 is a high surface brightness galaxy just East of the bright star HD 111310. It is also visible in the 2MASS image. The galaxy has a tiny bulge and a strong disk component. Type = late spiral, Sc.
HIPASS J1255-03 is an LSB dwarf irregular galaxy, not visible in the 2MASS image. Type = Im.
HIPASS J1258-33 is a late-type galaxy, similar to the LMC. Type = SBm.
HIPASS J1300-13B is similar to HIPASS J1258-33, except for an LSB extension of the disk to the North. Type = SBm(pec).
HIPASS J1337-39 is also a dwarf irregular galaxy in the Centaurus A group (see Banks et al. 1999). Type = Im.
HIPASS J1415-04B is a barred Sb or Sc galaxy. It has also recently been discovered by Colless et al. (2001;2dFGRS N145Z235, v sys = 2880±89 km s −1 ). A second galaxy, 2dFGRS N145Z228, closer to the H i position has a much higher velocity of 16912 km s −1 . The diameter is approximately 0. ′ 8 × 0. ′ 7. Magnitude = 14.8. Type = SBb/c. HIPASS J1424-16B is a late-type spiral galaxy. No bar or bulge is visible on the DSS -17 -II(R) image, but there is some evidence for a disk. Type = Sm/Im. HIPASS J1434-47 is a very LSB dwarf galaxy in a crowded field of stars. See also the ATCA image in Fig. 13. Type = Im.
HIPASS J2020-04 is a late-type spiral galaxy. Type = Sm/Im. Table 2). et al. 2002) in Galactic coordinates. The 87 newly cataloged galaxies (×) and the 51 known galaxies with no velocity measurements prior to the Parkes multibeam H i surveys (⋄) are marked. There are three potential optical counterparts. By integrating separately over the two velocity ranges (grey contours: 3000-3100 km s −1 ; black contours: 3100-3200 km s −1 ) we can associate the bright H i emission with the Im-type galaxy near the center, whereas the other two galaxies are probably contained within the lower intensity H i envelope to the East. The contour levels are at 60, 70, 80 and 90% of the maximum H i flux density (7.8 Jy beam −1 km s −1 ). Fig. 7.-a) H i spectrum of HIPASS J1225-08. b) DSS image centered on HIPASS J1225-08. The contour levels are at 60, 70, 80 and 90% of the maximum H i flux density (5.2 Jy beam −1 km s −1 ). There are two potential optical counterparts to HIPASS J1244-08 (see Table 2). Jy beam −1 km s −1 ). There are at least four potential optical counterparts to HIPASS J1244-08 (see Table 2). . There are three potential optical counterparts to HIPASS J1647-00 (see Table 2).
-25 - Fig. 10.-H i velocity distribution of the newly cataloged galaxies in the HIPASS Bright Galaxy Catalog (dotted histogram). The solid histogram shows the newly cataloged galaxies with |b| > 10 • . Fig. 11.-H i mass distribution of the newly cataloged galaxies in the HIPASS Bright Galaxy Catalog (dotted histogram). The solid histogram shows the newly cataloged galaxies with |b| > 10 • .
-26 - Fig. 12.-Comparison of the measured 50% H i velocity width, w 50 , versus the 20% H i velocity width, w 20 , for all the newly cataloged galaxies from the HIPASS Bright Galaxy Catalog. Newly Cataloged galaxies with |b| > 10 • are also marked (×). Total 138 Table 1: Number distribution of HIPASS BGC galaxies presented in this paper.
|
2014-10-01T00:00:00.000Z
|
2002-01-01T00:00:00.000
|
{
"year": 2002,
"sha1": "9dab9fa3ab8741607667b40306e5a75cf804755d",
"oa_license": null,
"oa_url": "https://espace.library.uq.edu.au/view/UQ:38537/UQ38537_OA.pdf",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "f3e12825dbd0f3b0095dfd7192ad28e7e7cb7659",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
263976602
|
pes2o/s2orc
|
v3-fos-license
|
Clinical features of isolated gestational proteinuria progressing to pre-eclampsia: retrospective observational study
Objectives Some women with isolated gestational proteinuria (IGP) later develop hypertension and are diagnosed with pre-eclampsia (PE). This study was performed to determine whether clinical features of such proteinuria preceding PE (P-PE) differ from those of other PE (O-PE). Design Retrospective observational study after approval of the institutional review board of ethics. Setting A single university hospital. Proteinuria was defined as a protein-to-creatinine ratio (mg/mg; P/Cr) of ≥0.27 in the spot urine specimen. IGP was defined as proteinuria in the absence of hypertension. P-PE was defined as PE in which proteinuria preceded hypertension by more than 2 days. Participants All of 10 and 18 consecutive women with P-PE and O-PE, respectively, who gave birth between January 2008 and August 2013. Results Proteinuria appeared earlier (at 30.2±3.0 vs 35.3±4.3 weeks, p=0.001), the P/Cr level was greater at birth (7.28±2.14 vs 3.19±2.49, p<0.001), net maternal weight gain during the last antenatal 1 week was greater (3.1±1.8 vs 1.3±1.7 kg, p=0.023) and length of pregnancy was shorter (32.5±1.9 vs 36.1±3.6 weeks, p=0.001) in women with P-PE than in O-PE. The duration of IGP was 10.0±5.9 days (range 3–20), and the time interval until delivery after diagnosis of PE was 6.1±8.2 days (range 0–23) in 10 women with P-PE. The P/Cr levels at birth were significantly inversely correlated with the antenatal lowest antithrombin activity and fibrinogen levels among the 28 women with PE. Conclusions Women with P-PE were likely to exhibit greater proteinuria in the urine, greater water retention in the interstitial space and more enhanced coagulation–fibrinolysis, thus suggesting that they may constitute a more severe form of PE than women with O-PE do.
INTRODUCTION
Pre-eclampsia (PE) remains a leading cause of maternal and perinatal mortality and morbidity. 1 Hypertensive disorders are responsible for approximately one quarter of all maternal deaths in Latin America and the Caribbean. 2 Hypertensive disorders encoded O-10 to O-16 by ICD-10 accounted for 8.1% and 4.4% of all maternal mortalities in Japan in 2005 and 2010, respectively. 3 The current criteria for diagnosis of PE are systolic blood pressure (SBP) ≥140 mm Hg or diastolic blood pressure (DBP) ≥90 mm Hg with proteinuria of ≥0.3 g/day. Thus, assessment of proteinuria is an important constituent of antenatal care to detect women with PE. Although proteinuria has Strengths and limitations of this study ▪ This study included all women with preeclampsia without known renal or autoimmune diseases who gave birth at a single centre during the period between January 2008 and August 2013. ▪ This study focused on the timing of the onsets of hypertension and significant proteinuria defined as a protein-to-creatinine ratio (mg/mg) ≥0.27 in the spot urine specimen, classifying women into two groups: proteinuria preceding pre-eclampsia and other pre-eclampsia. ▪ This study confirmed that a considerable number of women with pre-eclampsia (36% in this study) develop significant proteinuria prior to the development of hypertension. ▪ This study indicated that pre-eclamptic women with isolated gestational proteinuria constitute a severe pre-eclampsia group with respect to degrees of proteinuria, oedema and enhanced coagulation-fibrinolysis. ▪ This study demonstrated that women with proteinuria preceding pre-eclampsia were likely to give birth at an earlier stage of pregnancy compared to those with other pre-eclampsia. However, this may have been distorted by incidental bias due to the small size of the study population.
been considered as a late sign of the clinical course of PE, 4 it was recently demonstrated that some women exhibit proteinuria first in the absence of hypertension, so-called 'isolated gestational proteinuria (IGP)', and develop hypertension later and are diagnosed as having PE. 5 6 The IGP was proposed to be an early clinical sign of PE. 7 Then, the established PE risk factors, such as higher body mass index, younger age, nulliparity and twin pregnancy, were shown to be independently associated with the occurrence of IGP in late gestation in healthy term pregnancies, suggesting that IGP represents an early manifestation of PE. 8 As approximately 15-26% of women with new onset of hypertension alone in pregnancy develop proteinuria later and are diagnosed as having PE 5 9 ; there are two types of women with PE: those who develop proteinuria first significantly earlier than hypertension, and those who develop hypertension first or hypertension and proteinuria simultaneously. Frequent laboratory workup, at least twice in a week, is recommended in women with PE. 10 However, there have been no extensive studies regarding whether clinical and laboratory features of such women with proteinuria preceding PE (P-PE) differ from those of other women with PE.
METHODS
This study was conducted with the approval of the Institutional Review Board of Hokkaido University Hospital, a tertiary teaching hospital managing mainly high-risk pregnant women. Laboratory tests, including complete blood count, biochemistry and parameters of coagulation-fibrinolysis, such as antithrombin activity, are performed routinely in pregnant women visiting our clinic even for minor symptoms as well as those admitted to the hospital for management of various obstetric and incidental complications.
Definition of terms used in this study and study protocol Pregnant Japanese women receive 14-15 antenatal care visits before delivery and undergo determination of blood pressure, proteinuria by dipstick testing and body weight at each visit. 11 A spot urine specimen with ≥1+ on dipstick test at antenatal visits was used for determination of protein and creatinine concentrations in this study. Proteinuria was defined as a protein-to-creatinine ratio (mg/mg; P/Cr) ≥0.27 in the spot urine specimen, and IGP was defined as proteinuria in the absence of hypertension. Hypertension was defined as the occurrence of SBP ≥140 mm Hg and/or DBP ≥90 mm Hg on at least two occasions recorded more than 12 h apart. Gestational hypertension (GH) was defined as hypertension occurring on and after gestational week (GW) 20 in the absence of proteinuria. PE was defined as the presence of hypertension and proteinuria. The GW at new onset of hypertension and proteinuria was specified in each patient, and each woman with PE was classified into either type of PE according to timing of the onsets of hypertension and proteinuria: P-PE in which the duration of IGP was 3 days or more and other PE (O-PE) in which the duration of IGP was 2 days or less. Subanalysis was performed in women with O-PE classified into two groups: those with a duration of IGP or GH of 2 days or less (simultaneous PE, S-PE) and those with a duration of GH of 3 days or more (hypertension preceding PE, H-PE). Women with IGP or GH were followed up once or more per week on an outpatient basis in principle, but underwent laboratory workup at each visit, including complete blood count, coagulation-fibrinolysis and blood chemistry. Women with a diagnosis of PE were admitted to our hospital and underwent determination of blood pressure at least three times a day, body weight daily and laboratory workup, including blood tests and P/Cr test, at least twice a week.
Patient selection
A total of 1665 women gave birth on or after GW 22 at Hokkaido University Hospital during the study period between January 2008 and August 2013. A total of 28 PE women without known renal or autoimmune diseases were abstracted from the database of discharge records. We were provided data from the institutional central laboratory regarding blood and urine test results for these 28 women. The data included complete blood count, biochemistry and coagulation-fibrinolysis parameters, and protein and creatinine concentrations in the urine. The clinical data, including patient age, parity and clinical outcomes, were obtained from medical charts.
Indication for an early delivery in women with PE Induced labour or caesarean section were indicated in women with PE with one or more of the following: uncontrollable hypertension exceeding SBP ≥180 mm Hg and/or DBP ≥110 mm Hg even with antihypertensive agents; gradual increase in proteinuria exceeding P/Cr (mg/mg) ≥5.0 in the spot urine; liver dysfunction evidenced by an elevated aspartate aminotransferase (>45 IU/L) level concomitant with a reduced platelet count (<120×10 9 /L) and/or a reduced antithrombin activity level (<65% of normal activity level); and a net weight gain ≥5 kg/week.
Statistical analyses
Data are presented as means±SD. Statistical analyses were performed using the JMP10 statistical software package (SAS, Cary, North Carolina, USA). Differences between the means were tested by analysis of variance and Tukey-Kramer HSD (honestly significant difference) test between each group, and categorical variables were compared using the χ 2 test or Fisher's exact probability test. In all analyses, p<0.05 was taken to indicate statistical significance.
RESULTS
Of the 28 women with PE, 10 (36%) had a duration of IGP ≥3 days, and therefore had P-PE, and the remaining 18 women (64%) exhibited no IGP or had a duration of IGP <3 days, and therefore had O-PE (table 1). Of the 18 women with O-PE, the duration of GH was 3 days or more in 8 women (H-PE group). There were no significant differences in clinical background between women with H-PE and S-PE (in whom the duration of IGP or GH was ≤2 days; table 1). Women with P-PE were significantly more likely to develop proteinuria and hypertension earlier and gave birth earlier compared to those with O-PE.
The P/Cr levels increased with advancing gestation in P-PE and O-PE groups (figure 1). The P/Cr level at birth was significantly higher in women with P-PE than in O-PE. The antenatal duration of proteinuria in women with P-PE, 16.1±10.3 days, did not differ from the antenatal duration of hypertension of 10.9±9.8 days in women with O-PE. The time interval until delivery after diagnosis of PE did not differ between the two groups (6.1±8.2 days for P-PE vs 5.9±6.8 days for O-PE; figure 1). Proteinuria disappeared in 4 (40%) of the 10 women with P-PE and 13 (72%) of the 18 women with O-PE at 1-month postpartum and in 7 (70%) and 18 (100%), respectively, at 2 months postpartum. Three women with P-PE and proteinuria at 2 months postpartum exhibited no proteinuria at 3 months postpartum.
Net maternal weight gain in pregnancy did not differ between the P-PE and O-PE groups, although the gestation period was significantly shorter in women with P-PE than in O-PE (table 1). This may have reflected a significantly greater net maternal weight gain during the last antenatal 1 week in women with P-PE compared with O-PE (figure 2). Net maternal weight gains for the last 1 week (3.1±1.8 vs 1.3±1.7 kg, p=0.023) and for the last 3 days (2.2±1.4 vs 0.5±1.4 kg, p=0.005) were significantly greater in women with P-PE than in those with O-PE.
Although most blood test results did not differ significantly between women with P-PE and O-PE, creatinine level was significantly higher and antithrombin activity and fibrinogen level tended to be lower in women with P-PE than in those with O-PE (table 2), suggesting that these parameters were associated with P/Cr levels. Indeed, the highest antenatal P/Cr levels were significantly inversely correlated with the lowest antenatal antithrombin activity and fibrinogen levels, but not with the lowest antenatal platelet counts or the highest antenatal levels of uric acid and creatinine ( figure 3).
DISCUSSION
In this study, 36% of all the 28 women with PE during the study period developed proteinuria first and hypertension later, confirming that a considerable number of women with PE develop proteinuria first and subsequently develop hypertension. This phenomenon was first described in the literature in 2008 5 followed by reports that the appearance of IGP as determined by dipstick test is associated with established risk factors for PE, such as higher pre-pregnancy body mass index, nulliparity, twin pregnancy and elevation of blood pressure in late pregnancy, 8 supporting the hypothesis that IGP is part of the disease spectrum of PE. In addition, women with IGP defined by repeated positive dipstick test results in two successive antenatal visits show increased likelihood of developing PE. 12 The P/Cr levels have been validated as a proxy for 24 h collection in non-pregnant patients, and determination of P/Cr is the preferred method for evaluation of the amount of protein loss in the urine outside of obstetrics. 13 A P/Cr level around 0.27 is a reasonable cut-off level to exclude or detect proteinuria of 0.3 g/day, 14 although the prognostic value of different quantities of urinary protein is unclear and further studies to identify diagnostic thresholds of proteinuria that are accurate for predicting clinically important outcomes is needed. 10 A recent study examining the occurrence of adverse maternal and perinatal outcomes with different thresholds of proteinuria (0.3-0.49 and >0.5 g/day, respectively) in women with PE confirmed that 0.3 g/day is an appropriate threshold. 15 It may be clinically beneficial to be aware of physiological changes in the amount of protein loss in the urine of pregnant women. Among otherwise healthy women who do not develop PE, mean P/Cr level increases with advancing gestation in singleton and twin pregnancies, and women with twins are significantly more likely to have high P/Cr level than those with singleton pregnancies 16 : the mean P/Cr of 0.06 and 0.10 for singletons and twins, respectively, at GW 16-20 increased to 0.09 and 0.12 at GW 20-28 and to 0.15 and 0.22 at GW 34-38, respectively. In the 10 women with P-PE in this study, the mean P/Cr level continued to increase until delivery after showing P/Cr ≥0.27 (figure 1). These results suggested that the P/Cr increases linearly in some pregnant women in whom it exceeds a cut-off level of 0.27 in the absence of hypertension.
As mentioned above, P/Cr increases with advancing gestation. The P/Cr at birth was significantly greater in women with P-PE than in O-PE, perhaps reflecting the significantly longer antenatal duration of proteinuria in the former than in the latter group (figure 1). However, the time interval after onset of proteinuria until delivery in women with P-PE did not differ from that after onset of hypertension until delivery in women with P-PE (16.1 ±10.3 vs 10.9±9.8 days, respectively). In addition, the time intervals after diagnosis of PE until delivery were similar in both groups (6.1±8.2 vs 5.9±6.8 days, respectively). These observations suggested that IGP should be considered as an important clinical sign equal to hypertension.
In the present study, women with P-PE showed significantly greater maternal weight gain in the last antenatal 1 week compared to those with O-PE. The net weight gains of 4.2 and 3.1 kg in the last 2 weeks and 1 week in women with P-PE accounted for approximately 34% and 25% of the total weight gain of 12.2 kg in pregnancy, respectively, while respective values of 1.6 and 1.3 kg in women with O-PE accounted for 12% and 9.6% of the total weight gain of 13.6 kg. As the mean±SD weekly weight gain, defined as (maternal weight at delivery −pre-pregnancy maternal weight)/(GW at delivery−2), is 0.26±0.12 kg for Japanese women with singleton pregnancies not complicated with pregnancy-induced hypertension, 17 and weight gain in the last antenatal 2 weeks is 0.9±0.9 kg for otherwise healthy Japanese women with singleton pregnancies 18 ; these results indicated that although oedema due to the accumulation of excess water in the interstitial space occurred in late pregnancy in P-PE and O-PE, its degree was greater in women with P-PE than in those with O-PE. Excessive weight gain during the final stage of pregnancy is suggested to be a risk factor for eclampsia. 18 The daily protein loss in the urine was greater in women with P-PE than in those with O-PE, and the degree of proteinuria was inversely correlated with reductions of antithrombin activity and fibrinogen level in this study. A decrease in antithrombin activity has also been suggested to be a risk factor for eclampsia. 18 As enhanced coagulation-fibrinolysis is a characteristic laboratory finding in women with PE, 19 20 the degree of reduction in fibrinogen level may reflect the severity of PE. The circulating plasma volume was reduced in women with PE, 21 22 and serum creatinine level was significantly higher in women with P-PE than in those with O-PE in the present study, suggesting a greater decrease in circulating plasma volume in women with P-PE. Taken together, these results suggest that P-PE constitutes a more serious condition than other types of PE.
STUDY LIMITATIONS
IGP developed at GW 30.0±3.0, while GH developed at GW 32.9±3.8 in this study (table 1), suggesting that IGP occurs earlier than GH. However, this may not be a general characteristic and may have been distorted by incidental bias due to the small size of the study population. Although the definition of proteinuria differed between the present study and our previous study, 5 IGP occurred at a mean GW of 31.1 in 19 women with IGP preceding PE, while GH occurred at mean GW of 28.2 in 5 women with GH preceding PE in the previous study. 5 In another study to determine the fraction of women who developed PE among those with GH, 9 GH developed at GW 33±4 in 26 women with GH preceding PE. As IGP and GH may occur at any time point on or after GW 20, and as there have been no reports regarding the number of women who develop GH and IGP according to GW, the mean GW at the onset of IGP and GH in the present study and previous reports may have been subject to incidental bias due to the limited number of study patients.
CONCLUSIONS
We compared the clinical features of 10 consecutive women who first developed IGP and later developed hypertension with those of 18 consecutive women who developed hypertension first or hypertension and proteinuria simultaneously and were diagnosed with PE. The P/Cr level at birth and net maternal weight gain in the last antenatal 1 week were greater in the former than the latter group. The P/Cr levels at birth were significantly inversely correlated with the lowest antenatal antithrombin activities and fibrinogen levels among women with PE. All of these results suggested that women with PE along with IGP followed by hypertension constitute a more severe PE group.
Contributors RA was involved in drafting the manuscript, data collection and data analysis. TY was involved in study design, data collection and data analysis. MM was involved in study design and data collection. RN was involved in data collection. HM was involved in study design and supervision of this study.
Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None.
Ethics approval The Institutional Review Board of Hokkaido University Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No additional data are available.
Open Access This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 3.0) license, which permits others to distribute, remix, adapt, build upon this work noncommercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http:// creativecommons.org/licenses/by-nc/3.0/
|
2018-04-03T01:20:18.375Z
|
2014-04-01T00:00:00.000
|
{
"year": 2014,
"sha1": "9d6a140b38dab7e62fc838b6026909dd3fc8306f",
"oa_license": "CCBYNC",
"oa_url": "https://bmjopen.bmj.com/content/bmjopen/4/4/e004870.full.pdf",
"oa_status": "GOLD",
"pdf_src": "Highwire",
"pdf_hash": "4773f1ea6f355a63c81e7b53df605eec6ec30b93",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
268195512
|
pes2o/s2orc
|
v3-fos-license
|
A preliminary assessment of butterfly diversity from Mekhliganj town, Cooch Behar District, West Bengal, India
: In the present study, butterfly diversity from Mekhliganj town, which is located on the Teesta River bank of Cooch Behar District, West Bengal, India was studied. A total of 55 species of butterflies were recorded from the two study sites, out of which 22 species were observed for the first time from Cooch-Behar District, not recorded earlier. Out of these, five recorded species were legally protected under the Wildlife (Protection) Act (WPA), 1972 like Chliaria othona, Lampides boeticus, and Hypolimnas misippus . Therefore, efforts should be made for habitat conservation of the Teesta River bank.
INTRODUCTION
Butterflies play a number of critical roles in the maintenance of environmental quality in terrestrial ecosystems (Ghazoul 2002).Conservation biologists now utilize numerous species of butterflies to identify the important habitats that must be protected because they are highly sensitive to environmental parameters such as temperature, light, humidity, and rainfall (Spitzer et al. 1997;Thomas 2005;Bonebrake et al. 2010;Brereton et al. 2011).As an important pollinator, they face numerous conservation challenges as a result of the ongoing augmentation of anthropogenic activities such as industrialization, urbanization, usage of numerous pesticides in various agricultural, horticultural fields, deforestation along with monoculture plantation and overgrazing (Tiple et al. 2007;Roy et al. 2012Roy et al. , 2022)).
There are over 18,000 species of butterflies worldwide, out of which around 1,300 butterfly species are found in India (Samanta et al. 2017;Smetacek 2017).The northern region of West Bengal, which includes the districts of Cooch-Behar, Jalpaiguri, Darjeeling, Dakshin Dinajpur, Uttar Dinajpur, Alipurduar, Malda, and Kalimpong is well-known for its diverse fauna and flora (Pal 2017).However, very few studies of butterfly diversity from this area have been documented.
In this present study, butterfly diversity was studied in the town of Mekhliganj, which is located in the Cooch Behar District.Some authors have previously reported 66 species of butterflies from other areas of the Cooch-Behar District (Das et al. 2020;Roy et al. 2022), however, their diversity from Mekhliganj town remains undocumented and hence the present study was taken up.For this study, two geographically distinct sites were chosen.
MATERIALS AND METHODS
The butterfly diversity was studied at two geographically different study locations in the Mekhliganj city.Mekhliganj is a municipal city in Cooch Behar District located in northern part of West Bengal, covering an area of 3.88 km 2 , situated between 26.35°N and 88.92°E (Directorate of Census Operations V, West Bengal 2011).Field studies for butterfly diversity was conducted between January 2020 to August 2021.During this time each study site was visited twice a month from 0800 h to 1200 h.Butterflies were surveyed and photographed in these study areas.Butterfly survey and counting was conducted using the Pollard walk method (Pollard 1977).Butterflies were counted within 5 m on both sides of the transect walk.Photographs of butterfly specimens were taken with a NIKON D3500 DSLR camera.
Site 1: Town area (TA) consisted of ephemeral water bodies, ponds, marshes, bushes, wetlands, trees and shrubs, tea gardens and agricultural lands that are adjacent to human populations.Site 2: River bank (RB) is located in the Teesta riverbank (120-130 m from the water), and comprised of shrubs, agricultural grounds, aquatic plants and grasses as well as a few human settlements.The study area locations are listed in the Table 1 and photographs are given in the Images 1 & 2.
Three short forms were used to examine the occurrence status of each butterfly species.Butterflies that were very common and plentiful were designated as VC (more than 100), moderately abundant butterflies were designated as M (more than 30) and rare butterflies were designated as R (less than 30).Not even a single butterfly was harmed or killed during this study.
The colour patterns and wing designs of common butterflies were used to identify them on the spot.Other butterflies were carefully identified through photographs.Standard guides of entomological specialists, published literatures (Samanta et al. 2017;Mukherjee & Mondal 2020), field guide books (Smetacek 2017) and some websites (Know your insects 2022; Butterflies of India 2022) were used to confirm the identification of the butterfly species.Data analysis & all the diversity indices like Shannon Weiner index, Margalef index, and evenness index were calculated using PAST software version 4.10.
RESULTS
In the present study, a total of 55 species were recorded belonging to 44 genera of five families namely, Papilionidae, Hesperiidae, Pieridae, Lycaenidae, and Nymphalidae.Most number of species belonged to the family Nymphalidae (22 species) whereas least number of species belonged to the family Papilionidae (three species) (Image 3-5).A total of 53 species of butterflies were observed from TA whereas, 42 species were observed from RB (Table 2).
Results showed that alpha diversity of TA was little higher than the RB (comparing Shannon Weiner index).Margalef index showed higher diversity in TA (6.967) compared to RB (5.865).On the other hand, dominance was more in RB (0.05728) than TA (0.04523).Evenness index for both the study sites were close to each other.The Berger-Parker index, which indicates single taxa dominance was higher in RB (0.1454) compared to TA (0.1182).Table 5 summarises the different diversity indices of the butterflies from the two study sites.
DISCUSSIONS
As per our knowledge, this study is the first of its kind from this town and will shed some light on the region's ecosystem health and macro fauna conservation needs.Previously, three studies regarding butterfly diversity in Cooch Behar District were carried out.Thirty-three species out of a total 55 species of butterflies recorded during this study were also reported in those studies (Das et al. 2020;Roy et al. 2022).Whereas, 22 species were observed for the first time from Cooch Behar District, which were not recorded by previous authors (Das et al. 2020;Roy et al. 2022) Moreover, the number of species recorded in this study is consistent with other studies regarding butterfly diversity in various locations of West Bengal with similar landscape patterns (Ghosh & Siddique 2005;Mukherjee et al. 2015;Ghosh & Saha 2016;Mandal 2016;Mukherjee et al. 2016;Dey et al. 2017;Samanta et al. 2017;Das 2018;Pahari et al. 2018;Mahata et al. 2020;Mukherjee & Mondal 2020).The number of species recorded from the two study sites differed slightly maybe because TA was topographically more diverse than RB and also maybe TA was more suitable to support the host plants of the recorded butterfly species.
A total of five species were found to be included under the Wildlife (Protection) Act (WPA), 1972 (Table 4), viz., Chliaria othona included under schedule I and Lampides boeticus included under schedule II from family Lycaenidae; Euploea core included under schedule IV and Hypolimnas misippus included under schedule II from family Nymphalidae and Appias libythea included under schedule IV from family Pieridae.
The high diversity of butterfly fauna of Mekhliganj indicates the presence of preferable vegetation for different butterfly species.However, gradual urbanization of the town can lead to the disposal of host plants of butterflies resulting in decreased butterfly diversity.
Figure 1 .
Figure 1.Comparative abundance of recorded species among different butterfly families between Town Area (TA) and River Bank (RB).
Figure 2 .
Figure 2. Family-wise percent distribution of butterflies from Town Area (A) and River Bank (B).
Table 2 . Checklist of the butterflies reported from Mekhliganj.
VC-Very Common| M-Moderate | R-Rare.
|
2024-03-02T17:52:00.723Z
|
2024-02-26T00:00:00.000
|
{
"year": 2024,
"sha1": "6ec0ca3ab75eaca48f7901e2f594779716c96d57",
"oa_license": "CCBY",
"oa_url": "https://threatenedtaxa.org/index.php/JoTT/article/download/8137/9714",
"oa_status": "GOLD",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "6ec0ca3ab75eaca48f7901e2f594779716c96d57",
"s2fieldsofstudy": [
"Environmental Science",
"Biology"
],
"extfieldsofstudy": []
}
|
267468004
|
pes2o/s2orc
|
v3-fos-license
|
New record of Theophilea subcylindricollis Hladil, 1988 in Bulgaria (Insecta: Coleoptera: Cerambycidae)
: Theophilea subcylindricollis Hladil, 1988 was recorded for the second time in Bulgaria. A single male specimen was found on 7 May 2023 in Vinarovo Village, Vidin Province, Northwestern Bulgaria. Photos of the specimen ex situ are presented.
Introduction
Genus Theophilea (Cerambycidae) with type species T. cylindricollis Pic, 1895, was described by Pic in 1895 on the base of material from Armenia and subsequently it was reported from South European Russia, Ukraine and Hungary.After a comparative study of a large number of specimens Hladil (1988) considered that the specimens from Armenia are different from those collected in Hungary and described a new species T. subcylindricollis Hladil, 1988.Now the range of type species is limited to Armenia, Georgia, Turkey and Iran (Hoskovec et al., 2023).Currently, the genus Theophilea includes only these two species.
In 2023, a single specimen of T. subcylindricollis was observed and collected in a private yard in Vinarovo Village, Vidin Province.This note reports a second finding of the species in Bulgaria.
Methods
The specimen was photographed ex situ with an EOS 1200D (Canon) digital camera.The collected specimen was deposited in the collection of National Museum of Natural History in Sofia, Bulgaria.
Morphological notes: Theophilea subcylindricollis is apparently similar to Calamobius filum (Rossi, 1790), but it can be distinguished from the latter by the presence of long dark hairs on the ventral side of the antenna, the metallic luster of the elytra and the absence of a furrow on the median tibia (Bense, 1995;Dascǎlu, 2005).
Notes on biology: The larvae develop in the stems of grasses (Poaceae), e.g.Elymus repens, Poa angustifolia and Dactylis glomerata (Pil & Perić, 2012).The adults are active from April to July (Hoskovec et al., 2023) and are weak fliers (Pil & Perić, 2012).The habitats of the species are herbaceous communities in the flooded areas of the forest-steppe region, real steppes and loess steppes (Pil & Perić, 2012).
Notes on distribution: According to Zamoroka (2017) the original range of T. subcylindricollis was restricted to Pannonia and North Black Sea Region, but recently it is expanding westward, northward and eastward.According to Hoskovec et al. (2023) T. subcylindricollis is known from the Czech Republic, Hungary, North Macedonia, Moldova, Romania, Russia, Serbia, Slovakia, Ukraine and Kazakhstan.The species is also reported from Austria (Wiesbauer, New record of Theophilea subcylindricollis Hladil, 1988 in Bulgaria 2015), Bulgaria (Siering & Beier, 2019) and most recently from Albania (Kovács & Mesaroš, 2021).The species is rare and strictly protected in Serbia (Pil & Perić, 2012).
Both localities where T. subcylindricollis was recorded in Bulgaria are from the Danubian Plain, Northern Bulgaria, but are far away from each other.In Bulgaria the species was collected for the first time on 6 May 2017 around Ivanovski Manastir, Rusenski Lom River Valley, district of Ruse, Northeastern Bulgaria, and was subsequently reported by Siering & Beier (2019).The new locality in Vinarovo Village, Vidin Province is by far the northernmost in the country.Also there is a record from Gamzigradska Banja near Zaječar, Eastern Serbia (Popović et al., 2013), which is close to the border with Bulgaria and not far away from the new locality.
As the species was recently discovered in Bulgaria, further research is needed to establish its exact distribution in the country.It is possible that T. subcylindricollis can be found in other localities, most likely in Northern Bulgaria.
|
2024-02-06T17:25:31.999Z
|
2024-02-01T00:00:00.000
|
{
"year": 2024,
"sha1": "0a85c4df2b185aa0eaa9515c723754eff76e0c0f",
"oa_license": "CCBY",
"oa_url": "https://www.nmnhs.com/historia-naturalis-bulgarica/pdfs/000538000462024.pdf",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "62a0389d9a5ec22d9be57d460cd84d3467b316f1",
"s2fieldsofstudy": [
"Biology",
"Environmental Science"
],
"extfieldsofstudy": []
}
|
52111512
|
pes2o/s2orc
|
v3-fos-license
|
Mechanistic and structural studies of KDM‐catalysed demethylation of histone 1 isotype 4 at lysine 26
N‐Methylation of lysyl residues is widely observed on histone proteins. Using isolated enzymes, we report mechanistic and structural studies on histone lysine demethylase (KDM)‐catalysed demethylation of N ε‐methylated lysine 26 on histone 1 isotype 4 (H1.4). The results reveal that methylated H1.4K26 is a substrate for all members of the KDM4 subfamily and that KDM4A‐catalysed demethylation of H1.4K26me3 peptide is similarly efficient to that of H3K9me3. Crystallographic studies of an H1.4K26me3:KDM4A complex reveal a conserved binding geometry to that of H3K9me3. In the light of the high activity of the KDM4s on this mark, our results suggest JmjC KDM‐catalysed demethylation of H1.4K26 may be as prevalent as demethylation on the H3 tail and warrants further investigation in cells.
Widespread N-methylation of both arginyl and lysyl residues occurs on the tails of core histones (histones H2A, H2B, H3 and H4), where it has roles in regulating gene expression [1]. Lysyl residues can undergo successive methylations on the N e -nitrogen to give three methylation states (mono-N e -methyllysine, di-N emethyllysine and tri-N e -methyllysine respectively), as catalysed by S-adenosylmethionine-dependent methyltransferases (KMTs) [2,3]. In many cases, KMTs catalyse methylation at multiple sites to give different methylation states, on both histones and non-histone proteins [1,4]. Enzyme-catalysed demethylation of methylated lysyl residues also occurs at many sites on histones H3 and H4 [1]. Thus, at least at certain sites, lysyl methylation is dynamically regulated by the interplay between the two enzyme-catalysed processes.
Demethylation is catalysed by two families of histone lysine demethylases (KDMs) [5]. Members of the larger family of KDMs, JmjC KDMs, catalyse methyl group oxidation coupled to oxygen-dependent oxidative decarboxylation of 2-oxoglutarate (2OG), forming succinate and carbon dioxide (Fig. 1A) [5]. The available evidence implies that the resultant hemiaminal is normally unstable, fragmenting to give the demethylated lysyl residue and formaldehyde [6]. Dysregulation of JmjC KDM family members is implicated in many diseases, including developmental disorders and many cancers. The JmjC KDMs are thus the focus of drug development programmes [7], although their physiological roles are, in many cases, incompletely defined. There is emerging evidence that certain KDMs have multiple substrates including non-histone proteins and potentially methylated arginyl residues [8,9], suggesting the possibility for in vivo pleiotropy in KDM catalysis.
Though less well studied, lysyl methylation has been observed on histones other than H3 and H4. Histone 1 isotype 4 (H1.4), as opposed to the core histones H2A, H2B, H3 and H4, is one of multiple 'linker' histones that bind to nucleosomal DNA, but which do not form part of the histone octamer [10]. Linker histones have been traditionally viewed as playing important structural roles in stabilising higher order chromatin structure [11]. More recent evidence suggests that histone 1 isotypes and their post-translational modifications also have additional functional roles within the cell, including in transcriptional regulation [11,12]. To date, a number of sites of histone H1.4 lysyl N e -methylation have been identified (Fig 1B), with H1.4K26 being the most abundantly methylated site on H1.4 [13]. The methylation status of H1.4K26 is reported to vary dynamically during the cell cycle [14], and to undergo KDM4-catalysed demethylation, as supported by in vitro studies on mouse KDM enzymes [15]. The methylation status of the H1.4K26 residue is commonly associated with repressed genes, and is known to regulate binding of the family of heterochromatin protein 1 (HP1) proteins and related chromodomains, such as CDYL2 [16,17]. Loss of H1.4K26 (by substitution with a histone H1.4 variant bearing an alanine residue at this position, H1.4K26A) results in a reduction of cell proliferation, deregulation of gene expression and a reduction of stabilisation of H1.4 in heterochromatic regions [14]. Methylation of H1.4K26 also enhances phosphorylation of H1.4K27 by the Aurora B kinase, which in turn hinders recruitment of HP1 to H1.4K26me2 [17,18]. H1.4K26 is also reported to be acetylated by an unknown acetyltransferase, and to be deacetylated by SIRT1 [19]. Thus, emerging, and incompletely characterised, results indicate that complex regulatory loops exist for H1.4, in addition to those established with the core histones.
To address these questions, we report studies investigating human JmjC KDM-catalysed demethylation of H1.4K26 methylation. Our findings reveal that demethylation of methylated lysyl residues in H1.4K26 peptide fragments is catalysed by all members of the human KDM4 subfamily of JmjC KDMs (KDM4A-E), supporting previously reported cell-based studies (predominantly with mouse enzymes) [15]. Unexpectedly, we also observed low level H1.4K26me2 demethylation activity by KDM7A and KDM7B (PHF8), in the context of fragment peptides, though at levels that may not be biologically relevant. While demethylation of H1.4K26me2/1-containing peptides is relatively inefficient for KDM7A and KDM7B, kinetic analyses indicate that H1.4K26me3 is a comparably efficient substrate for KDM4A as H3K9me3 (at least in the context of fragment peptides) and is a better substrate than H3K36me3. This assignment is supported by crystallographic studies indicating that the H1.4K26me3 peptide binds to the KDM4 active site in a similar manner to the H3K9me3 peptide. Overall, H1.4K26me3 appears to be a good substrate for KDM4A, and indeed other KDM4 enzymes, which is supportive of its assignment as a KDM4 substrate in mouse and human cells [15].
Peptide synthesis
Peptides were produced as C-terminal amides using a Liberty Blue automated microwave peptide synthesiser (CEM Corporation, Matthews, NC, USA), using standard fluorenylmethyloxycarbonyl-mediated solid-phase chemistry and NovaPEG rink amide resin (Merck, Kenilworth, NJ, USA). Following synthesis, peptides were cleaved from the resin by incubation with trifluoroacetic acid/water/triisopropylsilane/dimethoxybenzene (92.5 : 2.5 : 2.5 : 2.5) for 3 h followed by precipitation with ice-cold diethyl ether. For kinetic experiments, lyophilised peptides were purified by reverse-phase high-performance liquid chromatography using a Vydac C18 column (Solvent A: 0.1% trifluoroacetic acid in H 2 O, Solvent B: 0.1% trifluoroacetic acid in acetonitrile). Sequences are given in Table S1.
Kinetic analyses
Kinetic parameters for KDM4 enzymes were determined by use of an FDH (formaldehyde dehydrogenase)/NAD + -coupled) assay for quantification of the formaldehyde reaction byproduct, as previously described [27]. Assays were carried out in 50 mM HEPES, pH 7.5, 0.01% Tween-20, with addition of Fe (II) ammonium sulfate (10 lM), sodium ascorbate (100 lM), 2oxoglutarate (200 lM), NAD + (500 lM), KDM4 enzymes (1 lM for specific activities, 500 nM for Michaelis-Menten kinetics) and FDH enzyme (0.001 UÁlL À1 , Sigma-Aldrich, St Louis, MO, USA) in a 30-lL volume in black 384-well plates. Specific activities were measured using a 100-lM peptide. For Michaelis-Menten experiments, peptide concentrations were varied. Reactions were monitored using a PHERAstar FS plate reader (BMG Labtech, Ortenberg, Germany) with 355 nm excitation and 460 nm emission. Kinetic parameters were calculated from the reaction rate during the initial linear phase of formaldehyde production, which were used to calculate specific activities or fitted to Michaelis-Menten equations using GRAPHPAD PRISM (GraphPad Software, La Jolla, CA, USA).
X-ray crystallography
Co-crystals of KDM4A .Ni(II).H1.4(18-32)K26me3. NOG were obtained in sitting drops grown at 4°C with a ratio of 1 : 2 sample to well solution [0.1 M MIB buffer (malonic acid/imidazole/boric acid, pH 6.0, 25% w/v PEG 1500)] [26]. Drops contained 10 mgÁmL À1 KDM4A, 5 mM H1.4(18-32)K26me3, 5 mM NOG and 4 mM NiCl 2 . Crystals were cryoprotected with 25% glycerol then flash-frozen in liquid N 2 . Data were collected using a single crystal at 100 K at the Diamond I04-1 MX beam line and were processed with HKL2000 [31]. The structure was solved by molecular replacement using PHASER [32] (search model: PDB ID 2OX0) and was refined by alternative cycles of CNS [33] and PHENIX [34], with iterative rebuilding of the refined model using COOT [35]. All residues were in the allowed regions of the Ramachandran plot as calculated by PROCHECK [36]. Data collection and refinement statistics are given in Table S2. The crystal structure has been deposited under PDB accession code 6H8P.
Given the activity observed with KDM4E and KDM7A, we decided to investigate the intra-subfamily conservation of this demethylation activity. The demethylation activity of the other subfamily members (KDM4A-D and KDM7B) was therefore screened. For all KDM4 subfamily members, we observed H1.4K26 demethylation activity similar to that already observed with KDM4E (Table 1, Figs S8-S11); however, no H1.4K26 demethylation activity was observed with KDM7B at this enzyme concentration (Fig. S12). Note that as with H3K9, demethylation of the monomethylated H1.4K26 peptide was only observed with KDM4E, under the stated conditions [24]. [29,44]. To test whether similar targeting may occur with H1.4, promoting activity at H1.4K26, an H1.4 peptide was synthesised containing di-N e -methyllysine at K26 and tri-N e -methyllysine at K21 (i.e. at the equivalent position in the sequence to K4 in the H3 peptide, Fig. 1B, Table S1). Monomethylation at H1.4K21 has previously been observed by proteomic mass spectrometry [45]. With the aim of enhancing activity, the enzyme concentration was increased from 1 lM to 5 lM, and time course analyses were conducted with both KDM7A and KDM7B, both with the new peptide (H1.4(18-32) K21me3K26me2) and the H1.4 peptides methylated only at K26 (Fig. 2B). At this increased enzyme concentration, clear evidence for demethylation of H1.4K26me2 was observed with KDM7B in addition to KDM7A (Fig. S12E), but no significant difference was observed between the peptide with and without methylation at H1.4K21 (Fig. 2B). Thus, it appears unlikely that N e -trimethylation at H1.4K21 promotes demethylation at H1.4K26 by KDM7B. These time courses, and similar ones with KDM7A, confirmed that the H1.4K26 peptides were significantly poorer substrates for the KDM7 subfamily than the H3K9 peptides (the weak demethylation efficiency observed for KDM7A/B precluded further kinetic analyses).
Studies then focused on determining the proficiency of H1.4K26 demethylation by the KDM4 enzymes.
Specific activities of each enzyme with H1.4K26me3 were compared to those with H3K9me3, and more detailed studies were carried out with a representative KDM4 enzyme, KDM4A. 1 H NMR time course analyses with KDM4A and the trimethylated H1.4K26me3 peptide confirmed time-dependent demethylation of the Kme3 residue (Figs 2C, S13). The 1 H NMR experiments also revealed concomitant turnover of 2OG into succinate (emergence of a singlet 1 H resonance at d H 2.28 ppm, Fig. S13) that is consistent with the proposed demethylation mechanism [5].
Kinetic analyses were then undertaken using a fluorescence-based formaldehyde dehydrogenase (FDH)coupled demethylation assay, which monitors the formation of NADH during FDH-catalysed oxidation of formaldehyde [27]. Initially, specific activities for each KDM4 enzyme with either H1.4K26me3 or H3K9me3 (each at 100 lM) were determined for comparison (Figs 2D, S14). These results revealed that in all cases demethylation of the two different marks was roughly comparable.
Michaelis-Menten kinetics were then conducted with KDM4A. Samples containing KDM4A, 2OG, ascorbate, H1.4K26me3 peptide, NAD + and FDH were analysed over time and kinetic parameters were determined by recording the initial reaction rates of NADH production (corresponding to enzymatic production of formaldehyde) at varying peptide concentration. Similar experiments were carried out with H3K9me3 and H3K36me3 substrate peptides and the obtained values compared ( Table 2). The experiments revealed that the H1.4K26me3 and H3K9me3 peptides are similarly efficient substrates of KDM4A, giving similar K M (32.3 AE 6.4 lM and 25.5 AE 4.9 lM respectively) and k cat (0.32 AE 0.049 s À1 and 0.31 AE 0.045 s À1 respectively) values. Demethylation of the H3K36me3 peptide was the least efficient, with higher K M (66.8 AE 5.4 lM) and lower k cat (0.12 AE 0.004 s À1 ) values than with the other two peptides.
Crystallographic analyses were undertaken to investigate the binding mode of trimethylated H1.4K26 in [46]. Density was observed for H1.4 residues 24-29. Refinement revealed the H1.4K26me3 peptide bound in a cleft along the surface of KDM4A with the methylated K26 residue protruding towards the active site-bound metal centre (Figs 3A, S1), positioning the carbon of the methyl group~4 A from the active site metal, apparently productively poised for enzymatic hydroxylation and subsequent demethylation.
Comparison of the H1.4K26me3 complex structure with those for a KDM4A:Ni(II):NOG:H3K9me3 peptide complex (PDB: 2OQ6 [26], Fig. 3B) and a KDM4A:Ni(II):NOG:H3K36me3 peptide complex (PDB: 2P5B [47], Fig. 3C) reveals that all three peptides bind with the side chain of the trimethyllysine residue in the same orientation. All three peptides have the same N-to C-directionality through the active site. The two residues N-terminal to the trimethyllysine residue (A24, R25 in the H1.4 peptide, A7, R8 in the H3K9 peptide and G34, V35 in the H3K36 peptide) adopt a similar conformation/orientation in all three structures, with backbone hydrogen bonds between KDM4A E169 and the peptide residues at the À1 and À2 positions (Fig. 3B/C). However, differences arise in the conformations of the C-terminal regions of the peptides. The H3K9me3 and H1.4K26me3 peptides manifest similar conformations/orientations for the three residues C-terminal to the target trimethyllysine residue, sharing conserved interactions with KDM4A K241, D135 and N86 (black dashes, Fig. 3B). By contrast with the H3K9me n and H1.4K26me3 substrate structures, the peptide backbone of H3K36 adopts a more 'bent' conformation, likely enforced by the Fig. 3. Binding mode of H1.4K26me3 to KDM4A. The figure shows views from (A) an X-ray crystal structure of KDM4A bound to nickel (green, substitute for iron), NOG (pink, a 2OG mimetic inhibitor) and an H1.4(18-32)K26me3 peptide (yellow), alone, or overlaid with: (B) a structure of KDM4A bound to H3(7-14)K9me3 peptide (cyan, PDB ID 2OQ6) [26], and (C) a structure of KDM4A bound to an H3K36me3 peptide (violet, PDB ID 2P5B) [47]. Residues from KDM4A in the H1.4K26 structure are shown as pale grey sticks, those from KDM4A in the H3K9 structure are shown as pale green sticks and those from KDM4A in the H3K36 structure are shown as pale pink (A-C). Some of the key interactions between KDM4A and the H1.4K26me3 peptide are shown by black dashes. Interactions between the H1.4K26me3 peptide and KDM4A conserved in the H3K9me3 peptide structure, but not in the H3K36me3 peptide structure are marked as pink dashes. Interactions only observed in the H3K36me3 peptide structure are shown as green dashes. (D) Overlay of the peptide backbones (and Kme3 side chain) of each peptide. The amide flip between the H3K36 peptide, and the H3K9 and H1.4K26 peptides, is highlighted with a dashed black box. presence of a proline (H3P38) in its sequence. Thus, in the K36 peptide, the backbone carbonyl of the H3K36 to K37 amide bond is 'flipped'~180°compared to the analogous amide in the other two substrates (Fig. 3D). Unlike the H3K9 and H1.4K26 peptides (pink dashes, Fig. 3C), the H3K36 peptide displays no interactions with either KDM4A K241 or D135, nor with the sidechain of N86. Instead, the backbone -NH of the H3K36-K37 amide link interacts with the phenol of Y175 (green dashes, Fig. 3C).
Overall, the crystallographic analyses reveal the H1.4K26me3 substrate adopts a binding mode more similar to that of H3K9me3 than H3K36me3. This common binding geometry is supportive of their similar demethylation efficiencies.
Discussion
Though lysyl methylations have been reported on linker histones [10,45], the processes that both regulate these methylation levels and contribute to their functional roles are largely undefined. Demethylation of methylated lysine residues at K26 of linker H1.4 has been reported to be catalysed by the KDM4 subfamily of JmjC KDMs in cells [15,43]. In previous studies on H1.4, no activity was observed for the KDM3 or KDM6 subfamilies; however, the KDM7 subfamily, which, like the KDM4s, also act at H3K9 had not been tested for activity with H1.4 [29,48,49]. Our results demonstrate that at least in the context of histone fragment peptides, recombinantly produced members of both the human KDM4 and KDM7 subfamilies of JmjC KDMs are capable of catalysing demethylation of methylated lysine residues at K26 of linker H1.4. Demethylation of dimethylated H1.4K26 by KDM7A and KDM7B appears considerably less efficient than the well-characterised H3K9me2 substrates for these enzymes, possibly precluding any biological relevance [29,42,44]. However, kinetic analyses with KDM4A reveal that demethylation of the H1.4K26me3 peptide and the well-characterised KDM4 H3K9me3 substrate are comparably efficient, whereas the H3K36me3 substrate is less efficiently demethylated than either H3K9me3 or H1.4K26me3 under the conditions tested [23,24,50,51].
Crystallographic analyses of H1.4K26me3 bound to KDM4A imply very similar binding modes for the H1.4K26me3 and an H3K9me3 peptides [26,47], consistent with the kinetic studies which imply that they are substrates of approximately equal efficiency. Importantly, the H3K36me3 peptide, which at least in vitro is a less efficient substrate for KDM4A-C (and not a substrate at all for KDM4D/E) than either H3K9me3 or H1.4K26me3, binds differently at the KDM4A active site [23,47,50]. One clear difference is in the binding mode of the amide bond on the C-terminal side of the N-methylated substrate lysine for H3K9/H1.4K26 versus H3K36 (Fig. 3). Although how this observation relates to the different binding/catalytic efficiencies of the different peptide sequences is difficult to dissect, it suggests that subtle differences in binding conformation may have substantial effects on catalytic efficiency, an observation relevant to functional assignment work on JmjC KDMs [50,52]. Similar subtleties have been observed for chromatin reader domains; in some cases they bind similarly to the closely related H3K9 and H1.4K26 sequences (e.g. HP1) [17], whilst in others binding is only observed to one or the other (e.g. ankyrin repeats of G9A) [43].
The combined biochemical and structural studies thus identify methylated H1.4K26 peptides as substrates for JmjC KDMs with broadly comparable demethylation efficiencies to those reported for KDM4 substrates on core histones [23,24,50,51]. Given this potential for activity at H1.4, it is of interest to further investigate the consequence of methylation at this position in vivo, in particular in the context of ongoing medicinal chemistry studies targeting the KDM4 subfamily as a treatment for disease [7,53]. The methylation status of H1.4K26 is reported to regulate binding of HP1 and other chromodomain proteins [16,17], which are associated with formation of heterochromatin. Thus, the reversal of this methylation by KDM4 enzymes in cells may well have direct effects on chromatin compaction and gene expression. It remains unclear whether a complex and dynamic 'histone code' akin to that proposed for the H3 N-terminal tail, exists for H1. The dynamic modulation of PTMs would be a likely requirement for this. Regulation of PTMs on H1.4 has already been demonstrated to play important roles in gene regulation, as in the case of acetylation of H1.4K34, which may help recruit the general transcription factor IID through its bromodomain, and which increases the dynamic exchange of H1 [54], or citrullination of H1.2R54, which results in histone eviction and chromatin decondensation [55]. The ready demethylation of H1.4K26 as catalysed by the KDM4 demethylase family argues that this might also be the case for H1.4K26 methylation, at least for H1.4K26me3 [12]. Overall, we hope that our findings will stimulate further cellular work to characterise the full scope of JmjC KDM substrate selectivity and will be informative for ongoing studies with JmjC KDM inhibitors.
Supporting information
Additional supporting information may be found online in the Supporting Information section at the end of the article. Fig. S11. KDM4D catalyses lysine demethylation at H1.4K26. Fig. S12. PHF8/KDM7B only catalyses lysine demethylation at H1.4K26 at high concentration. Fig. S13. Analysis of KDM4A demethylation by 1 H NMR. Fig. S14. Specific activity determination for KDM4 enzymes. Table S1. Peptide sequences used in this study. Table S2. Crystallographic data processing and refinement statistics.
|
2018-09-15T21:18:11.738Z
|
2018-09-14T00:00:00.000
|
{
"year": 2018,
"sha1": "78d6ba8db62cbc06763a4211b0bbcbd9a93535a3",
"oa_license": "CCBY",
"oa_url": "https://febs.onlinelibrary.wiley.com/doi/pdfdirect/10.1002/1873-3468.13231",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "78d6ba8db62cbc06763a4211b0bbcbd9a93535a3",
"s2fieldsofstudy": [
"Chemistry",
"Biology"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
229923058
|
pes2o/s2orc
|
v3-fos-license
|
Proton strings and rings in atypical nucleation of ferroelectricity in ice
Ordinary ice has a proton-disordered phase which is kinetically metastable, unable to reach, spontaneously, the ferroelectric (FE) ground state at low temperature where a residual Pauling entropy persists. Upon light doping with KOH at low temperature, the transition to FE ice takes place, but its microscopic mechanism still needs clarification. We introduce a lattice model based on dipolar interactions plus a competing, frustrating term that enforces the ice rule (IR). In the absence of IR-breaking defects, standard Monte Carlo (MC) simulation leaves this ice model stuck in a state of disordered proton ring configurations with the correct Pauling entropy. A replica exchange accelerated MC sampling strategy succeeds, without open path moves, interfaces, or off-lattice configurations, in equilibrating this defect-free ice, reaching its low-temperature FE order through a well-defined first-order phase transition. When proton vacancies mimicking the KOH impurities are planted into the IR-conserving lattice, they enable standard MC simulation to work, revealing the kinetics of evolution of ice from proton disorder to partial FE order below the transition temperature. Replacing ordinary nucleation, each impurity opens up a proton ring generating a linear string, an actual FE hydrogen bond wire that expands with time. Reminiscent of those described for spin ice, these impurity-induced strings are proposed to exist in doped water ice too, where IRs are even stronger. The emerging mechanism yields a dependence of the long-time FE order fraction upon dopant concentration, and upon quenching temperature, that compares favorably with that known in real-life KOH doped ice.
I. INTRODUCTION
Ice famously intrigues experimentalists and theoreticians alike. The crystal structure of ordinary I h ice consists of hexagonal rings of water molecules, each molecule tetrahedrally hydrogen-bonded to other four. Every proton occupies one of two sites along the H-bonds between two oxygens, and each oxygen satisfies the ice rule (IR), with two incoming and two ongoing H-bonds 1,2 . At low temperatures and under ordinary conditions, protons are unable to reach equilibrium, and ice is kinetically stuck in a glassy state characterized by the celebrated Pauling entropy 3,4 resulting from an infinity of different, IRconserving, defect free proton configurations. Still debated is the possibility to attain, in pure ice at low T , the low energy, zero-entropy ferroelectric (FE) phase, ice XI, endowed with a macroscopic polarization order parameter, the protons occupying bond sites in a unique order. An intimate understanding of the possible FE ordering mechanism of pure ice is necessary in order to understand why and how it is avoided or reached. Of fundamental importance, that issue also bears potential implications in fields as disparate as astrophysics 5,6 and surface science 7,8 .
The well known extrinsic ingredient which experimentally permits realization of ferroelectricity in bulk ice is KOH-doping, which allows ice to undergo, upon cooling below T c ≈ 72K 9 , the transition from proton disorder to proton-ordered ferroelectricity. This transition has several remarkable features. First, its temperature is practically independent of the concentration and even of dopant type, suggesting the FE phase and its onset are in fact intrinsic equilibrium features of ice, escaping realization merely due to kinetic reasons when doping is absent 10 . Second, the ferroelectric order achieved after long annealing times is partial and its fraction depends weakly on the dopant concentration in a wide range 11 . Third, the transition kinetics upon quenching is significantly dependent on the quenching temperature T q 12,13 . For example, at T q = 0.75T c no ordering is observed, but at T q = 0.89T c a fraction of the FE phase appears. As a simplifying note, we mention that nuclear quantum effects, generally relevant to hydrogen-bonded systems 14 , may be provisionally neglected here, since deuteration affects only modestly the transition temperature 13,15 .
The bulk of these observations is rationalized by the understanding that an FE state of low temperature hexagonal ice is thermodynamically favored, but its realization is hindered by a kinetic slowdown, likely due to the IR. The slowdown is overcome only in special conditions such as doping, accompanied by quenching from a sufficiently high T and subsequent annealing. Moreover, in many conditions the transition to the FE phase is incomplete, indicating that the slowdown mechanism can act also at the mesoscale. To physically clarify this scenario, it is desirable, due to the complexity of the problem, to resort to some simplified and yet microscopic model. The model should offer a comprehensive descrip-tion of the non-ergodic proton disorder, of the ordered FE state, of their properties and ideal phase transition. It should also allow the introduction of dopants, with access to the transformation kinetics to partial FE ordering which they permit.
Microscopic model descriptions of ice are abundant, including off-lattice descriptions, both force-field based, such as Ref. 16 and ab initio. In particular, density functional theory (DFT) calculations also using graph invariants predict the existence of an FE transition at T c =98K in pure hexagonal ice 17 . A better estimation of T c ≈70-80K was obtained by recent DFT-based Monte Carlo (MC) simulations using hybrid functionals 18 . An FE ordered phase is also predicted in cubic ice by ab initio calculations, and is comparable in energy to the corresponding FE phase of hexagonal ice 19 . A recent study compares measured infrared spectra with theoretical results from classical molecular dynamics and ab initio simulations suggesting evidence of partial FE proton ordering in cubic ice 20 . A possible FE transition at low T in pure hexagonal ice was studied by lattice models using empirical water potentials, yet with relatively inconclusive results owing to strong dependence upon details of the potential models 21,22 . Alternatively, MC ice simulations using a point charge lattice model led to a nonferroelectric state at low T 23 but it was argued that the failure to find an FE state could be due to the simplifications of the model. In addition, many specific observations have been made concerning the importance of rotational Bjerrum defects 3 , the role of Coulomb 24,25 , of multipolar interactions 26 , and other aspects 27,28 including off-lattice hydroxide configurations 29 .
None of these realistic off-lattice concepts and considerations, however, seems so far to lead to a well defined model description of the subtle kinetic phenomena connected with establishing partial FE order, or the lack of it due to persistent and glassy proton disorder, formidable problems that a simplified lattice model is more likely be able to tackle. That calls therefore for a fresh attempt.
We have developed a bare-bone lattice model of ice, and a MC technique which allows simulating large samples at arbitrarily low temperatures (see Calculation Details). The main feature of this model is that while embodying "dipole-dipole" interactions, its finite temperature ensemble contains only configurations which satisfy exactly the IR. Even if a lower energy FE state is stabilized by dipolar interactions, standard MC fails to evolve configurations of this system even at finite temperature, causing it to retain the statistical distribution of proton rings and the Pauling entropy down to arbitrarily low temperatures, as in real neat ice. This ice-rule obeying disordered state provides an ideal framework where the effect on kinetics of an added idealized dopant can be studied.
The physics of our ice model, built on a diamond lattice, bears similarities to that of spin ice in pyrochlores, whose lattice is dual to diamond, making the two isomorphic upon identification of the orientation of the spins with the location of hydrogen atoms on the bonds between oxygens 30 . Unlike spin ice models our model "water oxygens" possess interacting dipoles whereas the only interaction between hydrogens ("spins") come from the topological ice-rule constraints. These analogies and differences underpin those that will appear in the equilibrium phase diagram, reminiscent but not identical to the 3D Kasteleyn transition 31 of spin ice in a field 32,33 , as well as to the out-of-equilibrium, proton ordering behaviour. The kinetic process by which the dopant triggers proton ordering is an avalanche of proton hoppings, which breaking up closed rings, generate strings of collinear hydrogen bonds, all pointing in the same direction along a winding line, thus upsetting the ring landscape of the disordered phase. The barrier characterizing this simple process, which is of the order of the dipolar interaction, is therefore the rate limiting step for the formation of the string in ice.
Before introducing the details of our work, it should be stressed that our model study omits, by deliberate choice, many details that are known to play a role in real ice. In spite of that, we shall nonetheless throughout the paper compare the model's main results with known experimental facts. Points of agreement and disagreement between them will gauge the model's ability to address mechanisms that underlie some of the unexplained behaviors of ice ferroelectricity. In particular we will show that the string formation qualitatively reproduces several known facts in real KOH doped ice, providing bare-bone mechanisms for the dependence of the ferroelectric fraction on the dopant molar concentration, and on the quenching temperature.
II. MODEL
Our model system is a diamond lattice of N "water molecules" where each oxygen is connected tetrahedrally with four neighboring ones by H-bonds, as sketched in Fig. 1(a). This is the connectivity of cubic ice I c . The cubic and hexagonal (I h ) phases of ice differ by the stacking order of the hexagonal bilayers that form the lattice, but their topologies are similar 19 . Since the model Hamiltonian, which is presented below, depends on the connectivity of the lattice sites but not on the distance between particles, the results obtained here should be valid for both I c and I h .
Each oxygen at site i in the diamond lattice has four bonds to nearby sites labeled by j. The variables of our model are the proton configurations on all bonds. We represent them by a set of 4N variables ϕ ij , one for each directed bond. We have ϕ ij = 1 if in that bond there is a proton closer to oxygen i and ϕ ij = −1 if not. Note that in general ϕ ji is independent from ϕ ij . In pure ice, where we exclude Bjerrum defects 34 , all bonds possess one and only one proton, and all oxygens two protons, ϕ ij = −ϕ ji , and the independent variable number shrinks to 2N . A bond Ising-type variable σ j i is defined as: σ j i = ϕ ij e ij , where e ij is a vector pointing from oxygen site i to site j. Using these variables we define the dipole associated with oxygen site i as: That definition corresponds to a dipole of modulus one in each site satisfying the IR but is also valid for sites where it is not satisfied, where the dipole moduli are now smaller than one.
The Hamiltonian of our model is where the two control parameters, J and k, are both positive. The first term represents the nearest-neighbor (nn) dipole-dipole ferroelectric interaction between oxygen tetrahedra. The second term penalizes configurations that violate the IRs. Indeed, j ϕ ij = 0 only if two protons are close to site i and two are far. We will mostly describe the properties of this model, which to the best of our knowledge has not been studied for ice, in the special case k → ∞, where violations of the IR are forbidden. Yet, we will make use of finite k in replica-exchange accelerated MC. The physical order parameter is the FE polarization, defined as The model can be mapped, translating from site to bond variables, to an Ising-type Hamiltonian The first term is a large (practically infinite) ferroelectric coupling between nearest neighbor bonds, promoting frustration and disorder through its strong topological IR constraints. That effect is mitigated by the second term, also ferroelectric, between second neighbor bonds, contributing instead to stabilize a possible FE ordered state at low T. It should be noted that off-lattice configurations 29 as well as multipolar terms and long-range interactions 24-28 , are omitted. Testing the effects of removing these drastic approximations is beyond the scopes of this first study. Mainly justified by simplicity, the short-range interaction assumption is at least encouraged by screening of longrange electrostatic tails, which is induced by polarization. It may also be noted that, unlike first neighbor interactions, always ferroelectric, the sign of long-range interactions is not uniform, but rather oscillates between ferro and antiferro depending on direction, suggesting a certain level of cancellation. Indeed, MC studies of the dipolar spin ice model actually showed that medium to long range interactions are screened out, suggesting that short range physics should remain qualitatively valid 30,[35][36][37] .
The KOH impurities, which play a fundamental role in determining the kinetics, are introduced in our model as follows. In ice the K + impurity replaces a proton in one bond. This turns the proton-deficient molecule into a cation-hydroxide pair K + (OH) − . That is simulated in our model by a single, fixed proton vacancy in a bond ij (represented by setting ϕ ij = ϕ ji = −1), an action which simultaneously deprives oxygen site i of an outgoing proton, (this is the Bjerrum defect mimicking K + , which we keep fixed), and deprives oxygen site j by one incoming proton -this is a mobile IR breaking defect, mimicking (OH) − . We have no charges in our model, but for the sake of illustration, we will call these two defects K + and OH − . The presence of charge dopants in real doped ice induces lattice distortion, which could lower the relaxation barrier of the local structure, speeding the interconversion from paraelectric (PE) to FE. However, the formulation of our model does not allow assessing how important this particular effect is.
III. CALCULATION DETAILS
We carried out MC simulations on a diamond lattice with 12x12x12 cubic cells containing N = 13824 sites, representative of static oxygen sites in cubic ice. A zdirected electric field is coupled to the polarization for breaking the symmetry of the isotropic Hamiltonian. The field is removed after equilibration and is small enough (| E| ≈ J/10) in order not to modify the transition temperature. First of all we confirmed that the IR term causes frustration that prevents our defect free ice model from reaching thermodynamic equilibrium within standard MC sampling, where proton variables ϕ ij change one at a time while chosen randomly through the lattice. 2N proton move attempts performed sequentially represent our MC step or pass. This well known problem was addressed long ago by Rahman and Stillinger 2 who dealt with the IR by performing random walks on the lattice and noticing that paths involving crossing of the periodic boundaries bring in a change in total dipole moment. Time-honored as it is, that method involves the necessity of very large simulation sizes, which we prefer to avoid. We thermalize the system by a Hamiltonian Replica Exchange Method (HREM) 38 . We simulated m = 1, 2, ...M replicas in parallel, at the same temperature and same J but with different values of the ice-rule penalty parameter k. The original replica m = 1 has a prohibitively large value of k = 30J, practically infinite. As m increases the parameter k is reduced successively until the last replica m = M = 40, which corresponds to k = 0. The IR-violating defects (excess or lack of protons attached to an oxygen site) will therefore occur with increasing probability in replicas with decreasing k. After a prescribed number of MC steps the instantaneous configurations for adjacent replica are allowed to swap with a probability dictated by the standard Metropolis exchange criterion 39,40 .
As a direct extension of the intrinsic, defect free model, whose equilibrium properties will be shown to agree well with those of pure ice, we subsequently introduced defects, and studied the kinetics which they generate. This is done in a lattice of 9x9x9 cubic cells containing 5832 sites. To represent the effect of "KOH type" impurities, a small number of L ≪ N fixed proton vacancies were introduced, randomly distributed in lattice bonds ij, by setting ϕ ij = ϕ ji = −1, i.e. no proton either near oxygen i or near oxygen j, as if it had been moved to the nearby interstice in order to mimic the role of K +3 . In Fig. 3 the proton vacancy in a given bond is depicted by an interstitial K + ion schematically replacing the proton H + . In real ice, the KOH impurity produces two mobile defects: an ionic OH − defect and a Bjerrum L−defect (proton vacancy) 3 . In our model, for the sake of simplicity, the Bjerrum L−defect is fixed and only the hydroxide defect is able to move. Thus, the introduced impurity generates a traveling hydroxide which can trigger transitions involving the nearby protons, transitions otherwise impossible (see Fig. 3(b)). It's worth noting here that we expect similar kinetic effects from a mobile Bjerrum L−defect as those observed with a traveling hydroxide.
To address the kinetics of the doped ice model, we performed non-equilibrium simulations by a quenchingannealing (QA) simulation protocol, carried out with standard MC moves -which mimicks to some extent the real time evolution of experiments -and by comparison also with the HREM protocol, which artificially speeds up evolution towards equilibrium. Each calculation performed was an average of 50 runs with different randomnumber generator seeds. Starting with the doped system initially thermalized with HREM at very high T ∼ 3 T c , we quench it down to a temperature T q below T c (quenching) and then let it thermalize at the quenching temperature T q till the system comes as close as possible to equilibrium (the annealing process). That was done for a range of T q and of doping concentrations, so as to address the known experimental dependence of ice ferroelectricity upon these parameters.
A. Equilibrium phase diagram and proton rings
We first discuss the equilibrium phase diagram for defect-free bulk ice model, where no violations to the IR are allowed. In order to thermalize this model we use a Hamiltonian Replica Exchange Method (HREM) (see Calculation Details), in which a set of replicas differing only by the IR-controlling parameter k, are simulated in parallel. In the first replica k is extremely large, as appropriate to the IR conserving model we want to address, in the other replicas k is succesively smaller and smaller. Configurations of different replicas are exchanged according to a replica exchange protocol, which allows the simultaneous thermalization of all the replicas. Fig. 1(d), shows the average value of the polarization P as a function of temperature. There is a transition be-tween an ordered FE state (P ∼ 1) and a disordered PE state (P ∼ 0) at an equilibrium transition temperature T c ≈ 3J. As expected, T c is proportional to the strength of the oxygen dipolar-interaction parameter J (see Eq. 1). The transition appears to be strongly first order. Even without the IR constraint (k=0), the symmetrydictated universality class of this transition would differ from straight Ising. Indeed, the Hamiltonian (Eq. 1) possesses six equivalent FE ground states (polarization along x, -x; y, -y; z, -z), making it closer (yet not identical) to a Potts model, a family many members of which support first order phase transitions in high dimensions (see e.g. Ref. 41 ). Conversely, the fully IR conserving Hamiltonian would, once the dipole-dipole interaction was removed (J=0) and the protons were coupled to an electric field, display a Kasteleyn-type transition 31 as in spin ices. With nonzero J and large k, our model is richer, even if retaining some qualitative similarities to Potts and Kasteleyn transitions. A note of caution here is that while the experimental FE transition of real ice is, as in this model, first order 3,5 , there are in ice secondary order parameters, such as strain coupling, that are absent in the model but that would play a role making the transition first order.
Next, the equilibrium entropy evolution with temperature is a crucial information. We obtain it at each temperature as the integral of the specific heat at constant volume over T . Strictly speaking, this procedure is correct only if no first-order phase transitions are encountered along the path. However, in finite size systems like those analyzed in this work, first order transitions are avoided, the thermodynamic potentials vary continuously, and the procedure is therefore justified. The inset of Fig. 1(d) shows how at T c the entropy of the defect free IR conserving model correctly rises from essentially zero (the model has no acoustical modes) to the Pauling value S ∼ ln(3/2) across the transition. The free energy F as a function of the polarization can be estimated, at a given temperature, from the histogram of the polarization observed in the first replica. F is shown in Fig. 1(c). At T c , F shows two minima with same free energy separated by a barrier, which confirms the firstorder character of the transition. At T = 0.93T c the free energy retains a secondary minimum at P ∼ 0, signaling a metastable (equilibrated) PE state which, however, no longer exists at T = 0.85T c . An analogous metastable FE state must also exist above T c , but is already lost at T ∼ 1.03T c . Thus, free-energy barriers vanish shortly below and just above T c . Associated with the transition there is a change in the proton configuration inside the 12 hexagonal rings which thread each lattice site. The role of rings and directed H-bonds is widely discussed in ice and water 16,43 . Here, we must in addition distinguish different ring types according to their polarization. For that, we associate an arrow to each H-bond, pointing from the oxygen possessing a close-by proton to the other oxygen in that bond (see Fig. 1(e)). We then count, for each ring, the number of arrows pointing in a specific clockwise direction, and define from that a directed order parameter s = (6 − | 6 i=1 ϕ il |)/2, where i runs through the six-site ring clockwise, l = i + 1, and ϕ il is the proton variable of the H-bond il. The four different kinds of proton rings are schematically depicted in Fig. 1(e) labeled as S s , therefore S 3 , S 2 , S 1 , and S 0 , corresponding to s = 3, 2, 1 and 0, respectively. This ring classification differs from a previous one (see Supp. Inf. of Ref. 43 ), except for the case of the ring S 0 .
A schematic representation of a typical microscopic configuration in the PE phase of ice is depicted in Fig. 1(b). Inspecting all hexagonal rings in the equilibrium state of the defect-free ice model, we extract the average population < n s > of rings of each s = 0, ..3. Fig. 1(e) shows the results obtained as a function of temperature. For each site 3 s < n s >= 12 in pure ice at all temperatures. Well below T c , all rings have s = 3, thus n 3 = 12, accompanied by a net local polarization along z. In this FE state, schematically displayed in Fig. 1(a), where only z-polarized S 3 rings are present, the symmetry between the six possible polarizations of the system (along the x, y or z axis, and corresponding negative directions) is spontaneously broken by long-range order. At T c , < n 3 > has a sharp drop which accompanies the collapse of the order parameter P , while all other rings concurrently surge and proliferate as shown in Fig. 1(e). Finally, in the PE phase above T c , all ring populations acquire steady values, almost constant with further temperature increase. The S 3 rings do not disappear, but we find them equally polarized in all directions in accordance with the vanishing order parameter. In the following, S 2 , S 1 , and S 0 are called "disordered" rings because they only appear in the disordered phase. The slight residual temperature dependence can be attributed to finite size in our simulations.
Entropy reveals another effect of small size. Our calculated entropy at T > T c is ≈ 10% higher than the Pauling entropy as shown in the inset of Fig. 1(d). Pauling's entropy is known to be only a lower bound 33 . Our HREM calculations capture the additional proton correlations along the closed rings, which cause entropy to rise higher for smaller sizes 44 .
The population distribution of rings in the disordered phase is also similar to that found in a recent ab initio molecular dynamics study of hexagonal ice 43 . For instance, the relative abundance of S 0 rings is about 15.8% in our calculation which is nearly equal to the corresponding averaged-value obtained in Ref. 43 for hexagonal ice, ≈ 16.5%.
B. Non-equilibrium kinetics and FE polarization in doped ice
Thus far we described the static properties, both equilibrium and metastable, of the ice model. It is now possible to address the nonequilibrium MC kinetics of transformation between PE and FE states. The equilibrium transition being first order, the transformation will occur by nucleation. Yet, this process is very strongly influenced by IR constraints, which render ordinary homogeneous nucleation impossible, at least for k large enough. In that limit, only inhomogeneous nucleation is possible. We therefore study the transformation from the metastable and proton disordered state, into an ordered or partly ordered FE state, taking place once model impurities, meant to play a similar role to KOH, are introduced. To that end, we first equilibrate with HREM the IR conserving state at high T ∼ 3T c , and then quench it down to some T q below T c where we let it thermalize with standard MC moves. We call this procedure a quenching-annealing (QA) simulation protocol (see Calculation Details).
Initially, after a certain number of thermalization time steps at high T , the system reaches a state where all the ionic defects (hydronium -hydroxide pairs and even molecular states with zero or four protons) introduced by the random initial configurations managed to recombine and disappear. We checked that after thermalization at high T in the KOH-doped system with L extrinsic impurities, we have precisely L mobile hydroxide defects in the system because all intrinsic ionic defects permitted by finite k have recombined (see Fig. 3). After equilibration at high T with HREM, all the simulations continue with a QA protocol using standard MC (unless otherwise stated) on the replica with the largest k value. Fig. 2(a) shows the non-equilibrium evolution of the instantaneous polarization in a QA simulation after quenching at T q = 0.57T c , in a range of different conditions. As a first check, in pure ice (L = 0) the system remains stuck in a non-equilibrium glassy state with P ≪ 1 as shown by the blue dashed curve in Fig. 2(a). If HREM is instead kept active throughout, then the low-temperature thermodynamic equilibrium with P = 1 (FE order) is quickly recovered after quenching, as expected and as shown by the green curve in Fig. 2(a). The next step is the simulation of doped ice with L impurities representing KOH impurities (see Model and Calculation Details). Unlike the undoped case (L = 0), results for L = 1, 4, and 8, show a kinetic evolution with frank onset of the FE order parameter (see Fig. 2(a)). At large MC step number (conventionally representing long evolution times), the polarization P reaches ≈ 60 − 75%, almost independent of the impurity number L. This is in qualitative agreement with neutron diffraction measurements of doped deuterated ice, where a volume abundance of ≈ 48% of ice XI is observed in the bulk at T = 0.89T exp c 12 . The critical temperature and the ice XI fraction locally formed are practically indepen-dent of the impurity concentration [9][10][11]13,45 , a nontrivial outcome which is reproduced by our model. The L = 1 case corresponds to a molar fraction of 1/5832, similar to that of the doped-ice samples used in Ref. 10 , 1/5540, and leads to extensive FE ordering in the model that is qualitatively similar to experiment 9,10,12 . Moreover, neutron diffraction of annealed KOD-doped deuterated ice after low T quenching showed a sustained intensity growth of the characteristic 131-Bragg peak of the FE phase XI. Its intensity, proportional to the volume fraction of ferroelectric ice-XI, tends to a definite limit at long annealing times 12 , also decreasing when the quench temperature was lowered, as shown in Fig. 2(c). We conducted additional extensive weak-doping simulations, with L = 1, exploring how a change of T q affects the kinetics of FE onset. As Fig. 2(b) shows, the calculated long-time FE fraction diminishes as the quenching temperature is lowered, in qualitative agreement with the neutron diffraction data of Fig. 2(c). There are therefore good hopes that our model could shed light on the underlying reasons.
C. Microscopic mechanism of string nucleation
We now analyze the microscopic mechanism of IR defect-induced disorder nucleation in the FE phase and conversely, the impurity-induced nucleation and growth of FE clusters in the PE phase. First, we address the impurity-triggered nucleation of disorder by regular MC simulations of an ordered FE crystal at T = T c /2. At this low temperature, as shown in Fig. 3(a), the hydroxide is unable to propagate freely through the crystal. However, when temperature rises above T c (≈ 2T c ), the hydroxide departs from the impurity site, and travels through the lattice (see Fig. 3(b)). In its journey, it flips onto the xy plane the dipoles from their original FE z-polarization. That generates a chain, or string, of xy-dipoles, shown by the blue line in Fig. 3(b), with origin in the fixed initial impurity site and end at the moving hydroxide. This chain bears a resemblance to the so-called "Dirac" string associated with a magnetic monopole of model spin ice systems [46][47][48] . While the IR is of course the topological constraint that water ice and spin ice have in common which gives rise to strings in both cases, the two model systems are far from identical, as we will underline later.
As shown in Figs. 3(a) and 3(b), in the early stages of string formation the hydroxide can only progress upwards (see the blue path) along z and against the total polarization. As the ice rules are satisfied everywhere and the system is in the ordered phase, it can only receive one of the two protons from the top neighbouring water molecules. The preference will be to receive the one that creates locally a basal dipole aligned to that of the previous step in the hydroxide path, thus creating a chain of dipoles aligned in the x or y directions, no longer along z. This has an energy cost of ∆E = 2J per step. Otherwise, the resulting basal dipole would be perpen- with the characteristic S3 rings colored with mustard. The planted proton vacancy representing the doping by a KOH impurity produces a hydroxide defect (colored with turquoise) which remains in its site at this simulation temperature. Notice that we added a fixed K + atom colored with brown next to the proton vacancy for the sake of clarity in the picture as explained in Calculation Details (see also Microscopic Mechanism of String Nucleation); (b) immediately after the suddenly raise of T above Tc showing the hydroxide displacement through single proton jumps following the blue path and transforming two z-polarized mustard rings S3 into two "disordered" green rings S0 and S2; (c) after a longer simulation time above Tc showing the creation of a PE cluster (green rings) along the blue path of the hydroxide. Violet (green) arrows at different oxygens represent z-polarized (xy-polarized) dipoles di. dicular to that of the previous step with a higher cost ∆E = 3J. Yet, since MC moves are randomly generated and accepted according to Boltzmann weights, the chain-end hydroxide progresses, owing to finite temperature, not only in the z direction but also in the x or y directions, as in Figs. 3(b) and 3(c). As noted, the disordering mechanism produced onto the initially per-fect FE system by the hydroxide string with a probability that bifurcates at each step along the path resembles that of the three-dimensional Kasteleyn-like transition of spin ice in external field 49 . However, in our ice model the transition is not induced globally by an external field, but by the J-induced local field created by the growing seed. Unlike Kasteleyn's strict case of spin ice models, where an infinite number of configurations are degenerate and excitations have a large gap 30 , here nucleation brings the system closer to a lower free energy state due to the local dipole-dipole interactions. A second difference is that the local dipole field makes the probability to move a proton in the two possible branches uneven, affecting qualitatively the nucleation dynamics. The disorder produced by the hydroxide migration may also be characterized by the transformation of z-polarized S 3 (mustard) rings into S 2 , S 1 or S 0 (green) "disordered" rings with a certain degree of xypolarization (see also Fig. 1). For instance, Fig. 3(b) shows the formation of an S 0 ring with a three-step hydroxide jump, and that of an S 2 ring with a single-step jump. Thus, the traveling hydroxide nucleates in its path a disorder "contagion" cloud, formed by green rings, elongated in the z direction and zigzagging in the x and y directions, as shown in Fig. 3(c). The green cluster shown in this figure has a substantially smaller polarization than the FE bulk, and can thus be considered a seed of the PE phase inside the FE bulk. It is worth noting here that after the hydroxide has passed, the green PE cluster cannot further spread expanding its frontier perpendicularly to the blue line because the strong IR constraints frustrate any proton move attemp across the cluster boundary during the standard MC simulation, as they presumably would in real time evolution. In other words, the PE cluster can only progress as an elongated string through the traveling-hydroxide tip.
In reverse, and crucially, we finally address the nucleation mechanism of FE order inside the disordered phase. Thermalizing the system with a single impurity (L = 1) at a high T = 3T c , with an initial HREM MC simulation lasting 20000 steps, we choose the replica with the largest k value, therefore with well-respected IRs. With that, a regular (Metropolis) MC simulation is continued for the same number of steps. In this thermalization, the hydroxide migrates following a completely random path and losing track of the initial impurity site. Fig. 4(a) shows a typical configuration formed after thermalization. Completely disordered, it displays all types of S β rings as described earlier. This disordered configuration is then suddenly quenched to a low temperature below T c (see Figs. 4(b) and 4(c)). The reduced mobility of the hydroxide and the decrease of entropic contributions in favor of enthalpic ones reflects in the tendency of the hydroxide to migrate preferentially in one direction, that will in fact define the incipient polarization direction which we denote as z, as in Figs. 4(b) and 4(c). In its way along the new path, the hydroxide transforms disordered green rings into ordered z-polarized mustard ones. For instance, Figs. 4(a) and 4(b) show the transformation of a S 0 into a z-polarized S 3 ring as the hydroxide progresses along the blue path. This is precisely the reverse process to that displayed in Figs. 3(a) and 3(b). In its journey, the hydroxide may also take some steps that do not transform disordered rings into S 3 ordered ones, changing disordered rings into other disordered ones. Al-ternatively, this quenched evolution may also enlarge an existing ordered cluster by expanding its frontier, where again disordered rings turn into S 3 ones. All three possibilities were observed and appear in the snapshot taken from the simulation of Fig. 4(c). The net total result is the nucleation of ordered mustard rings along the blue path of the hydroxide.
V. DISCUSSION
We have described how FE and PE states transform into one another in a bare bone lattice model of ice, where only IRs and near-neighbor dipolar interactions are retained.
This model, it should be clear, has no ambition of describing real water ice in all chemical details, a field in itself whose literature is immense. The model however, is amenable to solution by simulation; and that makes it, as is often the case, quite instructive.
First we find, by means of an adequate MC protocol, that there is an equilibrium first order phase transition between the two states, with the correct Pauling entropy jump and an instructive proton ring distribution in the FE and PE states. The equilibrium transformation between the two does not take place by regular nucleation as in normal first order transitions because, as is known for a very long time, IRs make ordinary nucleation 50 ineffective: leaving pure, defect free bulk ice in a metastable PE state endowed by Pauling's entropy and a very characteristic proton ring distribution down to the lowest temperatures.
By introducing impurities, mimicking dopants such as KOH known experimentally to nucleate the transition, we examine the very special FE-PE and PE-FE heterogeneous nucleation mechanism in an IR-obeying system . The dopant generates an itinerant hydroxide-induced defect whose string-like evolution inside the bulk effectively punctures, as it were, the otherwise infinite barrier between the two states, ending the kinetic invulnerability of the metastable PE state at low temperatures. Starting with the PE state, the growth of the hydroxide string provides a quasi one-dimensional heterogeneous nucleation mechanism, with a propagating winding cloud of FE rings inside the initially proton-disordered bulk. This is in turn reflected by the increase of the FE order parameter as time (in our case MC time) evolves after quenching, as simulations show (Fig. 2(a)). Snapshots in Fig. 4 (and Fig. 3) moreover show a predicted FE (PE) nucleation landscape proceeding along the string of flipped protons which acts as the backbone. These strings and in fact the qualitative nature of the nucleation process are reminiscent of Kasteleyn-like transitions in spin ice modelsnot surprisingly, because the ordered (disordered) phase onset is again IR-dominated 30 . Nonetheless, the differences are important. Already at equilibrium, Fig. 1(d) compares the temperature dependence of the order parameter of the ice model with that of a spin ice model (see Gohlke et al. 42 ) with the same IRs but with an external field instead of our local dipole-dipole interaction (i.e., with J = 0 in our language). In spin ice there is a 3D Kasteleyn transition in the low-field regime with a characteristic second-order behavior at T > T c , in contrast to the first-order behavior of our phase transition. Beyond that, the evolution kinetics of strings in our ice model is controlled by J, again an element absent in spin ice models.
A number of results suggested by the present lattice ice model encouragingly resemble those known either experimentally or in more elaborate off-lattice models of water ice.
The static structure and ring correlations and the correct Pauling entropy of clean bulk ice appear to describe well the disordered PE state, as summarized by Fig. 1. The capability of metal hydroxide dopants to give rise to growing FE strings inside the PE state and viceversathus functioning as unconventional inhomogeneous nucleation agents -is demonstrated, as in Figs. 3 and 4. The long-time FE polarization fraction grows as the quenching temperature increases approaching T c (Fig. 2(b)), in nontrivial agreement with neutron diffraction experiments and contrary to usual ferrodistortive structural transitions where clusters with reversed order parameters below T c lead to a decrease of the average order parameter as T increases approaching the transition 51-53 . Again similar to real ice, the dependence of FE polarization upon the dopant concentration is minimal. In the model, inhomogeneous nucleation occurs with any number of extrinsic centers, and the residual increasing effectiveness appears simply to reflect a speed-up kinetics once the system is close to the transition point. Finally, the slowing down in the growth rate of FE polarization order parameter with MC time also resembles that observed in water ice in real time-see Fig. 2. In that slowing down however a multiplicity of elements can be simultaneously at work. The string nuclei cannot, owing to their nanoscale transverse size, convert, in a system with strict IRs, a PE state to complete ferroelectricity. Probably even more important in practice, string nuclei might suffer a decrease of their growth rate when their tips hit other existing FE clusters, or, in real ice, grain boundaries and other lattice defects. In our simulations, that kind of effect is involuntarily introduced by finite size. Even ignoring these important realistic aspects, one could note that the partially polarized system free energy is progressively closer to the FE equilibium state than the disordered starting point, yielding a decreasing thermodynamic force felt by the growing string tip ends.
Beyond purely on-lattice models like ours, the moving hydroxide can, besides moving on in-lattice configurations 27,28 , also visit (and be arrested by) offlattice interstitial configurations 29 . That event, not described in our model, will slow down the hydroxide mobility and also introduce kinks with possible bifurcations in the strings evolution. An evolution which nevertheless our model depicts in its most elementary form.
In conclusion, we have presented a soluble lattice model depicting the onset of ice ferroelectricity as a first order phase transition, and demonstrating how nucleation and growth mechanisms, otherwise universal in the kinetics of first order phase transitions, are profoundly changed by topological ice-rule constraints that control proton ordering. Results sheds light on, and support further understanding of, the onset and demise of ferroelectricity in ice.
|
2020-12-24T09:04:27.455Z
|
2020-12-21T00:00:00.000
|
{
"year": 2020,
"sha1": "f97aab412da75597678f4a748e75e94fa1aa9a40",
"oa_license": null,
"oa_url": "https://www.pnas.org/content/pnas/118/1/e2018837118.full.pdf",
"oa_status": "BRONZE",
"pdf_src": "Arxiv",
"pdf_hash": "9dbd218359b0dc3eefe1b037983788c82e873c80",
"s2fieldsofstudy": [
"Physics",
"Environmental Science"
],
"extfieldsofstudy": [
"Medicine",
"Physics"
]
}
|
227305646
|
pes2o/s2orc
|
v3-fos-license
|
An Extension of Tychonoffs Fixed Point Theorem to Quasi-point Separable Topological Vector Spaces
In this paper, we introduce the concepts of m-quasiconvex, originally m-quasiconvex,and generalized m-quasiconvex functionals on topological vector spaces. Then we extend the concept of point separable topological vector spaces (by the topological dual spaces) to quasi-point separable topological vector spaces by families of generalized m-quasiconvex functionals. We show that every pseudonorm adjoint topological vector space, which includes locally convex topological vector spaces as special cases, is quasi-point separable. By the Fan-KKM theorem, we prove a fixed point theorem on quasi-point separable topological vector spaces. It is an extension of the fixed point theorem on pseudonorm adjoint topological vector spaces proved in [8]. Therefore, it is a properextension of Tychonoffs fixed point theorem on locally convex topological vector spaces, which are demonstrated by some examples.
Introduction
In 1935, Tychonoff generalized the Schauder's fixed-point theorem from Banach spaces to locally convex topological vector spaces. Since then, many authors have extensively studied the properties of locally convex topological vector spaces. Some counter examples of topological vector spaces, which are not locally convex, were provided (see [1−4], [8−9], [12], [14−16]). In 2020, the present author introduced the concept of pseudonorm adjoint topological vector spaces, on which a fixed point theorem is proved. Meanwhile a concrete example was provided in to show that is a proper extension of the Tychonoff's fixed point theorem (see [8]).
Suppose that a pseudonorm adjoint topological vector space (X, ) is equipped with a family of -continuous quasi-pseudonorms. In [8], we notice that the totality of the family of -continuous quasi-pseudonorms plays a very important role for the existence of fixed points of continuous mappings on nonempty convex and compact subsets of X. It provides very useful ideas to extend the pseudonorm adjoint topological vector spaces to more general topological vector spaces.
In contrast, of the concept of point separable topological vector spaces by their topological dual spaces, in this paper, we use the term "quasi-point separable" for totality, which means that the spaces are point separable by families of generalized m-quasiconvex functionals. It is a generalization of the concept of point separable topological vector spaces (see section 3).
By using the concept of families of generalized m-quasiconvex functionals, we introduce quasipoint separable topological vector spaces. Similar to the fixed point problems on pseudonorm adjoint topological vector spaces studied in [8], quasi-point separable topological vector spaces have the following properties to ensure the fixed point property (see sections 3 and 4 for details): (a) In [8] and some other papers, the families of generalized m-quasiconvex functionals that point separates the considered topological vector spaces are also said to be total; (b) The family of generalized m-quasiconvex functionals is not required to induce (define) the original topology; (c) In the definition of quasi-point separable topological vector spaces, we have (i) the mquasiconvex functionals are continuous with respect to the original topology; (ii) the generalized m-quasiconvex functionals and the weighted functions are not required being continuous; (d) Every pseudonorm adjoint topological vector space is a quasi-point separable topological vector space; (e) Every locally convex topological vector space is a quasi-point separable topological vector space.
From (d) above, quasi-point separable topological vector spaces are extensions of pseudonorm adjoint topological vector spaces. Meanwhile, from section 3, we see that the constructions of quasi-point separable topological vector spaces are similar to that of pseudonorm adjoint topological vector spaces. Even though that the extension seems not very significant, but the conditions for a topological vector space to be quasi-point separable are easier checking. Hence, we believe that it is worth to write the new ideas of quasi-point separable topological vector spaces.
By using the Fan-KKM theorem, some authors (see [5−8]) proved the Tychonoff's fixed point theorem, the Brouwer's fixed-point theorem and the Schauder's fixed-point theorem; Park in [10−13] proved some fixed point theorems; and the present author in [8] proved a fixed point theorem on pseudonorm adjoint topological vector spaces. In this paper, we apply the Fan-KKM theorem to prove a fixed point theorem on quasi-point separable topological vector spaces. It is indeed a proper extension of the Tychonoff's fixed point theorem, which is on locally convex topological vector spaces.
Preliminaries
In this section, we first recall the concepts about pseudonorm adjoint topological vector spaces defined in [8].
Definition 2.1 [8]
Let X be a vector space with origin . A mapping p: X→ ℝ + is called a pseudonorm on X if it satisfies the following conditions: W3. For any elements x1, x2 of X, and 0 ≤ ≤ 1, one has Definition 2.2 [8]. Let X be a vector space. A mapping q: X → ℝ + is called a quasi-pseudonorm on X if there are a pseudonorm p on X and a strictly increasing continuous function : Here, q is said to be adjoint with the pseudonorm p and the weighted function .
Definition 2.3 [8].
Let (X, ) be a topological vector space. If X is equipped with a family of -continuous quasi-pseudonorms { } ∈Λ associated with a family of -continuous pseudonorms { } ∈Λ and a family of weighted functions { } ∈Λ , then (X, ) is called a pseudonorm adjoint topological vector space. Definition 2.4 [8]. A family of quasi-pseudonorms { } ∈Λ equipped on a topological vector space (X, ) is said to be total whenever, for x ∈ X, (x) = 0 holds, for every ∈ Λ, then it is necessary to have x = . A pseudonorm adjoint topological vector space is said to be total if it is equipped with a total family of quasi-pseudonorms.
Lemma 2.5 [8]. Every locally convex topological vector space is a pseudonorm adjoint topological vector space.
The, in rest of this section, we briefly recall the definitions of point separable topological vector spaces and give the definition of m-quasiconvex functionals, by which, we define quasi-point separable topological vector spaces.
Let (X, ) be a topological vector space with origin and let X* denote the dual space of X (the vector space of linear and continuous functionals on X). If, for x ∈ X, (2.1) then, X is said to be point separable (by its dual space X*), or X is called a point separable topological vector space.
(2.1) is equivalently defined as follows: X is point separable (by its dual space X*), if and only if, for any distinct elements x, y ∈ X, there is h ∈ X* such that h(x) ≠ h(y).
It is well known (see [11], or see Theorem 2.18 in [4]): Proposition 2.6. Every Hausdorff locally convex topological vector space is point separable.
M-quasiconvex functionals and quasi-point separable topological vector spaces
Throughout this section, unless otherwise is stated, let (X, ) be a topological vector space.
Definition 3.1. Suppose that u is a real valued functional defined on X. For any x1, x2 ∈ X, and 0 ≤ ≤ 1, then, u is said to be quasiconvex.
(ii) if u satisfies that then, u is said to be m-quasiconvex.
Here the letter m in front of the word quasiconvex means that it is with respect to magnitude. Similar to the property of quasiconvexity, the m-quasiconvexity of u is equivalent to the following fact, for any nonnegative number a, the set {x ∈ X: | ( )| ≤ a} is a convex subset of X.
Definition 3.3.
If u satisfies that, for any x1, x2 ∈ X, and 0 ≤ ≤ 1, then, u is said to be absolutely convex.
We provide the following two counter examples to show that, neither the quasiconvexity, nor the m-quasiconvexity includes other one.
Example 3.4.
Let u be a function on ℝ defined as Then, u is convex on ℝ, so is quasiconvex. But u is not m-quasiconvex.
Example 3.5.
Let u be a function on ℝ defined as Then, u is m-quasiconvex. However, u is not quasiconvex. Proof. The proof is straightforward and it is omitted here.
Lemma 3.7.
Suppose that u is an originally m-quasiconvex functional on X. Then, for any x ∈ X, the function | ( )| is an increasing function with respect to t > 0.
Proof. Take arbitrary 0 < t1 < t2. Then Proof. The proof is straightforward and it is omitted here. Lemma 3.10. Every -continuous pseudonorm p: X → ℝ + is -continuous originally m-quasiconvex on X.
Lemma 3.11. Let q be a -continuous quasi-pseudonorm on X adjoint with a pseudonorm p and associated with a continuous weighted function defined in [8]. Then q is -continuous generalized m-quasiconvex on X.
Proof. The proof is straightforward and it is omitted here. Proof. The proof follows from Lemma 3.9 immediately. From Lemmas 2.6 and 3.14, we immediately have that To prove (4.2), we first prove that, for any > 0, For > 0 as given above, assume, on the contrary, that (4.3) does not hold; that is, Based on the mapping f, we define a set-valued mapping F: C →2 \{∅} as follows: From the hypothesis (4.9), we see that x ∈ F( ), and therefore, F( ) ≠ ∅, for every x ∈ C. The -continuity of f: C → C and the -continuouity of the m-quasiconvex functionals { } ∈Λ imply that, for every x ∈ C, F( ) is -closed. Next, we show that the mapping F: C →2 \{∅} is a KKM mapping.
In particularly, if we take x = f(z0)) ∈ C, we get It is a contradiction. So equipped with a quasi-point separating space { , , } ∈Λ needs to be Housdorrf. It is worth to study whether or not that quasi-point separable property of topological vector spaces implies the Housdorrf property. It lefts to interested authors to consider.
In [12], Park studied fixed point problems on point separable topological vector spaces for a class of functions, such as half-continuous functions. Here, we use Theorem 4.1 and Lemma 3.10 to obtain the following fixed point theorem. Since every Hausdorff locally convex topological vector space is a point separable topological vector space, the following theorem is also an extension of the Tychonoff's fixed point theorem to point separable topological vector spaces. Tychonoff's fixed point theorem [12]: Let X be a Hausdorff locally convex topological vector space. For any nonempty compact convex set C in X, any continuous function f : C → C has a fixed point.
In this section, one-step further, we give a simple proof showing that lr, is point separable; and from Lemma 3.9 or 3.14, it follows immediately that lr, for 0 < r < 1, is quasi-point separable. Define a functional q on lr as It is known that the functional q induces a metric on lr by , for every { }，{ } ∈ lr. Hence, the space lr is point separated by { }. Therefore, lr is point separable (by * ). , for { } ∈ l r .
One can show that the functional p induces a metric on the space l r as follows: , for every { }，{ } ∈ l r . (5.3) Let l r * denote the topological dual space of l r . Then, the metric vector space l p is point separable (by l r *); so is quasi-point separable with a quasi-point separating space { , , ℝ + } ∈N .
Proof. To show that the functional defined in (5.3) defines a metric on l p , one may consider the following steps: (a) f(t) = 1+ is a strictly increasing function on ℝ + ; (b) (t + s) r ≤ t r + s r , for t, s ∈ ℝ + (this should be proved in Example 5.1); (c) , for t, s ∈ ℝ + .
Rest of the proof is almost the same to the proof of Example 5.1. For every n = 1, 2, … , define a functional un on l r as in (5.2). The space l r is point separated by { }. From { } ⊆ l r *, l p is point separated by l r * and l r is point separable, so is quasi-point separable.
|
2020-12-07T02:00:59.055Z
|
2020-12-04T00:00:00.000
|
{
"year": 2020,
"sha1": "96fc3c1850ec2b12975160be449752d3a0e5392b",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Arxiv",
"pdf_hash": "96fc3c1850ec2b12975160be449752d3a0e5392b",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.