Split (1)

train · 100k rows

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
209446998	pes2o/s2orc	v3-fos-license	Mild Cognitive Impairment in the Migrant Population Living in Europe: An Epidemiological Estimation of the Phenomenon Background: The construct of mild cognitive impairment (MCI) is triggering growing clinical and research interest. The detection of MCI may be affected by diverse ethno-cultural determinants possibly influencing the personal and social perception of the individual cognitive functioning as well as the reliability of objective cognitive assessment. These challenges may acquire special relevance in subjects with a migration background and composing ethnic minority groups. Objective: The present study is aimed at providing an estimate of the number of MCI cases occurring in the migrant population living in the extended European Union (EU) in 2018. Methods: The number of MCI cases in older migrants living in Europe and in each of the 32 considered countries was estimated by multiplying the number of migrants, provided by Eurostat, with the age-specific prevalence rates, derived by the harmonized data produced by the COSMIC collaboration and based on different operational definitions of MCI. Results: Nearly 686,000 cases of MCI were estimated in the extended EU by applying age-specific prevalence rates based on the International Working Group criteria. Higher figures were obtained when the Clinical Dementia Rating- and the Mini Mental State Examination-based criteria were applied. The proportion of MCI cases in migrant subjects ranged from 1.1% (Romania) to 54.1% (Liechtenstein) (median: 8.4%; IQR: 4.7%–14.2%). Conclusions: MCI represents and will increasingly constitute a relevant issue in the migrant population living in Europe. The present data reinforce the need of developing approaches and models of care that may be diversity-sensitive and inclusive for a culturally variegated population. INTRODUCTION Mild cognitive impairment (MCI) is commonly intended as a decline in the individual's cognitive performances not resulting in a significant reduction of functional independence and social or occupational functioning [1]. It is therefore conceived as an intermediate stage between normal cognition and dementia [2]. MCI is stimulating growing research and clinical interest. In fact, based on the available evidence, it represents a robust risk factor for future dementia [3]. At the same time, it is increasingly regarded as a promising phase for implementing dementia prevention strategies, as also indirectly suggested by the observed potential for clinical improvement/reversion to normal cognition [4]. As a proof, 349 randomized controlled trials testing novel pharmacological or non-pharmacological interventions targeting MCI are currently registered on the clinicaltrials.gov website (as of June 2019). The detection of MCI is commonly triggered by a concern/complaint regarding a change in cognition from the subject, an informant, or a clinician. The clinical diagnosis then requires a standardized evaluation of the person's cognitive functioning with the aim of providing the evidence of an objective impairment, in relation to normative parameters, in one or more cognitive domains [1]. Accordingly, cognitive testing has a central role in the so far adopted operational definitions of this condition [1,5]. Moreover, the neuropsychological assessment is essential to properly characterize and classify this clinical construct (e.g., amnesic MCI versus non-amnesic MCI; single-domain MCI versus multi-domain MCI) and to more precisely stratify the individual risk profile in terms of conversion to dementia [6]. The identification and management of MCI may, however, be affected by diverse ethno-cultural determinants possibly influencing the personal and social perception and judgment of the one's cognitive functioning as well as the reliability of objective cognitive assessment. These reflections may therefore assume a special relevance in migrants and subjects composing minority groups. In fact, it has been shown that these individuals may have different attitudes toward mental and cognitive disturbances [7,8] and frequently have delayed contact with dedicated healthcare and social services [9]. Moreover, in Western countries, there is still a paucity and a scarce adoption of instruments and tools supporting a culture-sensitive (and, thus, more reliable) evaluation of cognitive performance [10,11]. The considerations are gaining further relevance in light of the ongoing sociodemographic transitions consisting in the progressive aging not only of native populations but also of subjects with a migration background [12]. The aim of the present study is to provide an estimate of the number of MCI cases occurring in the migrant population living in the extended European Union (EU) in 2018 to start gaining awareness of this phenomenon. Older migrants in Europe In the present study, "migrants" were operationally defined as those individuals living in a given European country but born abroad, regardless of the length of stay and the causes for the migration [13]. Data provided by the Statistical Office of the European Union, Eurostat (http://ec.europa.eu/eurostat/web/ population-demography-migration-projections/popu lation-data/database; database: "Population on 1 January by age group, sex and country of birth" [migr pop3ctb]) were used to calculate the number of migrants, aged 60 years or older, living in Europe. Information was available for the 28 countries of the EU and the four countries composing the European Free Trade Association (i.e., Iceland, Liechtenstein, Norway, and Switzerland). All data were updated to August 2019 and referred to the subjects living in each country on 1 January 2018. Mild cognitive impairment prevalence rates The age-specific prevalence rates of MCI were derived by the data provided by the Cohort Studies of Memory in an International Consortium (COSMIC) collaboration [14]. These estimates were calculated by applying different MCI classifications to the harmonized data coming from 11 longitudinal population-based studies on cognitive aging from USA, Europe, Asia, and Australia. Specifically, three commonly adopted operational definitions of MCI were applied to the participants aged 60 to 89 years recruited in these studies: 1) International Working Group criteria (IWG) [5]. MCI is identified by the presence of four criteria: absence of dementia (mostly ascertained with the DSM-IV criteria); no or minimal functional impairment (i.e., dependence in ≤two instrumental activities of daily living); subjective memory/cognitive complaints or concerns; and objective cognitive impairment (i.e., a score within the bottom 6.681%, or equivalently more than 1.5 SDs below the mean, of the scores for a given cognitive domain within the relevant study's sample). MCI is further classified into amnesic (aMCI), when the memory domain is impaired, and non-amnesic (naMCI), when the impairment occurs in any of the other cognitive domains. 2) Mini-Mental State Examination (MMSE) score from 24 to 27 (inclusive) [15]. 3) Clinical Dementia Rating (CDR) of 0.5 [16]. Estimated cases of mild cognitive impairment among migrants in Europe The number of MCI cases in migrants from 60 to 89 years old living in Europe and in the 32 considered countries was estimated by multiplying the number of migrants with the age-specific prevalence rates. For each nation, we also estimated the proportion of MCI cases occurring in migrants (calculated as the ratio between the estimated cases in migrants and in the overall population). RESULTS A total of 12,730,960 migrants aged 60-89 years (women 55.1%) lived in Europe in 2018 (Table 1), with national estimates ranging from 3,326 in Iceland to 3,741,052 in Germany (Table 2). Nearly 686,000 cases of MCI were estimated in this population by applying age-specific prevalence rates based on the IWG criteria. Higher figures were obtained when the CDR-and the MMSE-based criteria were applied, increasing to almost 1.1 million and 1.5 million cases, respectively ( Table 1). As evident in Table 2 The proportion of MCI cases in migrant subjects (i.e., the ratio between the estimated cases in migrants and in the general population) ranged from 1.1% (Table 3 and Fig. 1). DISCUSSION To the best of our knowledge, the present study constitutes the first attempt to explore the magnitude of the issue of MCI occurring in migrants living in Europe. Our estimates, combined with those relating to dementia cases in the same population [17], suggest that more than one million migrants living in our continent is expected to be affected by a cognitive disorder, thus potentially referring to clinical and social facilities in the host countries. Accordingly, in several nations, a relevant proportion of cognitive disturbances is probably involving foreign-born individuals with important implications in terms of diagnostic accuracy, provision of care, and social support. These estimates are strongly influenced by the adopted operational definitions of MCI and are projected to markedly increase in the next future due to the current sociodemographic transformations [12]. In this regard, adopting the same analytic methods and considering the more stringent IWG criteria, it can be assumed that the number of MCI cases in migrants living in the extended EU has shown a 34%increase in just 4 years, passing from 511,624 in 2014 to 685,902 in 2018. Some potential limitations should be considered when interpreting these findings. First, the existence of possible differences in terms of MCI prevalence across diverse ethnic groups could not be neglected. We assumed that both natives and individuals migrating from different World regions shared the same risk of MCI, thus applying the same age-specific prevalence rates to the heterogeneous autochthonous and immigrant populations. Indeed, there is emerging evidence that cognitive disturbances may be more prevalent in specific groups and ethnicities. For instance, a study recruiting 2,254 participants aged 55 years or older in the Netherlands revealed that MCI was three times more frequent in most non-western migrants compared to the native Dutch population. In addition, a relevant variability of MCI prevalence across different immigrant groups was observed [18]. Such heterogeneity can be attributed to various determinants (e.g., vascular risk factors, educational level, lifestyles, physical activity, social interactions [19][20][21]) that have robustly been associated with the risk of cognitive disturbances and that were unavailable for the present analysis. The decision to base our analysis on the data produced by the COSMIC collaboration [14] (and not on other MCI prevalence data available in the literature) stems from the aim of providing different estimates of the phenomenon of interest according to alternative operationalizations of MCI that are widely adopted in the routine practice. For instance, the MMSE, despite being poorly sensitive and not recommended for the detection of MCI [22], is still largely used for its identification in clinical settings [23]. This approach has allowed us to frame the phenomenon taking into account the variability in its measurement. However, it should be noticed that the present estimates are not far from those that can be obtained from more recent MCI prevalence data. In fact, nearly 1,350,000 MCI cases can be estimated in the same European migrant population by applying the agespecific rates provided by the American Academy of Neurology (AAN) guideline on MCI, derived from studies targeting both MCI and related constructs (e.g., cognitive impairment no dementia (CIND)) [24]. Unfortunately, the adopted prevalence rates were available only for subjects aged less than 90 years. Therefore, we could not estimate the number of MCI cases in the oldest migrants, thus producing a possible underestimation of the phenomenon. The choice of not presenting our findings separately for men and women was instead supported by the fact that, in the pool analysis used as reference, MCI prevalence rates were substantially unaffected by sex [14]. In conclusions, MCI represents and will increasingly constitute a relevant issue in the migrant population living in Europe. This phenomenon remains to be characterized at the "real-world" level, thus merging the present epidemiological estimates with information coming from clinical and social services. Our data should inform clinicians, researchers, and policymakers on the need of developing approaches and models of care that may be diversity-sensitive and inclusive for a culturally variegated population. Given the centrality of the neuropsychological assessment in the detection of MCI, cross-cultural tools for the cognitive assessment should increasingly be used. In parallel, the possible role and involvement of professionals like interpreters and cultural mediators in the field of cognitive disturbances should be considered. Moreover, a greater effort should be made in order to understand the migrants' attitudes, beliefs, and perceptions toward cognition and cognitive disorders. Marco Canevelli is supported by a research grant of the Italian Ministry of Health for the project "Dementia in immigrants and ethnic minorities living in Italy: clinical-epidemiological aspects and public health perspectives" (ImmiDem)(GR-2016-02364975).	2019-12-19T09:17:21.967Z	2019-12-16T00:00:00.000	{ "year": 2019, "sha1": "3d87228394f7cddb85243547b2dd9d84261b3dde", "oa_license": "CCBY", "oa_url": "https://content.iospress.com/download/journal-of-alzheimers-disease/jad191012?id=journal-of-alzheimers-disease/jad191012", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "8aebddd6f1fdc1a8318ea4b64e9e92dab5ef5034", "s2fieldsofstudy": [ "Medicine", "Sociology" ], "extfieldsofstudy": [ "Medicine", "Psychology" ] }
237206294	pes2o/s2orc	v3-fos-license	Predicted B Cell Epitopes Highlight the Potential for COVID-19 to Drive Self-Reactive Immunity COVID-19, caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), whilst commonly characterised as a respiratory disease, is reported to have extrapulmonary manifestations in multiple organs. Extrapulmonary involvement in COVID-19 includes autoimmune-like diseases such as Guillain-Barré syndrome and Kawasaki disease, as well as the presence of various autoantibodies including those associated with autoimmune diseases such a systemic lupus erythematosus (e.g. ANA, anti-La). Multiple strains of SARS-CoV-2 have emerged globally, some of which are found to be associated with increased transmissibility and severe disease. We performed an unbiased comprehensive mapping of the potential for cross-reactivity with self-antigens across multiple SARS-CoV-2 proteins and compared identified immunogenic regions across multiples strains. Using the Immune Epitope Database (IEDB) B cell epitope prediction tool, regions predicted as antibody epitopes with high prediction scores were selected. Epitope sequences were then blasted to eight other global strains to identify mutations within these regions. Of the 15 sequences compared, eight had a mutation in at least one other global strain. Predicted epitopes were then compared to human proteins using the NCBI blast tool. In contrast to studies focusing on short sequences of peptide identity, we have taken an immunological approach to selection criteria for further analysis and have identified 136 alignments of 6–23 amino acids (aa) in 129 human proteins that are immunologically likely to be cross-reactive with SARS-CoV-2. Additionally, to identify regions with significant potential to interfere with host cell function-or promote immunopathology, we identified epitope regions more likely to be accessible to pathogenic autoantibodies in the host, selected using a novel combination of sequence similarity, and modelling protein and alignment localization with a focus on extracellular regions. Our analysis identified 11 new predicted B-cell epitopes in host proteins, potentially capable of explaining key aspects of COVID-19 extrapulmonary pathology, and which were missed in other in silico studies which used direct identity rather than immunologically related functional criteria. INTRODUCTION In March 2020, the World Health Organization (WHO) declared the disease known as COVID-19, caused by the Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV-2), as a global pandemic. As of the 2nd of August 2021, there have been over 198.2 million confirmed cases and over 4.3 million deaths reported (WHO, 2020a). COVID-19 patients typically present with symptoms such as fever, dry cough, tiredness, as well as others such as myalgia, sore throat, loss of taste or smell, red eyes, diarrhoea and skin rash (Huang et al., 2020;Meng et al., 2020;WHO, 2020b). The SARS-CoV-2 genome encodes a series of structural and non-structural proteins (Chan et al., 2020;Wu et al., 2020). There are four structural proteins consisting of the surface glycoprotein (spike, SP), nucleocapsid phosphoprotein (nucleoprotein, NP), membrane (M) and envelope (E). There are additionally eight open reading frames (Orf) encoding the nonstructural proteins: Orf1ab, Orf3a, Orf3b, Orf6, Orf7a, Orf7b, Orf8 and Orf10. Multiple studies have used serological assays to measure antibody responses to the SARS-CoV-2 virus (Long et al., 2020;Zhao et al., 2020;Post et al., 2021). Antibodies are usually evaluated for reactivity against the spike and the nucleoprotein with the aim of understanding seroconversion, as well as correlating disease severity with neutralizing antibody titres (systematically reviewed in Post et al. (2021)). Additional reports have identified increases in antibodies to other SARS-CoV-2 proteins, including Orf proteins, after infection (Hachim et al., 2020). Since the beginning of the pandemic, variants of SARS-CoV-2 have been emerging and circulating worldwide (van Dorp et al., 2020). Some mutations, such as K417N and E484K in the spike protein, are immune evasion mutations (Altmann et al., 2021). Since the second half of 2020 variants of SARS-CoV-2, containing mutations such as these, have arisen with increased transmissibility, associated with more severe disease and a reduction in neutralizing antibodies. These variants have been called variants of concern (VOC) or variants of interest (VOI). As of July 2021, there are four VOCs and four VOIs classified by the Centers for Disease Control and Prevention (CDC) (SARS-, 2021). The VOCs consist of the strains originating in the United Kingdom (Alpha strain), Brazil (detected in Japan, Gamma strain), South Africa (Beta strain), and India (Delta strain). Whereas VOIs consist of strains originating in Canada (Eta strain), United States (Iota strain), India (Kappa strain) and Peru (Lambda strain). Viral infections generally can play a role in promoting or exacerbating autoimmune diseases (reviewed in Smatti et al. (2019)). Proposed mechanisms for this association include molecular mimicry and bystander activation (Fujinami et al., 2006;Smatti et al., 2019). Molecular mimicry refers to the phenomenon of immune cross-reactivity between a foreign pathogen and self-protein, where an immune cell recognizes both due to their sequence similarity (Fujinami et al., 2006;Smatti et al., 2019). Bystander activation refers to the activation of self-reacting immune cells driven by the liberation of self-antigens (the targets which stimulate immune responses) from virus-lysed cells which otherwise would not be typically exposed to the immune system (Fujinami et al., 2006;Smatti et al., 2019). Given that a range of extrapulmonary pathologies have been reported in COVID-19 (Cheng et al., 2020;Cheung et al., 2020;Han et al., 2020;Hundt et al., 2020;Nalleballe et al., 2020;Oxley et al., 2020;Poyiadji et al., 2020;Xie et al., 2020;Zheng et al., 2020), molecular mimicry has been hypothesized to be playing a role (Angileri et al., 2020a;Cappello et al., 2020;Marino Gammazza et al., 2020). Additionally, conditions similar to the autoimmune diseases Guillain-Barré Syndrome (Ameer et al., 2020;Korem et al., 2020) and Kawasaki disease (Bordet et al., 2020;Jones et al., 2020) have been reported in COVID-19 patients. Within this Kawasaki-like disease, termed Multisystem Inflammatory Syndrome in children (MIS-C), autoantibodies targeting a range of antigens including, but not limited to, anti-La, anti-Jo-1, anti-MUC15, anti-P2RX4, anti-MAP2K2 and anti-CSNK1A1 have been reported (Consiglio et al., 2020;Gruber et al., 2020). Additionally, autoantibodies targeting other antigens including type I interferons (IFNs) (Bastard et al., 2020) have been identified in COVID-19 positive patients (Zhou et al., 2020a;Bastard et al., 2020;Vlachoyiannopoulos et al., 2020) alongside key autoimmune associated antigens which are often associated with tissue damage, for example, anti-nuclear antigen (ANA) (Zhou et al., 2020a;Vlachoyiannopoulos et al., 2020). It is therefore highly likely autoantibodies that recognize a range of other self-proteins are induced by infection with SARS-CoV-2. Computational biology methods such as immune epitope mapping and Basic Local Alignment Search Tool (BLAST) allow for relatively quick predictive screening analysis. They can help narrow findings and hypotheses, leading to more targeted laboratory-based validation work. In the context of COVID-19, understanding immune epitopes, the sequences that are recognized by antibodies, have been used to predict potential peptide-based vaccine candidates or to increase sensitivity in serological assays (Ahmed et al., 2020;Amrun et al., 2020;Poh et al., 2020). In the initial stages of the COVID-19 outbreak, B-cell epitope mapping was performed in SARS-CoV-2 to check for potential cross-reactivity of immune responses with other coronaviruses (Ahmed et al., 2020;Grifoni et al., 2020). Recently, there have been a small number of studies reporting sequence similarities between SARS-CoV-2 proteins and human proteins, in support of the molecular mimicry hypothesis (Angileri et al., 2020a;Angileri et al., 2020b;Ehrenfeld et al., 2020;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Lyons-Weiler, 2020;Marino Gammazza et al., 2020). With the exception of one study, which used longer epitopes (Lyons-Weiler, 2020), most of these studies sought to find short regions (5-6aa) of exact identity between the SARS-CoV-2 proteins and human proteins (Angileri et al., 2020a;Angileri et al., 2020b;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020). Whilst these studies suggested a number of regions, most of the identical sequences identified would be unlikely to sustain a pathogenic autoreactive response (Angileri et al., 2020a;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020). Specifically, while assessing its potential accessibility for antibody binding, these studies did not discriminate the localization of the epitope within the protein or within the cell (Angileri et al., 2020a;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Lyons-Weiler, 2020;Marino Gammazza et al., 2020). Moreover, except in the case of Kanduc (2020), who identified identical hexamers between SARS-CoV-2 and human proteins associated with various system disorders, these studies only report small numbers of human proteins with similar (Lyons-Weiler, 2020) or short identical sequences (Angileri et al., 2020a;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020) to SARS-CoV-2 proteins (e.g. association with the immune system, molecular chaperones, brainstem proteins). Therefore, the limited nature of these studies does not show the full extent of similarities between SARS-CoV-2 and human proteins. In the present study we have pioneered a different immunoinformatics approach to overcome some of these limitations and further explore the potential for autoimmune cross-reactivity driven by SARS-CoV-2 molecular mimicry. B-cell epitopes within the structural proteins and several Orf proteins (shown to generate antibody responses in COVID-19 positive patients (Hachim et al., 2020)), were first predicted using the IEDB B Cell epitope tool. Epitopes predicted in the structural proteins were compared to the current VOCs (Alpha, Beta, Gamma and Delta) and VOIs (Eta, Iota, Kappa and Lambda). Eight of the fifteen epitopes contained at least one mutation in one variant. The identified 9-53aa long sequences, in both structural and Orf proteins, were then compared to sequences in human proteins using the NCBI protein BLAST tool (Altschul et al., 1990). Assessment of potential for cross-reactivity between SARS-CoV-2 and self-proteins with capacity to perpetuate autoimmune pathology was based on a combination of immunologically relevant sequence similarity (not just identity) (Angileri et al., 2020a;Kanduc, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020) and the localization of the protein itself, with a focus on extracellular targets. 11 human proteins, containing amino acid sequences similar to nine predicted SARS-CoV-2 B cell epitopes, were identified based on these selection criteria. These findings indicate that antibodies induced by SARS-CoV-2 could directly interfere with cell function, including that of immune cells, and could help explain some of the additional pathologies identified in COVID-19 patients (Cheng et al., 2020;Cheung et al., 2020;Han et al., 2020;Hundt et al., 2020;Nalleballe et al., 2020;Oxley et al., 2020;Poyiadji et al., 2020;Xie et al., 2020;Zheng et al., 2020). Finally, comparing the sequences of both predicted spike epitopes and full length spike protein to various human proteins implicated in immune thrombocytopenia purpura (ITP) and thrombocytopenia syndrome (TTS), our results indicate that molecular mimicry is unlikely to be the cause of TTS, or vaccine induced prothrombotic immune thrombocytopenia (VIPIT) following vaccination with the COVID-19 adenovirus vector vaccines. To our knowledge, this is the first study to compare immune epitopes across the circulating VOCs and VOIs; highlight multiple similarities between the selected Orf proteins and human proteins; identify proteins with reported associations to autoimmunity as sharing sequences with SARS-CoV-2 epitopes; and to highlight novel extracellular human proteins which may have antibody cross-reactivity with SARS-CoV-2 immunogenic regions. Research Pipeline to Explore Potential Immune Cross-Reactivity In order to explore the similarities between potential immunogenic SARS-CoV-2 regions and human proteins, we applied a research pipeline as outlined in Figure 1. B cell epitopes were first predicted within a selection of SARS-CoV-2 proteins. Those selected were then compared to the human proteome. To explore potential immune cross-reactivity, we applied novel criteria that not only considered short identical sequences but also amino acid variations with conserved structural and charge changes that may not impact antibody binding. We also considered protein localization, with a focus on extracellular proteins. The alignments of interest found within human proteins were cross-checked as potential epitopes within their own protein sequence. Details of each step described below. Sequence Identification and Epitope Mapping Based on previous mapping and assay-based studies (Hachim et al., 2020;Grifoni et al., 2020), epitope prediction was carried out on proteins previously identified from the SARS-CoV-2 isolate Wuhan-hu-1 (Figure 1; Supplementary Table S1) (GenBank: MN908947.3) . As Orf3b is produced by a premature stop codon (Gordon et al., 2020;Lam et al., 2020) and the sequence could not be identified on GenBank, the sequence provided in Lam et al. (2020) was used for epitope mapping. Linear B cell epitope predictions were performed using the B cell Epitope prediction tool found within Immune Epitope Database and Analysis Resource (IEDB, http://tools.iedb.org/ bcell/ ). Using the Bepipred Linear Epitope Prediction algorithm (Haste Andersen et al., 2006;Larsen et al., 2006;Ponomarenko and Bourne, 2007), predicted epitopes for structural proteins (Spike, Envelope, Membrane and Nucleoprotein), were selected based on prediction score (>1) and having length greater than 6aa. Epitopes for the Orf proteins were predicted using the Bepipred Linear Epitope Prediction 2.0 algorithm (Jespersen et al., 2017) and were selected for having the highest peak points (prediction score >0.5) of length greater than 6aa. Each algorithm was selected based on the IEDB recommended at the time epitope mapping was performed, and prediction score cut offs to select for those more likely to be real epitopes. Where possible, protein structures were downloaded from the protein data bank rcsb.org (Berman et al., 2000). For SARS-CoV-2 proteins, where regions were uncrystallised, homology models were created using the MPI Bioinformatics Toolkit (https:// toolkit.tuebingen.mpg.de/). Specifically, templates were found using HHpred and models were created using Modeller (Zimmermann et al., 2018). Any unstructured regions were deleted and figures were created in The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC. All available human protein models were downloaded from the SWISS-MODEL Repository (Bienert et al., 2017). Sequence Alignment to Other SARS-CoV-2 Variants For SARS-CoV-2 variants protein alignment, we used eight different variants and the original Wuhan sequence (Supplementary Table S2). The FASTA sequences were retrieved from the GISAID database (https://www.gisaid.org/). In the EpiCov search section of the GISAID database there is an available tab that allows for selection of the major circulating variants. Of these we selected the current VOCs (Alpha, Beta, Gamma and Delta) and VOIs (Eta, Iota, Kappa and Lambda). The virus names listed in Supplementary Table S2 are GISAID nomenclature and the specific viruses were selected based on various conditions: for all variants we selected the conditions in which the sequence was complete, we excluded sequences with low coverage (>5% unidentified amino acids) and selected the specific clad of the variant being analyzed. For all variants, each one was chosen within a specific date based on their historical appearance. From these we selected the ones that underwent Illumina sequencing and for which the assembly method was clearly reported. In order to obtain the specific amino acid sequences of the various genomic regions based on the B cell epitope prediction, we retrieved the FASTA sequence of the entire viral genome from the selected variants and blasted these using BlastN by selecting standard databases and optimizing for highly similar sequences (megablast). Since not all submissions to GISAID have also been submitted to NCBI some did not exhibit 100% sequence identity. The percentage of sequence identity for each variant is listed in the table below (column NCBI blast identity in Supplementary Table S2). For the ones that exhibited lower than 100% sequence identity, with blastN we visualized the areas in which the differences were laying. Those specific sequences were then blasted to view where in the viral genome they were appearing. Using JalView (V2.11.1.4) (Waterhouse et al., 2009), we aligned the amino acid sequences for the surface glycoprotein, membrane protein and nucleoprotein areas of all eight variants plus the original Wuhan sequence, and set the latter as reference genome. We then performed alignment using the Multiple Sequence Alignment using Fast Fourier Transform (MAFFT) alignment program with default settings (V7.110) (Katoh and Standley, 2013). MAFFT is a high speed multiple sequence alignment program which utilizes the Fast Fourier Transform to optimize protein alignments based on the amino acid physical properties. Sequence BLAST to Compare Epitopes to Human Proteins and Identification of Associated Diseases, Expression Location and Function Using the NCBI protein-protein BLAST (blastp, https://blast. ncbi.nlm.nih.gov/ (Altschul et al., 1990)), predicted epitope FIGURE 1 \| Research pipeline of the project to explore the potential of immune cross-reactivity. Protein sequences for select SARS-CoV-2 proteins were obtained for epitope predictions (highlighted green. Spike: Surface glycoprotein. E, envelope; M, membrane; NP, nucleoprotein). Using the Immune Epitope Database (IEDB) epitope prediction tool, B cell epitopes were predicted and those selected were screened against human proteins using the NCBI Blastp tool. Human proteins with sequence similarities to the SARS-CoV-2 epitopes were investigated for their function, disease association and alignment localization. Potential immune crossreactivity between SARS-CoV-2 and human alignments was then explored by applying specific selection criteria. Alignments of interest from the human proteins were explored as potential epitopes within the human protein sequences. Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 sequences were blasted for the top 100 results, to the nonredundant protein sequences (nr) database, specifically looking at Homo sapiens (Taxid ID: 9606). All other algorithm parameters remained as default settings, including: automatically adjust parameters for short input sequences, and expect threshold 0.05. Resulting lists were narrowed by excluding isoform repeats, identical proteins with different nomenclature, uncharacterized/ hypothetical proteins and variable regions of B and T cell receptors. Associations of the self-proteins to diseases, their expression locations and functions were identified using UniProtKB, GeneCards and PubMed Resources. Any proteins identified through computational analysis or found on an unreviewed UniProtKB database page were further excluded. For any direct epitope to full protein comparison, "align two or more sequences" in the NCBI blastp tool was used. Prediction of Potential Immune Cross-Reactivity Selection Criteria Using the previously narrowed BLAST lists, potential B cell epitope cross-reactivity was identified based on the following criteria: 1) Covering at least six consecutive amino acids. 2) Where there were amino acid differences (including those interrupting a 6aa consecutive sequence), the similarity between the amino acid structure and charge was compared. Any variances that may impact antibody function or binding, such as structural and charge changes, were excluded. 3) Expression location or alignment region found in the extracellular domain. Confirmation of B Cell Epitope Within the Self-Proteins Complete protein sequences were obtained from the UniProtKB database and mapped for potential B cell epitopes using the IEDB B cell epitope prediction tool. To correspond to the prediction algorithm used for the SARS-CoV-2 proteins, the Bepipred Linear Epitope Prediction algorithm (Haste Andersen et al., 2006;Larsen et al., 2006;Ponomarenko and Bourne, 2007) was used for human proteins where the shared alignments were to structural SARS-CoV-2 proteins. Whereas proteins which shared alignment with non-structural Orf proteins were mapped using the Bepipred Linear Epitope Prediction 2.0 algorithm (Jespersen et al., 2017). To confirm if any epitopes had been validated, the UniProtKB ID was used to search the antigens and epitopes in the IEDB database (Vita et al., 2019). Identification and Selection of B Cell Epitopes in SARS-CoV-2 Proteins As previous studies have looked at the structural proteins (Grifoni et al., 2020), B cell epitope mapping was initially performed on the four structural proteins of SARS-CoV-2 Wuhan-hu-1 variant: spike (SP), membrane (M), envelope (E) and nucleoprotein (NP), using the IEDB B cell Epitope Prediction Tool. In B cell epitopes, 4-6aa often form the minimal binding region (Pieczenik, 2003), and most B cell epitopes have a length between 4 and 25aa (Gupta et al., 2013;Potocnakova et al., 2016) (but can be as long as 50aa (Gupta et al., 2013)). As some epitopes are composed of discontinuous multiple linear segments of 1-6aa (Rubinstein et al., 2008), we selected the predicted epitopes of 6aa or more. Additionally, we applied a cut off of at least score 1, selecting those specifically with the higher prediction score and therefore more likely to be real epitopes. Based on these criteria, we identified 15 epitopes of interest ranging between 10 and 53aa ( Figure 2; Supplementary Table S3): five spike epitopes of which two are C-terminally unstructured ( Figure 3A), one unstructured C-terminal membrane epitope, and nine nucleoprotein epitopes of which two are partially structured, and three are C-terminal unstructured motifs ( Figure 3B). During the course of the present study, four of the five spike epitopes (249-261, 597-606, 805-816, 1256-1265), and three of the nine nucleoprotein epitopes (164-216, 232-269, 361-390) we predicted, were further shown by others to be regions which overlap with new epitopes identified in laboratory studies Poh et al., 2020;Shrock et al., 2020;Wang et al., 2020;Yi et al., 2020;Yoshida et al., 2021), providing support to the selected epitope mapping approach. In addition to the structural proteins, epitopes were also predicted for Orf3a, Orf3b, Orf7a and Orf8 SARS-CoV-2 proteins since antibodies against these proteins have been identified in COVID-19 patients (Hachim et al., 2020). Similar to the structural proteins, epitope sequences were selected to have a minimal length of 6aa. As the Bepipred v2.0 algorithm (Jespersen et al., 2017) was used to predict these epitopes, the prediction score criteria were modified to have a score greater than 0.55, representing the higher predicted immunogenic regions in these proteins shown to have antibody responses. Based on these criteria a total of 10 epitopes were identified within the Orf proteins ( Figure 4; Supplementary Table S3); two partially structured, one C-terminal partially structure epitope for Orf3a ( Figure 5A), three structured epitopes for Orf7a and Orf8 (Figures 5B,C, respectively) and a single unstructured N-terminal epitope within Orf3b. Of these, sequences Orf3a 172-197, Orf3a 216-225, Orf8 23-45 and Orf8 48-56 have been validated to be within a region to which COVID-19 patients have antibody responses (Shrock et al., 2020;Wang et al., 2020). Comparison of Predicted Immunogenic Regions Among SARS-CoV-2 Variants of Concern and Variants of Interest With the multiple variants of concern and of interest circulating worldwide which can escape neutralization and cause more severe disease (Garcia-Beltran et al., 2021), we were interested in whether these predicted B cell epitopes overlapped with known escape mutations or contain any mutations which may impact antibody responses. As the antibody responses to structural proteins, specifically the spike and nucleoprotein, have been Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 5 extensively studied and are used for serological assays (Post et al., 2021), potential mutations in these proteins were of interest. Using the GISAID database to obtain sequences for the VOCs: Alpha, Beta, Gamma and Delta variants, the structural proteins (spike, membrane and nucleoprotein) were blast aligned with predicted epitopes to identify any mutations in these regions (Supplementary Figures S1, S2; Supplementary Table S4). Of the five predicted spike epitopes only SP675-687 contained any mutation ( Figures 3A, 6A). In this 13aa sequence the Beta variant contains two mutations, Q677H and R682W. At position 681, both the Delta and Alpha variants contained mutations P681R and P681H respectively. In each case, these mutations consist of a structural and/or charge mutation which may impact antibody binding. For example, the P681 mutations would be expected to increase flexibility in this region. No mutation was identified within the one membrane epitope predicted ( Figure 6B). Among aa1, 58, 115, 164, 232, 273, 338, 361, 408 were identified. All associated sequences can be found in Supplementary Table S3. Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 6 the nine nucleoprotein epitopes, NP115-127, 273-287, 338-347 and 408-419 did not contain any mutations. NP1-51 contained a mutation in the Alpha variant at D3L. Epitopes NP58-85, 232-269 and 361-390 each only have one mutation: P80R (Gamma), S235F (Alpha) and D377Y (Delta). Multiple mutations were identified in NP164-216 ( Figures 3B, 6C). The Delta variant contained two mutations S202I and R203M. Position 203 was additionally mutated in the Alpha and Gamma variants, R203K. These variants were also mutated at position 204, G204R. There was additionally a mutation in the Beta variant, T205I. Except for mutations R203K and T205I in NP164-216, all mutations identified result in a structural and/or charge change that may impact antibody binding, resulting in the more severe disease seen in these variants. In addition to looking at the VOCs, we further looked into whether any mutations occurred within the predicted epitopes in the VOIs: Eta, Iota, Kappa and Lambda (Supplementary Figures S3, S4; Supplementary Table S4). Within the five predicted spike epitopes, SP249-261 and SP675-687 contained mutations ( Figure 6D). In SP249-261, the Lambda variant was missing aa249-252 and contained the mutation D253N. At this same residue the Iota variant had the mutation D253G. In each of these cases a structural and or charge change occur. Both Eta and Kappa variants contained a mutation in SP675-687, Q677H and P681R, respectively. Each of these are shared mutations in VOCs Beta and Delta, respectively. As seen in the VOCs, no mutation was identified in the single predicted membrane epitope ( Figure 6E). Within the nine predicted nucleoprotein epitopes, NP1-51, NP164-216, NP232-269 and NP351-390 contain mutations. The Eta variant contained four mutations in NP1-51: Shift at position 1, S2M, D3Y and A12G. There was also a mutation in NP1-51 in the Lambda variant, P13F. None of these are shared with VOCs, and only D3Y, given the change in structure and charge may impact antibody binding. As with the VOCs, multiple mutation sites were found within NP164-215 ( Figure 6F). Shared with VOCs were R203M (Kappa), R203K (Lambda), G204K (Lambda) and T205I (Eta). Additional mutations were P119L (Iota) and G214C (Lambda). The last two epitopes containing at least one variant with a mutation were NP232-269 and NP361-390, with one and two mutations respectively. These were M234I (Iota), T366I (Lambda) and D377Y (Kappa). Of these, D377Y results in a structural and charge change that may impact antibody binding. Shared Sequence Alignments Between SARS-CoV-2 Predicted Epitopes and Human Proteins The similarities between the predicted SARS-CoV-2 B cell epitopes and the human proteome were identified using the NCBI protein-protein BLAST tool. Each of the 25 epitopes were compared, and final lists were narrowed to remove 1) the duplicates under alternative nomenclature or sequence ID 2) different isomer repeats 3) uncharacterised or hypothetical FIGURE 3 \| Predicted epitopes mapped to protein structure of Surface Glycoprotein and Nucleoprotein. (A) X-ray crystal structure of the Surface Glycoprotein (Cai et al., 2020) coloured via spectrum (N to C terminal) with the structured B cell epitopes highlighted via spheres; aa294-261 (cyan), aa597-606 (green), aa675-687 (yellow) and aa805-816 (orange). Also labelled are the mutation sites; L249, D253, Q677, P681 and R682. (B) Structure of the nucleoprotein coloured via spectrum (N to C terminal). The N-terminal regions are from the X-ray crystal structure (PDB code: 6vyo) followed by an unstructured linear (yellow line) to a C-terminal homology model. The structural B cell epitopes are highlighted via spheres; aa58-85 (cyan), aa115-127 (green), aa164-216 (yellow), aa232-269 (orange) and aa273-287 (salmon). Also labelled is the location of known mutation sites; P80, P199, R201, S202, R203, T205, G214, N234 and S235. Table S5). Of the 25 epitopes, two (NP1-51 and NP164-216) had no significant similarity when blasted to the nr database. Among the 23 other epitopes, a total of 281 alignments consisting of 256 self-proteins, were identified. Self-Proteins Share Disease Associations Which Relate to Symptoms Reported in COVID-19 Patients For each of the human proteins identified that share similar sequences with the SARS-CoV-2 predicted epitopes ( Supplementary Table S5), the function, association to diseases (including COVID-19) and the specific sequence alignment identified were investigated ( Figure 8). In doing so, it was found that some of the alignments identified were in computationally predicted sequences or unreviewed proteins in the UniprotKB database. These proteins (Supplementary Table S6), were removed from those of interest, resulting in a total of 246 alignments, consisting of 223 self-proteins remaining (Supplementary Table S7). Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 8 reported in other SARS-CoV-2 related papers (Aishwarya et al., 2020;Vastrad et al., 2020). Ankryin B, which aligns to SP597-606, has been reported to be transcribed in lungs of COVID-19 patients, where it is not usually expressed (Aishwarya et al., 2020), and ZNF354C, aligning to NP361-390, is a transcription factor which interacts with genes that are differentially expressed in SARS-CoV-2 infected patients (Vastrad et al., 2020). Additionally, 144 of the self-proteins are reported to be associated with a range of other diseases (Supplementary Table S8). Among the identified diseases, we observed similarities for different forms of the same disease, for example types of retinitis pigmentosa or types of epilepsy. Some of these diseases share similarities with symptoms reported within COVID-19 patients such as cardiovascular diseases (atherosclerosis, cardiomyopathy, hypertension etc.) (Zhou et al., 2020b;Madjid et al., 2020); respiratory issues (airway hyper-responsiveness, inflammation) (Huang et al., 2020); neurological diseases (cerebellar ataxia, epilepsy) (Mithani et al., 2021;Povlow and Auerbach, 2021;Werner et al., 2021); and myopathy (Manzano et al., 2020;Versace et al., 2021). We also found that 42 of these 144 proteins had an association with various types of cancer. To identify whether there were any similarities or common location associations between the proteins, we grouped each of the proteins based on the body system/s they were found to be associated with (Supplementary Table S9). In doing so, we identified a range of overlap between proteins and systems ( Figure 9). 52 proteins were found to be associated with the nervous system, six of which overlapped with the cardiovascular system, which is just under half of the cardiovascular-related proteins identified. Overlap could additionally be found with the respiratory system and gastrointestinal tract (GIT), systems associated with known COVID-19 complications. Additional locations/systems found to have protein associations included excretory system, facial region, skin, bone/muscle, thyroid, mitochondria/metabolic diseases and the immune system. Among the proteins associated with these, some showed overlap with other regions. This suggests that potential interruptions in some of these proteins could have multiorgan consequences, which may be associated with COVID-19. There is an Association Between Human Proteins, With Shared SARS-CoV-2 Sequences, and Autoimmunity As reports of autoimmunity in COVID-19 continue to emerge (Bordet et al., 2020;Korem et al., 2020;Unsworth et al., 2020;Lui et al., 2021), of key interest was the association between the identified human proteins and whether they have a role in autoimmune diseases or are known autoantigens. Of the 223 human proteins, 50 were associated with autoimmune diseases, in both human and animal model settings (Supplementary Table S10). Among these 50 proteins, we found that some overlapped with multiple autoimmune diseases ( Figure 10; Supplementary Table S11). Systemic lupus erythematosus (SLE) was found to have the most protein association, followed by multiple sclerosis (combined human and animal model, experimental autoimmune encephalomyelitis (EAE)). SLE shared the most overlap with other autoimmune diseases and some targets were shared across more than two autoimmune diseases. Many of the associations identified were due to gene single nucleotide polymorphisms (SNPs) and altered expression levels. However, eight of the proteins are known targets of autoantibodies and include key antibodies for assessing or diagnosing the associated diseases such as the myasthenia gravis autoantigen A-kinase anchor protein 12 (gravin), and histone 3, a nuclear target in SLE (Table 1). This suggests that the presence of some of these autoantibodies in COVID-19 patients without a history of autoimmune disease could be due to immune cross-reactivity. with the structured B cell epitopes highlighted via spheres; aa17-25 (blue), 33-51 (green) and aa71-96 (red). (C) X-ray crystal structure of Orb7a (Flower et al., 2021) with the structured B cell epitopes highlighted via spheres; aa 23-45 (blue), 1148-56 (cyan) and aa63-78 (light green). Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 9 Potential Cross-Reactivity of Predicted SARS-CoV-2 B Cell Epitopes to Human Proteins Short identical alignments (5-6aa) have been reported to be shared with SARS-CoV-2 and human proteins (Angileri et al., 2020a;Angileri et al., 2020b;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020). Using the alignments identified from blasting predicted SARS-CoV-2 immune epitopes, the potential for immune crossreactivity was explored. While point mutations in viruses are known to create antibody escape variants (Doud et al., 2018), this is not always the case, and not all mutations will affect antibody binding (Doud et al., 2018). Some antibodies are unaffected by conserved changes (Rodpothong and Auewarakul, 2012). Therefore, to look at potential immune cross-reactivity, we applied criteria that not only relied on a minimum 6aa length, but took into account amino acid variations that had conserved charge and non-structural changes and therefore is less likely to impact antibody binding. Applying these unique criteria, a total of 136 alignments from both intra-and extra-cellular proteins were identified as potential cross-reactive targets (Supplementary Table S12). Although antibodies have a potential to cross-react with any of these proteins, they are more likely to bind and crossreact with extracellular proteins on intact cells. Using the UniProtKB database, research into the localization of each protein from Supplementary Table S12 of potential crossreactive human proteins was performed. Those of further interest were proteins that are reported to be secreted into the extracellular domain or where the alignment region was reported to be extracellular. In doing this, a total of 11 proteins were identified ( Table 2). In these 11 human proteins, the alignment identities to the SARS-CoV-2 epitopes ranged between 62 and 100%. Additionally, the SARS-CoV-2 spike and nucleoprotein epitopes with aligned human proteins were not found to have a mutation in any variant of concern or variant of interest. Next, we further explored these sequences in the human proteins through identifying whether they are potential epitopes within the self-protein sequence and their structural accessibility. Complete human protein sequences were obtained from the UniprotKB database and epitope mapping performed ( Table 3). The alignments in seven of the 11 proteins were identified as epitopes, namely Bone morphogenetic protein 1, Lysozyme-like 1, Lysozyme-like 2, CCL22, COL6A3, tubby like protein 3 and alpha-internexin. Additionally, those of COL6A3, tubby like protein 3 and alpha-internexin were found within large sequences (greater than 50aa). For the proteins Reelin, LST-3 and lactase-phlorizin hydrolase, some, but not all, of the amino acids within the alignments were identified as potential epitopes. The final protein of interest, Mucin-12 did not yield any results when using the Bepipred Linear Epitope Prediction 2.0 algorithm, the algorithm used for the matching to SARS-CoV-2 protein, Orf3a. However, using version 1 of this prediction algorithm, the alignment partially sits within a predicted epitope. Given these epitopes were predicted, we searched IEDB to explore whether any of these B cell epitopes may have been previously validated, however none were. Despite this, most of these alignments do appear as potential epitopes within the human proteins, further suggesting a likelihood that antibody cross-reactivity may occur between SARS-CoV-2 and these targets. Of the 11 extracellular proteins, five had homology model structures available which covered the alignment regions of interest: CCL22, Lysozyme-like 1, Lysozyme-like 2, Reelin and Alpha Internexin ( Table 2). Each of these alignment motifs were mapped to the corresponding structures ( Figure 11). In every case, the motifs of interest were found towards the protein surface, making them accessible to antibody binding and therefore the potential of cross-reactivity. Furthermore, for those proteins which did not have a structure readily available, all alignment motifs were predicted to reside in unstructured protein regions. These unstructured regions, usually unstructured loops or N-/C-terminal tails, are also readily available for antibody cross-reactivity. Thrombosis and Thrombocytopenia Syndrome Following COVID-19 Vaccination is Not due to Molecular Mimicry Autoantibodies targeting PF4 have been implicated in the thrombosis and thrombocytopenia syndrome (TTS) induced in rare cases following vaccination with the ChAdOx1 nCoV-19 (AstraZeneca) or the Ad26.COV2.S (Johnson & Johnson) COVID-19 vaccines (Greinacher et al., 2021;Muir et al., 2021;Scully et al., 2021). Both these vaccines consist of adenovirus vectors encoding the spike protein of SARS-CoV-2 (Folegatti et al., 2020;Sadoff et al., 2021). To explore whether molecular mimicry is a potential mechanism causing the syndrome we used the blastp protein alignment tool to align the predicted spike epitopes to PF4 (UniprotKB: P02776). No similarity was found between any predicted epitope in the spike protein to sequences in the full-length PF4 protein. To further check if there may be similarity between the spike and PF4 outside the selected epitope regions, the complete spike protein sequence was compared to PF4, which resulted in no similarity results. PF4 interacts with a variant of CXCR3 (CXCR3B) (Lasagni et al., 2003), the receptor of a number of key chemokines (such as CXCL10, a proinflammatory cytokine important for chemotaxis and the activation of peripheral immune cells (Lasagni et al., 2003)). We therefore explored whether the SARS-CoV-2 spike epitopes were similar to CXCR3B and thus potentially involved with interrupting PF4-CXCR3B interactions. Spike epitopes were blasted to the CXCR3-B UniprotKB sequence (P49682-2), and no similarities were identified. Finally, as deficiency of ADAMTS13, a metalloprotease, has been implicated in patients with ITP, spike epitopes were blasted to the ADAMTS13 UniprotKB sequence (Q76LX8), where no sequence similarity was found. This suggests that molecular mimicry is unlikely to be the cause of the described cases of TTS. DISCUSSION Using in silico immunoinformatic tools, potential B cell immunogenic epitopes in the SARS-CoV-2 proteome were predicted and further used to compare to global variants as well as explore the similarity to human proteins. In doing so, we identified eight structural epitopes containing mutations in at least one strain within these immunogenic regions. When comparing the epitopes to the human proteome, a variety of human proteins were identified to share sequences similar to SARS-CoV-2 proteins. Many of the identified human proteins were found to be associated with diseases, some of have which been reported to be related to COVID-19 symptoms and complications. Additionally, we show associations of these FIGURE 7 \| Workflow to identify human proteins that share sequence similarities with SARS-CoV-2 immunogenic regions. The predicted SARS-CoV-2 B cell epitopes were compared to the human proteome using the NCBI protein BLAST tool. Two epitopes, NP1-51 and NP164-216 had no sequence similarities to human proteins. The top 100 sequence alignments from the remaining 23 epitopes were narrowed by removing duplicates (alternative nomenclature/sequence IDs), uncharacterized/hypothetical proteins and the variable regions of lymphocyte receptors. This resulted in a final list of 281 sequence alignments that was comprised of 256 human proteins. Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 proteins to autoimmune diseases, such as SLE and MS. We further identified sequence similarities between SARS-CoV-2 immunogenic regions and human proteins which are localized in the extracellular region. These similarities and potential ease of access to circulating antibodies suggests the potential damaging cross-reactivity that can perpetuate a pathological condition. Finally, we analyzed and found that molecular mimicry may not be the mechanism for the thrombosis and thrombocytopenia syndrome occurring following vaccination with the AstraZeneca and Johnson & Johnson COVID-19 vaccines. To the best of our knowledge, this is the first study to predict B cell epitopes and compare to the highlighted VOCs and VOIs known to escape the immune response and be more infectious. However, a recent study aligned 10,664 SARS-CoV-2 genomes, to identify conserved regions and predicted both B and T cell epitopes specifically within these regions (Ghosh et al., 2021). Of their highlighted B cell epitopes, only our single predicted membrane epitope crossed over. This may be due to different epitope prediction algorithms used or single mutations in these genomes eliminating regions of interest due to not being conserved. While this method, may help in the design of epitope-based synthetic vaccines, due to targeting conserved regions, it does not indicate the immunogenic regions of the full proteins and how mutations may impact immune responses. Typically, the mutations reported in the VOCs and VOIs are those that lie in the spike protein due to its key role in infection and pathogenesis (Shang et al., 2020). D614G was one of the earliest mutations in variants that emerged as more infectious than the initial SARS-CoV-2 variant and became globally dominant (Korber et al., 2020). Each of the VOCs contain this mutation along with various others. The Alpha strain contains a N501Y mutation in the ACE2 receptor binding domain (RBD) (Garcia-Beltran et al., 2021). Examples of other mutations characteristic to the different strains used in the present study include K417N/T and E484K (Garcia-Beltran et al., 2021), and P681H, L425R, P681R, E484Q, among others (COVID-19 Weekly Epidemiol, 2021). Of these, one mutation of interest in the spike protein, P681H/R, is located next to the furin cleavage Table S7). Two proteins have been reported to be associated with COVID-19, 144 have an association with diseases (Supplementary Table S8) and 50 with autoimmunity (Supplementary Table S10). FIGURE 9 \| Overlap of proteins between body systems. Proteins found to be associated with diseases were grouped based on body system location of the diseases. Key systems with known complications in COVID-19 disease were found to have overlapping protein associations. FIGURE 10 \| Proteins found to share alignments with SARS-CoV-2 epitopes have associations with various autoimmune diseases. Proteins associated with autoimmune diseases were grouped based on the specific disease or diseases (human and animal model combined) they are found to be associated with Systemic lupus erythematosus (SLE), multiple sclerosis (MS), rheumatoid arthritis (RA) and diabetes had the most protein associations. Proteins were also found to be associated with other autoimmune diseases to a lesser extent. Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 site which is important for invading host cells (Papa et al., 2021;Peacock et al., 2021). Specifically, P681R is of interest as it is found within the Delta variant. Following the wave of devastation in India, with its increased transmissibility, the Delta variant has become the dominant variant in multiple countries (Bolze et al., 2021;Mishra et al., 2021;Sheikh et al., 2021). An early study suggests mutation P681R is associated with the increased viral pathogenicity (Saito et al., 2021). This mutation site can be identified within the predicted epitope SP675-687. This epitope is the only spike predicted epitope that contained at least one mutation in multiple VOCs (Alpha, Beta and Delta) and VOIs (Eta and Kappa) but is also the only predicted spike epitope not overlapping a validated epitope. Despite this, there may still be antibody responses to this epitope in the population, and therefore the mutations within the SP675-687, such as P681R, may be impacting antibody binding due to structure and charge changes that could be impacting a functional role of these antibodies. Among the other predicted spike epitopes only SP249-261 contained at least one mutation in VOIs, Iota and Lambda. With no mutations identified in the other three spike epitopes, this suggests that although found overlapping validated regions, these antibodies may not play a role in viral neutralization. We additionally found mutations in at least one variant in six of the nine predicted nucleoprotein epitopes. Golgin subfamily B member 1 RA, SLE and Sjögren's syndrome Rodríguez et al. (1982), Hong et al. (1992), Nozawa et al. (2004) Orf3a 238-272 Centromere protein Q SLE and systemic sclerosis Song et al. (2013) Orf3b 9-28 A-kinase anchor protein 12 (Gravin) Myasthenia gravis Gordon et al. (1992) NP164-216 was found to contain multiple mutations among all VOCs and VOIs examined, all of which were around the same region (position 199-205 and 214). It has recently been reported that high anti-NP responses may be associated with poorer outcomes in COVID-19 (Batra et al., 2021). With more severe disease associated with these variants, it raises questions of whether the mutations in these epitope regions increase antibody binding, rather than decrease binding, or whether the different strains have unique epitope regions. COVID-19 is associated with a series of multi-organ complications (Huang et al., 2020;Zaim et al., 2020). Many of the human proteins identified in this study, that share amino acid sequence similarities with the SARS-CoV-2 virus, play key roles in cellular functions, which if interrupted may result in altered cell function and therefore pathology. We found that clusters of proteins could be grouped based on their relationship to similar diseases and overlap to multiple body systems, some of which have been implicated in COVID-19 pathology, including respiratory, cardiovascular, gastrointestinal tract and nervous systems. Some of the broad examples of such diseases include epilepsy, cardiomyopathy and cerebellar ataxia, all of which have been reported in COVID-19 patients (Siripanthong et al., 2020;Mithani et al., 2021;Povlow and Auerbach, 2021;Werner et al., 2021). However, some of the diseases associated with the similar proteins may not result in a complication but instead confer a higher risk. Alzheimer's Disease, macular degeneration and cardiovascular diseases were all diseases identified with proteins that shared sequence similarities to SARS-CoV-2 capable of making them the targets of autoantibodies. Preexisting diagnosis for each of these have been found to predict higher risk of infection and greater severity and risks in COVID-19 Huang et al., 2020;Ramlall et al., 2020;Yu et al., 2021). Many of the proteins identified to be associated with disease are intracellular and are therefore less likely to be immune targets. However, as the SARS-CoV-2 virus is an intracellular pathogen, the sequence similarities could alternatively have an impact on cellular functions which may result in the observed pathologies, independently of having the potential to be recognized by antibodies. Autoantibody (AAb) targeted proteins found within a range of autoimmune diseases (SLE, Myasthenia Gravis, T1D etc.) were found to share similar sequences to some of the new predicted SARS-CoV-2 epitopes, as well as known SARS-CoV-2 epitopes (Shrock et al., 2020;Wang et al., 2020). Studies have shown that within COVID-19 patients, known AAbs associated with autoimmune diseases, including but not limited to, anticardiolipin, anti-SSA/Ro and anti-nuclear antibody (Zhou et al., 2020a;Vlachoyiannopoulos et al., 2020) are increased, indicating a breaking of immune tolerance (the mechanisms which regulate responses and ensure immune cells do not attack self). As identified in the present study, histone H3 shares an identical 6aa sequence with the SARS-CoV-2 Orf8 protein, which additionally sits in a region identified as an epitope in COVID-19 patients , indicating the possibility of immune cross-reactivity. Gravin, the myasthenia gravis autoantigen, was also found to have sequence similarity with the Orf3b viral protein. While, to our knowledge, anti-gravin AAbs have not been reported in COVID-19 patients, there has been a case report of post-COVID-19 infection onset of myasthenia gravis (Huber et al., 2020). With several viral infections being associated with autoimmune diseases, these data support the possibility that the COVID-19 pandemic will lead to an increase in autoimmune diseases. Anti-La and anti-Jo-1, typically associated with SLE/Sjogren's Disease and inflammatory myopathies respectively, have been identified in children with COVID-19 Multisystem Inflammatory Syndrome (MIS-C) (Consiglio et al., 2020;Gruber et al., 2020). However, it is not only AAbs with known associations to autoimmune diseases found in COVID-19 patients (Consiglio et al., 2020;Gruber et al., 2020). AAbs to a range of tissue specific and immune related mediators have been identified in MIS-C patients. Additionally, increased levels of anti-interferon (IFN) antibodies have been reported to be associated with more severe COVID-19 disease (Bastard et al., 2020). This shows that it is not only important to understand the targets associated with autoimmune diseases, but that any similarity or recognition of self may contribute to additional COVID-19 pathology and/or greater disease severity. The majority of proteins found to fit within our distinct criteria for the potential of cross-reactivity were intracellular, with only 11 found to be extracellular. By applying these criteria, two proteins associated as antigens in autoimmune diseases, Tubby-related protein 3 and Alpha-internexin, were also identified as potential cross-reactive proteins. We additionally found that tubby-related protein 3, as well as two other potential cross-reactive targets, bone morphogenetic protein 1 and Mucin-12, have alignments similar to SARS-CoV-2 epitopes which overlap with in vivo validated epitopes Shrock et al., 2020;Wang et al., 2020), potentially indicating an opportunity for crossreactivity, especially as these targets are extracellular. Additionally, as seen in autoimmune diseases, many autoantibodies to various intracellular targets are identified (Suurmond and Diamond, 2015). During COVID-19 infection, the release of intracellular proteins may be playing a role in breaking immune tolerance, allowing for the potential cross-reactivity of SARS-CoV-2 antibodies to self, or the increase of AAbs. This may be perpetuated through the recognition of the regions that are extracellular and therefore more likely to be visible to circulating antibodies. Interestingly, the SARS-CoV-2 spike and nucleoprotein alignments with human protein matches conforming to our criteria, are ones that do not contain mutations across different global variants, suggesting the potential for cross-reactivity to these proteins irrespective of the virus variant. Furthermore, according to the IEDB, there are several confirmed discontinuous B cell epitopes within the spike protein, which may also further expand the number of cross-reactive epitope targets to human proteins to be explored in future studies. Other studies have reported sequence similarities between SARS-CoV-2 and human proteins (Angileri et al., 2020a;Angileri et al., 2020b;Ehrenfeld et al., 2020;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Lyons-Weiler, 2020;Marino Gammazza et al., 2020), the majority through identifying identical 5 or 6 amino acid segments found within immunogenic regions (Angileri et al., 2020a;Angileri et al., 2020b;Ehrenfeld et al., 2020;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020). Apart from the study of Kanduc (2020), who formed overlapping hexamers (offset by one residue) from SARS-CoV-2 sequences that are identical in validated SARS-CoV-1 epitopes, these studies began by aligning overlapping pentamers/hexamers to human proteins to identify identical sequences (Angileri et al., 2020a;Angileri et al., 2020b;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020). These identical alignments were then determined if they were found in immunogenic regions. In contrast, Lyons-Weiler (2020) predicted immunogenic SARS-CoV-2 epitopes which were then compared to human proteins. In the present study, we used a similar approach to Lyons-Weiler (2020) by identifying sequences in key SARS-CoV-2 proteins with higher predicted immunogenic regions. However, the comparison between these SARS-CoV-2 sequences and human proteins was further expanded, in contrast to the other reports (Angileri et al., 2020a;Angileri et al., 2020b;Ehrenfeld et al., 2020;Kanduc, 2020;Kanduc and Shoenfeld, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020), as we did not only rely on short identical sequences. Instead, to explore potential immune cross-reactivity, we applied a novel combination of criteria, which considered the ability of antibodies to potentially recognize amino acid differences, as well as the localization of the alignments for antibody accessibility. Although our criteria were unique compared to the previous studies, we'd expect some crossover with those who report identical hexamers (Angileri et al., 2020b;Ehrenfeld et al., 2020;Kanduc, 2020;Lucchese and Flöel, 2020;Marino Gammazza et al., 2020), and had a similar start methodology (Lyons-Weiler, 2020). However, to the best of our knowledge only five previously reported proteins were also found in our potential targets: Hermansky-Pudlak Syndrome I protein, Unconventional myosin XVI, Transmembrane Protein KIAA1109 (Ehrenfeld et al., 2020), Ankyrin repeat and sterile alpha motif domain containing 1A (Lyons-Weiler, 2020) and CLOCK (Kanduc, 2020). Each of these proteins reported the same alignment to the same SARS-CoV-2 protein, except for CLOCK which was reported with a different hexamer alignment suggesting some proteins may have multiple shared sites with SARS-CoV-2. Other similar findings may not have been identified as these reports focus only on human proteins of a certain function or location (e.g. molecular chaperones or adaptive immune system proteins), some of which were found to match SARS-CoV-2 proteins not studied in the present study (e.g. orf1ab). Differences in findings may also be due to different in silico tools used for the comparison of SARS-CoV-2 and human proteins, where we used the NCBI blastp tool while others used the Pir Peptide Match program (Kanduc, 2020;Kanduc and Shoenfeld, 2020;Marino Gammazza et al., 2020). It was found that the Pir Peptide Match program preferred short sequences (∼6aa length) to identify the similar human proteins but could not obtain the same results when using the full-length epitopes (>10aa long). Whereas the NCBI blastp tool allowed us to apply criteria that not only relied on identical sequences but considered how similar charge and structure between amino acid variations may not impact antibody binding and therefore explore sequence similarity between the full length of the predicted epitopes and human proteins. Additionally, not all pentamers or hexamers from the SARS-CoV-2 proteins, reported to align with human proteins, can be found within our predicted epitopes, further explaining differences in the human proteins identified. In some cases, this may be due to different prediction algorithms used, or the sequences were part of prediction epitopes that did not fit our specific epitope criteria. One sequence, SRSSSR, found within the nucleoprotein, has been highlighted in multiple reports as having a shared alignment with human proteins (Angileri et al., 2020b;Kanduc, 2020). This hexamer can be found within our predicted epitope NP164-216, however this was one of the epitopes that obtained no significant blast results. This example demonstrates how there can be major differences in outputs between different bioinformatic methods. Computational methods, such as epitope mapping and blasting are useful techniques. They can allow narrowing of questions and potential proteins of interest in hypotheses before doing experimental studies. However, while bioinformatics studies can narrow these lists, the current study is limited to predictions alone and ultimately need to be confirmed experimentally. Since the initial epitope predictions were performed, new literature emerged to identify new SARS-CoV-2 B cell epitopes in vitro that were also identified by our approach which have not been reported previously Poh et al., 2020;Shrock et al., 2020;Wang et al., 2020;Yi et al., 2020;Yoshida et al., 2021). Several of our predicted epitopes, including: Spike 249-261, 597-606, 805-816, 1256-1265NP 164-216, 232-269, 361-390;Orf3a 172-197, 216-225, Orf8 23-45, 48-56 overlap with the new epitopes highlighted in these reports. Regions within spike 553-579 and 806-835 have been identified across several studies as immunodominant regions Poh et al., 2020;Yi et al., 2020), whereas other epitopes show differing ranges of reactivity (Shrock et al., 2020). This indicates that, although not immunodominant or reported in the literature, these predicted epitopes may still be real, especially as individual immune responses are polymorphic. This is important to consider when mapping potential crossreactivities between SARS-CoV-2 and self-antigens, as there Frontiers in Bioinformatics \| www.frontiersin.org August 2021 \| Volume 1 \| Article 709533 may still be protein cross-reactivities to identify due to host or virus polymorphisms. Additionally, to allow for direct sequence similarities between virus epitopes and human proteins, only linear B cell epitopes were explored in the present study. However, B cell epitopes can be composed of multiple discontinuous segments (Rubinstein et al., 2008). As more protein structures covering the full-length SARS-CoV-2 proteins are identified, performing discontinuous epitope analysis may further uncover other potential human proteins that may become targets of antibodies through cross-reactivity. Furthermore, T cells may also cross-react with self-protein T cell epitopes through mechanisms such as bystander activation (Smatti et al., 2019). A future study looking specifically at T cell epitopes and crossreactivity to self may provide further evidence of the immune system's role in pathology. For the first time, to our knowledge, we further explored and identified that some of the alignments in self-proteins predicted to be cross-reactive with SARS-CoV-2 alignments are potential epitopes within their own complete sequence. Additionally, where possible, we applied structural analysis to further explore the potential for cross-reactivity to the human proteins. In doing so, we identified the motif alignments were towards the surface of the protein, or in unstructured regions, and therefore would provide ready access for antibody binding. As crystal structures become available for the human proteins of interest, both full coverage of those mapped, as well as the remaining proteins, it would be interesting to further apply this method. However, the Bepipred Linear Prediction Algorithm 2.0 derives its epitope predictions from crystal structures (Jespersen et al., 2017). It is therefore likely that the Orf proteins and the human cross-reactive alignments predicted to be epitopes, are more likely to be found on the surface where antibodies have easier access. The mechanism behind the thrombotic events occurring in some people following vaccination with the AstraZeneca and Johnson & Johnson adenovirus vectored vaccines is not known (Scully et al., 2021). Our findings suggest that molecular mimicry between the SARS-CoV-2 spike protein and proteins implicated in TTS and ITP, including PF4 and ADAMTS13, is unlikely to be the cause of these events. The vaccine induced thrombosis thrombocytopenia syndrome has been reported to be similar to autoimmune heparin-induced thrombocytopenia (aHIT) (Schultz et al., 2021), and therefore similar mechanisms may be involved. In aHIT, it has been suggested that structural changes in PF4 may be involved (Greinacher et al., 2017). Anti-PF4 antibodies are part of the diagnostic criteria for TTS (Greinacher et al., 2021). It has been hypothesized that free DNA in the vaccines may be a possible trigger of the anti-PF4 antibodies (Greinacher et al., 2021). However, further research into the formation of these anti-PF4 autoantibodies and the causes behind TTS following COVID-19 vaccination is required to identify potential interventions to prevent these events. In the present study, linear B cell epitopes in SARS-CoV-2 proteins were predicted and compared to human proteins, identifying a number of new targets for potentially crossreactive autoantibodies. Since many of the predicted epitopes reported in this study overlap with ones validated in COVID-19 patients, future studies may synthesize peptide sequences from the SARS-CoV-2 epitopes and the equivalent human sequences to perform in vitro analysis for validation. 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.	2021-08-19T13:11:50.830Z	2021-08-19T00:00:00.000	{ "year": 2021, "sha1": "2b23819822a5528dae5478b5294318e850e3b8f7", "oa_license": "CCBY", "oa_url": "https://www.frontiersin.org/articles/10.3389/fbinf.2021.709533/pdf", "oa_status": "GOLD", "pdf_src": "Frontier", "pdf_hash": "2b23819822a5528dae5478b5294318e850e3b8f7", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Computer Science", "Medicine" ] }
259672520	pes2o/s2orc	v3-fos-license	High-Throughput Determination of Stern–Volmer Quenching Constants for Common Photocatalysts and Quenchers Mechanistic information on reactions proceeding via photoredox catalysis has enabled rational optimizations of existing reactions and revealed new synthetic pathways. One essential step in any photoredox reaction is catalyst quenching via photoinduced electron transfer or energy transfer with either a substrate, additive, or cocatalyst. Identification of the correct quencher using Stern–Volmer studies is a necessary step for mechanistic understanding; however, such studies are often cumbersome, low throughput and require specialized luminescence instruments. This report describes a high-throughput method to rapidly acquire a series of Stern–Volmer constants, employing readily available fluorescence plate readers and 96-well plates. By leveraging multichannel pipettors or liquid dispensing robots in combination with fast plate readers, the sampling frequency for quenching studies can be improved by several orders of magnitude. This new high-throughput method enabled the rapid collection of 220 quenching constants for a library of 20 common photocatalysts with 11 common quenchers. The extensive Stern–Volmer constant table generated greatly facilitates the systematic comparison between quenchers and can provide guidance to the synthetic community interested in designing and understanding catalytic photoredox reactions. ■ INTRODUCTION −11 The mechanism for PET occurs in one of two ways: (i) reductive quenching, whereby the photocatalyst is quenched through acceptance of an electron from a reductant or (ii) oxidative quenching, whereby the photocatalyst donates an electron to an oxidant (Figure 1A). 7The quenching agent may be an integrated component of the transformation, such as one of the organic substrates or a cocatalyst or a sacrificial quencher that serves to convert the photocatalyst to a species that can oxidize or reduce the organic substrate, such as O 2 or alkyl amines. 11,12hedding light on quenching steps within a photoredox reaction is crucial to gain mechanistic insights; 13−17 this can include questions such as establishing which reaction component is acting as the main quencher and the magnitude of quenching that occurs under catalytically relevant conditions.−24 Emission quenching (also called Stern−Volmer) studies consist of measuring emission intensity fluctuations of the photocatalyst excited state in the presence of a quencher.In a typical Stern−Volmer study, a fluorometer is used to measure the emission intensity, and solutions of the photocatalyst with different concentrations of quenchers need to be prepared individually under a rigorously inert atmosphere to avoid the interference of oxygen quenching. 25Such constraints have limited the prevalence of Stern−Volmer studies compared to the extensive body of work on photoredox catalysis.A fully automated continuous flow platform has been developed to facilitate data collection and reduce manual labor but remains low throughput and requires the construction of a homemade flow setup. 26−29 To keep up with the exponentially increasing chemical space for photocatalytic transformations, new workflows that enable Stern−Volmer data collection at a higher throughput need to be developed. To that end, we have devised a protocol for the collection of Stern−Volmer quenching data using commercially available fluorescence plate readers and consumable 96-well plates.Plate readers are ubiquitous in biological studies and have been used in a variety of colorimetric and fluorometric assays, 30−32 as they allow for rapid (<1 min) reading of an entire 96-well plate.In our quenching studies, the use of well plates enables the collection of Stern−Volmer constants for one photocatalyst and up to 12 quenchers in less than a minute.We further expanded the throughput of this workflow by devising a protocol for a liquid handling robot to enable the automated dispensing of quenchers at increasing concentrations in a 96well plate format.−9 The 220 quenching rate table generated opens the door for deeper data analysis of quenching trends and mechanisms, and we believe that it will be useful to the community interested in mechanistic understanding of photoredox catalysis. ■ RESULTS AND DISCUSSION In photoredox reactions, the first step in the photochemical cycle is excitation of the photocatalyst by a photon of the appropriate wavelength.Upon excitation, the photocatalyst can decay back to the ground state in a radiative (via photon emission, i.e., fluorescence or phosphorescence) or nonradiative way (heat) (Figure 1A). 33In the presence of an increased concentration of a quencher, the photocatalyst excited state can also react via EnT or PET, which will compete with radiative decays and ultimately decrease photon emission (Figure 1B).The decrease in emission intensity is linked to the concentration of a quencher as described by the Stern−Volmer equation (eq 1) (1) where I 0 is the emission intensity in the absence of a quencher, I is the intensity in the presence of a quencher, k q is the quenching constant, τ 0 is the excited state lifetime, and [Q] is the quencher concentration. 34The product of k q and τ 0 is often described as K sv , the Stern−Volmer constant.K sv can be used to compare quenching rates between different quenchers (Figure 1C).The knowledge of K sv and [Q] for all of the components of a reaction enables the ranking of quenching power and ultimate identification of the main quencher for the photocatalyst. In a typical Stern−Volmer experiment, independent solutions of different concentrations of quenchers and a constant concentration of photocatalysts need to be prepared under a rigorously inert atmosphere, and the excitation intensity of each one must be measured separately to extract I and build the Stern−Volmer plot.We decided to take advantage of the parallelization offered by 96-well plates to propose a Stern−Volmer plate design, which involves increasing the quencher concentration moving down a column, with a different quencher (or replicate) in each column (Figure 2).Multichannel pipettors enabled the rapid dispensing of each component across the entire plate.To maintain a strict inert atmosphere around the samples, the entire sample preparation and measurement were performed under an inert atmosphere inside a nitrogen-filled glovebox in constant purge mode.This was made possible by the small footprint of the fluorescence plate reader, which can easily be introduced into a glovebox.To validate this design, we sought to test our approach by comparing K sv values obtained with our workflow to values reported in the literature.We probed the quenching of (2,2′-bipyridine)bis[2-(pyridinyl-κN)phenyl-κC]iridium-(III) (Ir(ppy) 3 ) by diethylbromomalonate, bromoacetonitrile, and indole in dichloroethane (DCE) and observed similar trends in the quenching values, with diethylbromomalonate, exhibiting a higher K sv than bromoacetonitrile by a factor of 1.7 (in the original report, the factor is 2), and indole, showing virtually no quenching (Figure 3). 35wever, we measured higher quenching constants overall by a factor of ∼2. 35This difference is likely because of our ability to consistently maintain the sample in an inert atmosphere throughout the entire measurement using this workflow. With this positive validation result in hand, we sought to showcase the magnitude of quenching data accessible through this high-throughput method.We decided to focus on generating Stern−Volmer constants for a library of commonly used photocatalysts and quenchers in a rapid fashion.We automated the plate preparation by writing a protocol to direct a liquid handling robot (Tecan Freedom EVO 100) to dispense the solutions of photocatalyst and quencher into 96well plates.First, one photocatalyst was dispensed across the entire plate.We maintained the quencher concentration as the variable across rows and selected 12 widely used quenchers for the columns.With this design, one photocatalyst could be analyzed on a plate in less than one minute.The automated liquid dispensing allowed us to prepare plates for each of 20 common photocatalysts in <2 h entirely under an inert atmosphere (Figure 4) for a total of 1920 samples to generate 220 quenching constants. The 12 quenchers were N,N-diisopropylethylamine, triethylamine, 4-methoxystyrene, Hantzsch ester, quinuclidine, bis-(pinacolato)diboron, phenylacetic acid, tris(trimethylsylil)silanol, and potassium cyclohexyltrifluoroborate as reductive quenchers and di-tert-butyl peroxide and diethylbromomalonate as oxidative quenchers (sodium persulfate was included as well but exhibited low solubility in MeCN such that no quenching was measured and reported).These quenchers were selected for their wide use in many different photoredox reactions. 2,3,9,11Trialkylamines are typically used as a sacrificial reductant to enable access to a highly reducing photocatalyst and, as such, are used in a wide range of reactions.The other reagents contain functional groups that will typically generate reactive radicals upon photocatalyst quenching and that will undergo the desired reactivity.The photocatalysts were selected to contain diverse chemical structures, such as cyanoarene-based photocatalysts, acridinium salts, ruthenium-(II) complexes, iridium(III) complexes, and other miscellaneous organophotocatalysts.This selection varies not only in structures and photophysical properties but also in redox potentials while encompassing the most commonly used catalysts for photoredox reactions (Figure 5).The K sv values for different photocatalysts are reported in Table 1.For low K sv values, the entry is shown as <5.In such cases, higher concentrations of quenchers will be needed to determine K sv . Interestingly, we observed entries exhibiting negative deviations from linearity in their quenching profiles, mostly with Hantzsch ester (Figure 6) and diethylbromomalonate.This type of behavior has been attributed to the presence of different conformers for the photocatalyst, which gives rise to fractional accessibility of the fluorophore. 36Some cases can be treated using a modified Stern−Volmer equation, namely, the Lehrer equation. 37It should be noted that the Hantzsch ester, unlike the other quenchers, also has a non-negligeable absorbance up to 400 nm and emits at 450 nm, 38 which overlaps with the absorbance and emission of a few of the selected photocatalysts, thereby creating an inner filter effect, which interferes with the Stern−Volmer measurement.Therefore, data for this particular quencher might not reflect the true photocatalyst quenching rate. Overall, we observed strong quenching from trialkylamines with typically oxidative excited photocatalysts (Figure 7A), which is unsurprising considering their wide use as quenchers in many photoredox reactions.DIPEA is the substrate that exhibits quenching with most photocatalysts in the collection.Conversely, oxidative quenchers like diethylbromomalonate tend to only react with strongly reducing photocatalysts such as Ir(ppy) 3 .However, further data analysis of K sv values of the quenchers in relation to the oxidation potential of the photocatalyst excited state shows that, interestingly, these two entities are not necessarily related.For example, quenching values for triethylamine show no clear correlation between K sv and excited state oxidation potential (Figure 7B).Other phenomena such as pH, binding events, exciplex formation, or hydrogen bonding have been shown to influence quenching. 17,23Solvent effects also have a role, as illustrated by the 1.7-fold higher K sv between Ir(ppy) 3 and diethylbromomalonitrile in acetonitrile compared to that in DCE.Additionally, phenylacetic acid showed very little quenching with all photocatalysts, but the quenching rate for this specific quencher is likely dependent on pH with the deprotonated carboxylate being more susceptible to oxidation. 4,39An interesting quencher is (TMS) 3 SiOH, which has been used recently in several metallophotoredox reactions 1,40 and which shows noticeable quenching patterns.Indeed, while most reported systems including (TMS) 3 SiOH involve Ir photocatalysts, we can observe much higher quenching with acridiniums and organocatalysts, such as N-Et-Flavin and DCA.These photocatalysts could therefore outperform Ir in systems relying on this quencher.Interestingly, two photocatalysts, namely, tetraMeO-Acr-N-Ph and tetraMeO-Acr-N-diMeOPh, show some degree of quenching with all of the quenchers tested (Figure 7C), which might explain their increased use as versatile photocatalysts. 9On the contrary, photocatalysts, such as Ru(phen) 3 Cl 2 and Ru-(bpy) 3 (PF6) 2 , one of the most important photocatalysts that launched the entire field of photoredox catalysis, 11 exhibit fairly poor quenching with most substrates.This observation could partially explain its decline in usage against more reactive Ir photocatalysts, although it should be noted that many other factors beyond photocatalyst quenching impact photocatalytic processes and can determine the failure or success of a photocatalyst. In conclusion, we demonstrated the use of a highthroughput workflow for the measurements of Stern−Volmer quenching constants using commercially available fluorescence plate readers and 96-well plates.We validated the method by comparing the results to reported quenching values.Automatization of the liquid dispensing step allowed for the rapid preparation and analysis of the Stern−Volmer quenching for a collection of common photocatalysts and quenchers.We believe that this workflow enables a consistent and rapid collection of quenching constants with readily available instrumentation, thereby finally bringing Stern−Volmer studies to the same level of high-throughput capabilities as those for reaction discovery for photoredox catalysis.We believe that this method as well as the data generated is truly enabling for the community interested in mechanistic understanding of photoredox reactions. Chemicals and Materials Unless stated otherwise, all of the reagents and solvents were purchased from commercial suppliers (Acros, Merck Millipore Sigma, TCI America, Fisher Scientific, etc.) and were of the highest analytical purity.UPLC grade solvents were used for all experiments.All operations were performed inside a N 2 -filled glovebox.and fluorescence spectrum measured with the plate reader.The emission data was then exported to Excel and fit to the Stern−Volmer equation using least squares regression with y-intercept set to 1.The slope values (i.e., quenching constants) were filtered by magnitude and by R 2 value to constrain the data to the observation of significant linear quenching.Plots with noticeable but nonlinear quenching were tested for outliers (two standard deviations away from the line of the best fit) and/or improved fit with the Lehrer equation. Data Availability Statement The data underlying this study are available in the published article and its Supporting Information. The supporting information contains the Tecan script for the automated dispensing and the raw emission data (PDF) ■ Figure 5 . Figure5.Quenchers and photocatalysts for the quenching studies.λ max is the absorption maximum for the photocatalyst and below are reported the potential for the redox couples of the excited photocatalyst. a (a) Nonlinear quenching, (b) absorbance overlap between the quencher and photocatalyst, and (c) fit with the Lehrer equation. Figure 7 . Figure 7. (A) Comparison of the average K sv over all of the photocatalysts for each quencher.(B) Plot of K sv against the photocatalyst excited state potential for triethylamine.(C) Number of quenchers with >5 K sv for each photocatalyst.Color coding: trialkylamines (purple circle solid), other reductive quenchers (teal circle solid), oxidative quenchers (yellow circle solid), cyanoarenes (blue circle solid), acridiniums (purple circle solid), Ru(II) complexes (maroon circle solid), Ir(III) complexes (orange circle solid), and miscellaneous organic photocatalysts (light yellow circle solid). Table 1 . K sv between Catalyst and Quencher Collection Reported in M −1a	2023-07-12T07:05:58.637Z	2023-06-29T00:00:00.000	{ "year": 2023, "sha1": "e59b7e3ca8b16d627bf94a4647f5440906fbdd80", "oa_license": "CCBYNCND", "oa_url": "https://pubs.acs.org/doi/pdf/10.1021/acsorginorgau.3c00019", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "4dcabf38280ff32132d78b44a600323b73681513", "s2fieldsofstudy": [ "Chemistry" ], "extfieldsofstudy": [] }
222317392	pes2o/s2orc	v3-fos-license	Simvastatin improves mitochondrial respiration in peripheral blood cells Statins are prescribed to treat hypercholesterolemia and to reduce the risk of cardiovascular disease. However, statin users frequently report myalgia, which can discourage physical activity or cause patients to discontinue statin use, negating the potential benefit of the treatment. Although a proposed mechanism responsible for Statin-Associated Myopathy (SAM) suggests a correlation with impairment of mitochondrial function, the relationship is still poorly understood. Here, we provide evidence that long-term treatment of hypercholesterolemic patients with Simvastatin at a therapeutic dose significantly display increased mitochondrial respiration in peripheral blood mononuclear cells (PBMCs), and platelets compared to untreated controls. Furthermore, the amount of superoxide is higher in mitochondria in PBMCs, and platelets from Simvastatin-treated patients than in untreated controls, and the abundance of mitochondrial superoxide, but not mitochondrial respiration trends with patient-reported myalgia. Ubiquinone (also known as coenzyme Q10) has been suggested as a potential treatment for SAM; however, an 8-week course of oral ubiquinone had no impact on mitochondrial functions or the abundance of superoxide in mitochondria from PBMCs, and platelets. These results demonstrate that long-term treatment with Simvastatin increases respiration and the production of superoxide in mitochondria of PBMCs and platelets. www.nature.com/scientificreports/ biosynthesis of mevalonate, farnesyl pyrophosphate, and ubiquinone (UQ). UQ, also known as coenzyme Q10, mediates the transport of electrons from complex I, and II to complex III in the electron transport chain (ETC) 24 . One hypothesis suggests that lower levels of UQ found in plasma, peripheral blood mononuclear cells (PBMCs), and skeletal muscle of statin users result in an impairment of the ETC, thereby reducing the rate of mitochondrial respiration 17,[25][26][27] . Here, we provide evidence that is not consistent with this hypothesis. On the contrary, we demonstrate that long-term Simvastatin therapy has the opposite effect, significantly improving mitochondrial respiration in patient-derived platelets, and PBMCs. We argue that the hypothesis of a negative relationship between statin usage, and mitochondrial function is based on suboptimal experimental setups, and we demonstrate that shortterm Simvastatin therapy at a dose vastly exceeding the pharmacologically relevant dose does in fact impair mitochondrial respiration in human hepatocarcinoma cells cultured in vitro. Therefore, the effect of statins on mitochondrial respiration appears to be context-, and dose-dependent. In this study, mitochondrial parameters were evaluated in patient-derived platelets, and PBMCs, with the aim to investigate statin-induced mitochondrial alterations. We used blood cells in this study because it is less invasive to access compared to muscle cells obtained from muscle biopsies. Moreover PMBCs and platelets are well-characterized surrogate models for assessing pharmacological effects on skeletal muscle mitochondrial function [28][29][30][31][32][33] . In addition to increasing mitochondrial respiration, we observed that Simvastatin increases production of mitochondrial superoxide, a representative reactive oxygen species (ROS) by-product of mitochondrial respiration, and that patient-reported myalgia increases with increasing abundance of superoxide, but does not trend with oxygen consumption rate (OCR) in mitochondria from patient-derived peripheral blood cells. Short term in vitro exposure to high concentration Simvastatin decreases mitochondrial respiration in Huh-7 cells. To investigate the effects of high dose Simvastatin treatment, the human hepatocyte-derived Huh-7 cell line was selected, as the effects of Simvastatin treatment are well described for this cell line 34,35 . In the present study, Huh-7 cells were exposed to 2.5-10 µM Simvastatin for 72 h in vitro, as this dose is widely used in the literature describing pleiotropic effects of Simvastatin treatment 36 . After 72 h of Simvastatin treatment, the effect on mitochondrial respiration was determined by evaluating basal respiratory rate, ATP turnover, reserve respiratory capacity, and maximal respiratory capacity (Fig. 1A,B,F). Basal respiratory rate is a measure of the rate of oxygen consumed by the cells when kept in media lacking inhibitors added (Fig. 1C). ATP turnover is measured as the decrease in rate of oxygen consumption after inhibition of the ATP synthase (Fig. 1D). This decrease is relative to ATP produced by oxidative phosphorylation. The reserve respiratory capacity is a measure of a theoretical extra capacity to produce ATP as a response to an increased energetic demand (Fig. 1E), and this has previously been correlated with the ability of cells to withstand periods of stress 37 . Reserve respiratory capacity is measured as the difference in OCR at basal, and that at maximal activity. Maximal respiratory capacity is a measure of total oxygen consumption possible by oxidative phosphorylation and it is measured as the difference in OCR following uncoupling by FCCP and after ETC inhibition by antimycin A (Fig. 1F). Our results show that basal OCR in Huh-7 cells decreased significantly with increasing concentrations of Simvastatin, being 30% (P = 0.0126) or 75% (P < 0.0001) lower than control cells in the presence of 2.5 µM or 10 µM Simvastatin, respectively. ATP turnover (P = 0.0011), reserve respiratory capacity (P = 0.0005), and maximal respiratory capacity (P = 0.0002) were significantly decreased compared to controls in the presence of 10 µM Simvastatin but showed no effect at lower concentrations of Simvastatin. These results demonstrate that short term in vitro exposure to high-dose Simvastatin significantly impairs mitochondrial respiration in Huh-7 cells. Short term in vitro exposure to high dose Simvastatin increases production of mitochondrial superoxide production in Huh-7 cells. Mitochondrial superoxide was 1.35-1.50 fold higher in Huh-7 cells exposed to 2.5-10 µM Simvastatin than in untreated controls ( Fig. 1G) (P = 0.01; P < 0.0021, and P = 0.0007 for 2.5, 5, and 10 µM Simvastatin, respectively). In cells exposed to 10 µM Simvastatin, superoxide production increased to 30% of the level induced by exposure to menadione, a level that could cause significant oxidative damage to nucleic acids, and other cellular macromolecules. These results indicate that in vitro exposure to high dose Simvastatin has the potential to cause significant respiratory and oxidative alterations. Establishment of the cohort. The effect of Simvastatin on mitochondrial respiration in human subjects was investigated in a cohort of 40 patients with at least 6 months history of Simvastatin use (20-40 mg per day), and 12 matched age-, weight-, BMI-, and body fat matched control subjects ( Table 1). The concentration of white blood cells (WBC), neutrophils and monocytes were slightly, but significantly increased in Simvastatin users compared to controls (P = 0.034; P = 0.047, and P = 0.027 respectively) ( Table 1). There was no difference in VO 2 max levels between Simvastatin users and controls, indicating similar levels of fitness (data not shown). Simvastatin usage is correlated with increased mitochondrial respiration. Mitochondrial parameters were measured in PBMCs, and platelets from the cohort of Simvastatin users and controls. The results show a 63% decrease in basal OCR in platelets, but not in PBMCs, from Simvastatin users (P = 0.0003) ( Fig. 2A,G). Reserve respiratory capacity was 1.37 fold higher in platelets, and 1.43 fold higher in PBMCs, respectively (P = 0.0038 and P = 0.0027) (Fig. 2C,I) in Simvastatin users than in controls. This indicates an improved ability to accommodate energy demand. No difference in ATP turnover rate was detected in platelets or PMBCs (Fig. 2B,H), demonstrating equivalent levels of oxygen used for ATP production. A significant increase in maximal OCR was observed in platelets (1.33 fold, P = 0.0071), and PMBCs (1.30 fold, P = 0.0004) (Fig. 2D,J) www.nature.com/scientificreports/ similar in platelets, and PBMCs from Simvastatin users, and controls ( Fig. 2E,K), while glycolytic reserve was 2.82 fold, and 1.53 fold higher in platelets and PBMCs, respectively (P = 0.0007 and P = 0.0015) (Fig. 2F,L). Glycolytic reserve is the difference in ECAR in the presence and absence of an ATP synthase inhibitor. The glycolytic reserve is, therefore, a measure of the glycolytic capability of a cell to respond to an inhibition of oxidative phosphorylation. Assuming that the glycolytic capacity is unaltered, a higher glycolytic reserve indicates a higher dependence on oxidative phosphorylation in platelets and PBMCs from Simvastatin users than controls. Simvastatin users were stratified into users with (N = 14), and without (N = 17) self-reported myalgia. We found no significant association between myalgia, and any measured parameter of mitochondrial respiration or glycolysis ( Figure S1). Complex I and complex IV activity is higher in Simvastatin users. To describe the functional background for the measured mitochondrial attributes, a dipstick assay was used to quantify the activity of ETC complexes I and IV in platelets. This assay detected 1.70 fold, and 1.94 fold more active ETC complex I and complex IV, respectively, in Simvastatin users than in controls (P < 0.0001) (Fig. 2M,N). These data are consistent with above results, showing higher OCR, and reserve respiratory capacity in Simvastatin users than in controls ( Fig. 2I,J). Oxygen consumption was quantified in Huh-7 cells in the presence of 0, 2.5, 5 or 10 µM Simvastatin, as indicated. OCR was measured before and after addition of the ATP synthase inhibitor oligomycin and complex III inhibitor, antimycin A (A), or before and after addition of the uncoupler FCCP and antimycin A (B); the data were used to calculate basal respiratory rate (C), ATP turnover (D), reserve respiratory capacity (E) and maximal respiratory capacity (F). Mitochondrial superoxide was quantified in Huh-7 cells treated with 0, 2.5, 5 or 10 µM Simvastatin; values were normalized to control (0 µM Simvastatin) and menadione was used as a positive control (G). Panels A-F show mean OCR ± standard deviation. Panel G shows the mean superoxide level ± standard deviation. At least three biological replicates were performed. www.nature.com/scientificreports/ Mitochondrial DNA content is similar in PBMCs from Simvastatin users and controls. The amounts of mitochondrial and nuclear DNA (mtDNA, nDNA) were quantified by real-time quantitative PCR (qPCR) in PBMCs from 20 Simvastatin users, and eight controls ( Figure S2) and the mtDNA:nDNA ratio was calculated. There was no significant difference in this ratio in PBMCs from Simvastatin users and controls. The mtDNA, and nDNA content was not quantified in platelets, because platelets are anuclear. Mitochondrial superoxide is higher in Simvastatin users than in controls. Mitochondrial superoxide was measured as a representative of mitochondrial produced ROS in platelets and PBMCs from Simvastatin users and controls. The results indicated a 2.52-fold (P = 0.0003) increase in superoxide in platelets, but no significant difference in superoxide was detected in PBMCs from Simvastatin users, and controls (Fig. 3A,C). Superoxide level was compared in Simvastatin users with myalgia (N = 12), and Simvastatin users without myalgia (N = 22). The results revealed that the elevated superoxide levels are strongly associated with reported myalgia, with a 3.43-and 1.58-fold higher level of superoxide in platelets and PBMCs, respectively in Simvastatin users with SAM than in Simvastatin users without (Fig. 3B,D) (P < 0.0001; and P = 0.0075). Similarly, there was a significant decrease in superoxide in platelets and PBMCs when comparing Simvastatin users not reporting SAM to Simvastatin users reporting myalgia (P < 0.0035; and P = 0.045) (Fig. 3B,D). Ubiquinone has no effect on mitochondrial respiration or mitochondrial superoxide. To investigate the effects of UQ on mitochondrial respiration, 13 Simvastatin users were treated with 2 × 200 mg UQ per day for eight weeks, while 9 Simvastatin users were given a placebo 38 . Mitochondrial OCR, basal respiratory rate, ATP turnover, and reserve respiratory capacity were quantified before and after treatment with either UQ or placebo; however, the results revealed no significant differences in these mitochondrial parameters ( Figure S3). Similarly, the abundance of mitochondrial superoxide in platelets, and PBMCs was not significantly different before, and after dosing with UQ or placebo (Fig. 3E,F,H,I), irrespective of the presence or absence of SAM. These data strongly suggest that UQ, under the conditions tested here, has no impact on the level of mitochondrial superoxide in platelets or PMBCs (Fig. 3G,J). Discussion Statins reduce hypercholesterolemia and are important for the prevention of CVD. However, statin use is frequently associated with myalgia leading to non-compliance with the recommended treatment protocol. To better understand the mechanisms underlying SAM, we assessed mitochondrial parameters and the abundance of superoxide in Simvastatin users, and matched controls, using platelets and PBMCs as a surrogate model for skeletal muscle [28][29][30][31][32][33] . In contrast to previous studies [16][17][18][19][20][21][22][23] , we report here that statin usage is correlated with improved oxidative phosphorylation properties of a tissue. We demonstrate that reserve respiratory capacity, maximal capacity, glycolytic reserve, and complex I and IV activity are higher in platelets and PBMCs from Simvastatin users than in control subjects. At the same time, basal OCR was similar or lower in Simvastatin users than in controls, and cells of Simvastatin users demonstrated a higher glycolytic reserve. These results are consistent with our proposal that platelets and PBMCs from Simvastatin users demonstrate higher respiratory efficiency than control cells, in that they produce more ATP molecules for each oxygen molecule consumed. An increase of reserve respiratory capacity of PBMCs has been associated with gait speed, physical performance, muscle strength and muscle quality in adults and older adults 28,39 . The biological role of the increased reserve respiratory capacity as well as other increased aspects of mitochondrial respiration is unclear. A previous study of hyperlipidemic patients treated for four weeks with 40 mg Simvastatin per day, demonstrated an approximately 50% increase of celullar oxidative consumption and a 25% increase of mitochondrial membrane potential of their polymorphonuclear leukocytes 40 . This supports our findings of a positive correlation between Simvastatin usage and ability of blood cells to perform mitochondrial respiration. However, it is not clear why we see an increased capacity for respiration. The complexes of the ETC can assemble into structures called supercomplexes, which modulates electron transport and ROS production (reviewed in 41 ). Supercomplex formation is believed to be a very dynamic process and of great importance for regulation of mitochondrial respiration 42,43 . Recently, it has been demonstrated that endothelial cells changed their supercomplex formation in response to atorvastatin 44 . Supercomplex formation can be mediated by increased levels of ROS 45 . Conversely, it has been demonstrated that a change of supercomplex conformation resulting in a dissociation of complex I from the supercomplex results in an increased mitochondrial ROS production 46 . In this study we find an increased production of mitochondrial superoxide in platelets and PBMCs of Simvastatin users, and we can demonstrate an increased activity of complex I in these cells. Both factors support our suggestion of a regulatory involvement of supercomplexes, however, an elucidation to this requires a dedicated study. As a caveat to this study, we acknowledge evidence for differential effects of statins on different tissues as has been seen demonstrated by others 47 ; it is therefore possible that statin use improves various functions of mitochondria in blood cells, as reported here, while decreasing these in skeletal muscle. We and others, have reported a decrease of inflammation markers in response to long term statin treatment 38,48 . It is therefore possible that the altered mitochondrial characteristica of PBMCs is linked to lower levels of inflammation. This decrease is not reflected in levels of WBC, neutrophils and monocytes which are slightly elevated in Simvastatin users, but more work is needed to describe the role of statin modulated mitochondria in inflammation. Nevertheless, even though other reports suggest that mitochondrial respiration is impaired in muscle, blood, and liver of Simvastatin users [18][19][20][21][22] , we argue that these discrepancies could reflect different experimental approaches, and conditions used in previous studies. www.nature.com/scientificreports/ 1000-fold higher than the physiologically-relevant dose, e.g. 2.5 to 10 µM vs. 1.6-4.3 nM Simvastatin 36,49 . Using Huh-7 cells exposed to 2.5-10 µM Simvastatin for 72 h, we also observed impaired respiration and increased production of mitochondrial superoxide, as reported previously by others. We therefore argue that there is an unmet need to investigate the mitochondrial respiration of both muscle and blood cells in persons receiving Simvastatin. Preferably in a cohort that is monitored since before the first Simvastatin administration and at least 6 month into the treatment. Importantly, we report here that increased myalgia trend with a significant increase in mitochondrial superoxide in PBMCs and platelets from statin users, while myalgia did not trend with other measures of mitochondrial respiration. This finding agrees with a recent study demonstrating a 41% increase in mitochondrial superoxide in mouse skeletal myotubes exposed to atorvastatin 50 . The study also demonstrated a statin-induced increase in glutamate efflux mediated by the xCcysteine/glutamate antiporter, and argues that these factors could explain SAM 50 . In a previous study on the same cohort, we demonstrated a five-fold increase in serum levels of UQ 51 . However, here we report that mitochondrial respiration and the abundance of superoxide in blood cells were similar in Simvastatin users dosed with UQ or placebo for 8 weeks independent of the presence or absence of SAM. Admittedly, the sample size for this experiment is very small (N = 7 Simvastatin users with myalgia); therefore, it remains possible that a beneficial effect of UQ on myalgia could be observed in a future study on a larger patient cohort. Nevertheless, other studies generally agree with our findings, indicating limited benefit of UQ towards SAM 38,52 . In conclusion, this study demonstrates that long-term treatment with Simvastatin increases mitochondrial respiratory capacity in PBMCs and platelets. Reported side effects of Simvastatin usage are correlated to increased levels of mitochondrial superoxide and do not correlate with any mitochondrial respiratory parameters measured to date. Therefore, elevated mitochondrial superoxide in peripheral blood cells has potential as a biomarker for SAM. Participants. This study is part of the interdisciplinary project "Living with statins" (LIFESTAT) 7 . The participants used in this study were also examined in other studies [53][54][55] including a study by Dohlman et al., where muscle tissue was studied in detail 16 . Informed consent have been obtained from all participants in this study. The experiments presented in this study was conducted based on two sub-studies from LIFESTAT; A crosssectional study, and an interventional study. In the cross-sectional study statin-users with or without myalgia were compared to hypercholesteraemic controls (no therapy). The intervention consisted of eight weeks of UQtherapy in Simvastatin-users, and controls, as described previously 38 Platelet and PBMC purification. Venous blood was sampled in 6.0 mL BD Vacutainer blood collection tubes containing EDTA (BD Biosciences). The participants were fasted overnight and sampled the following morning. PBMCs were isolated as previously described 56 with minor changes, including a platelet purification step as described 57 . Platelets, and PBMCs were used immediately for extracellular flux assay, and mitochondrial superoxide measurement, or cryopreserved in freezing medium (70% FBS, 25% Roswell Park Memorial Institute (RPMI) 1640, (Thermo Fisher Scientific-Life tech.), 5% DMSO) for later testing of ETC complex I, and complex IV activity as well as mtDNA levels by qPCR. SYSMEX haematology analyser. Blood samples were analysed for complete blood cell count, red blood cells, leukocytes, and platelets using a Sysmex XN automated haematology analyser XN (Sysmex Corporation, Kobe, Japan). Furthermore, platelets, and PMBCs were purified, and monocytes, lymphocytes, and neutrophils were counted in whole blood, and the PBMC fraction. Determination of mitochondrial respiration and extracellular acidification rate. OCR and ECAR were quantified using Agilent Seahorse Extracellular Flux (XF) technology on an XF Analyzer (Agilent Technologies). Cells were seeded in a Seahorse XF plate using Cell-Tak adherent (Corning) with 4 × 10 4 Huh-7 cells, 2 × 10 7 platelets or 6 × 10 5 PBMCs per well, and resuspended in Seahorse assay media (Seahorse Bioscience, Agilent) containing 25 mM glucose, 2 mM pyruvate, 2 mM glutamine, adjusted to pH 7.4. OCR and ECAR was measured in the presence of oligomycin (1 µM for Huh-7 cells, and 0.5 µM for platelets and PBMCs) or Carbonyl Cyanide-4-(triFluoromethoxy) Phenylhydrazone (FCCP) (0.75 µM for Huh-7 cells, 0.9 µM for platelets, and 0.6 µM for PBMCs). All samples were then treated with 2 µM antimycin A as a control. Samples were measured as the median of 4 or 5 technical replicates. www.nature.com/scientificreports/ Determination of the specific activity of electron transport chain complex I and complex IV. Complex I (rotenone insensitive NADH-dehydrogenase specific), and complex IV dipstick assays (Abcam kits 109720, and 109876) were performed using cryopreserved platelets, as described previously 58 . Determination of mitochondrial produced superoxide. Mitochondrial Determination of relative mitochondrial DNA levels by quantitative PCR. Genomic DNA was extracted from 1 × 10 6 cryopreserved PBMCs from Simvastatin-treated subjects with or without myalgia, and from controls. Extraction was performed using the GeneJET genomic DNA purification kit (Thermo Fish. Sci. Life). mtDNA to nDNA ratio was analyzed by qPCR using the StepOnePlus real-time PCR system (Applied Biosystems, and threshold (double delta Ct) values, as described previously 59 ). tRNALeu (UUR) was the reference mitochondrial gene, and β-2-microglobulin (β2M) was the nuclear reference gene, as described previously 60,61 . The mtDNA/nDNA ratio was determined by threshold (double delta Ct) values as described 59 Statistics. For comparison of mitochondrial respiration, ECAR, complex I, and IV activity, mtDNA content, and mitochondrial superoxide, significant difference between controls, and Simvastatin users was calculated using two-tailed unpaired Student's t-test. For comparison of effect of increasing Simvastatin treatment on mitochondrial respiration, and superoxide levels, as well as between controls, Simvastatin users experiencing myalgia, and Simvastatin users not experiencing myalgia, single classification analysis of variance (ANOVA) was used. Assumptions of normality were checked by visual inspection prior to ANOVA. When the ANOVA indicated significant differences, Dunnett's (mitochondrial respiration) or Tukey's (mitochondrial superoxide) honestly significant method was used to test for differences. P values below 0.05 were considered significant. Results from Huh-7 cells are presented as mean ± SD, while results from human subjects are presented as mean ± SEM. Statistical analyses were performed using GraphPad Prism version 8.3 (GraphPad Software, La Jolla California USA).	2020-10-14T13:05:38.340Z	2020-10-12T00:00:00.000	{ "year": 2020, "sha1": "10be00b4b5a79ec897980b4fc8457b04c84594e1", "oa_license": "CCBY", "oa_url": "https://www.nature.com/articles/s41598-020-73896-2.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "834bc8d9ff70b6aab22cc8d5c10898a718c76c1e", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
221402321	pes2o/s2orc	v3-fos-license	A Combined Measurement Method for Large-Size Aerospace Components Automated and high-accuracy three-dimensional (3D) shape measurement is required in quality control of large-size components for the aerospace industry. To eliminate the contradiction between global measurement and local precision measurement control in 3D digitalization for the key local features of the large-size components, a combined measurement method is proposed, including a 3D scanner, a laser tracker, and an industrial robot used as an orienting device, to achieve high-accuracy measurement. As for improving the overall measurement accuracy, an accurate calibration method based on coordinate optimization of common points (COCP) and coordinate optimization of global control points (COGP) is proposed to determine the coordinate systems. Firstly, a coordinate optimization method of common points (COCP) is recommended. Then, a coordinate optimization method of global control points (COGP) based on the angular constraint is proposed for minimizing the measurement errors and improving the measurement accuracy of the position and orientation of the 3D scanner. Finally, a combined measurement system is established, and validation experiments are carried out in laboratory within a distance of 4 m. The calibration experiment results demonstrate that the max and mean errors of the coordinate transformation have been reduced from 0.037 and 0.022 mm to 0.021 and 0.0122 mm. Additionally, the measurement experiment results also show that the combined measurement system features high accuracy. Introduction Large-size components with a large number of supports are commonly seen in modern advanced manufacturing, especially in the aerospace industry. The machining quality of the key local features on the support mounting surface directly impacts the quality of assemblies between the large-scale component and external instruments. On-line machining is necessary because high-precision and high-efficiency machining cannot presently be achieved through off-line processing. Therefore, to ensure on-line machining accuracy, it is essential to measure the key local features on-line at the product design coordinates. The key 3D and geometrical information derived from the measurement includes the component's position and orientation, the 3D shape, and the location of the key local features on the supports. Furthermore, the measuring range (3 × 3 m to 5 × 6 m) is large, which makes it challenging to locate the support with high accuracy and simultaneously acquire 3D information of the key local feature (25 × 25 mm) with high accuracy (±0.035 mm). The requirement of measuring the key local features in large-scale on-site machining usually cannot be met by the employment of a single measuring device. Additionally, all information needs to be obtained at the same time and transformed into a unified coordinate system, which also makes the large-scale 3D measurement extremely challenging. Various methods and systems have been proposed for the large-scale 3D measurements [1,2]. The three-coordinate measuring machines (CMMs) have been extensively applied in 3D measurement due to their high accuracy and excellent stability. With the development of noncontact optical measuring equipment and computer vision techniques, visual sensors have been integrated with traditional CMMs [3][4][5]. However, due to the limitation of CMM measurement efficiency, this allows only a small percentage of products to be sampled and inspected. Furthermore, CMM is less used in on-site and on-line measurement due to its structure. When 3D shape information of large-sized components needs to be measured, the 3D shape measurement system integrated with photogrammetry and fringe projection [6] is widely used. However, reflective markers on the target must be attached before the measurement, which interferes with the morphology of the measured part and affects the measurement efficiency. Recently, industrial robots have been extensively applied in the manufacturing field as economical and flexible orienting devices. Therefore, an increasing number of visual sensors are being integrated into robots. Laser scanning, a technology for large-scale 3D shape measurement [7][8][9][10], is more available and economical. However, laser scanning only collects data along limited lines for each measurement, which may result in the robot scanning results containing ripples. Thus, the measurements of the key local features remain barriers to obtaining high accuracy. To extend the measuring range of the laser scanner at designated measurement positions, a movement mechanism [11,12] has been integrated into the laser scanning system. In addition, a linear track or a rotary system will be required to be put into use. However, the movement mechanism inevitably gives rise to errors, reducing the measurement accuracy. The movement mechanism should be calibrated to ensure measurement accuracy. Compared with the laser scanning, structured light profilometry [13][14][15][16][17][18][19][20] has been well developed and widely used for scanning the surface of the object rapidly, as well as acquiring a high-density and high-quality point cloud of a region for each measurement. If the calibration process of the visual sensors is well designed and implemented, their measurement accuracy can be guaranteed [21,22]. Furthermore, compared with the line scanning method, structured light profilometry has a much bigger scanned area and thus is more efficient. Due to the large-size geometry of the component and the finite measuring range of a single station, it is difficult to guarantee the overall measurement accuracy. To further expand the measurement range for measuring large-size objects and guarantee the overall measurement accuracy, more external measurement devices are being integrated into 3D shape measurement systems [23][24][25][26], such as indoor GPS (iGPS) systems, total stations, and laser trackers. Du, F. et al. [23] developed a large-scale 3D measurement system that combines iGPS, a robot, and a portable scanner. However, the overall measurement accuracy is limited by the measurement property of the iGPS. Paoli, A. et al. [24] developed a 3D measurement system that combines a 3D scanner, a total station, and a robot for automating the measurement process of hull yacht shapes. Several optical target corner cube reflectors are mounted on the robotic system basis and tracked by a total station. However, the robot positioning error is inevitably introduced, and the measurement accuracy is restricted by the robot positioning accuracy. Leica developed a large-scale 3D shape measurement system [25] by combining a laser tracker, T-scan, and a robot. However, it was too expensive to be widely adopted. Du, H. et al. [26] proposed a robot-integrated 3D scanning system, which combines a 3D scanner, a laser tracker, and a robot. The scanner is carried by the robot heading to the planned measurement position during operation. Its end coordinate system is created by rotating the 3D scanner, which is tracked by the laser tracker. However, the laser tracker cannot detect and control the measurement errors during the measurement process. As the requirements for accuracy have continued to increase, the current measurement methods and systems mentioned above cannot meet the present requirements for high-accuracy on-line measurement of key local features. Besides, the study of the error control in measurement systems is Sensors 2020, 20, 4843 3 of 18 limited. Therefore, a combined measurement method for the large-scale 3D shape measurement of key local features is proposed, combining a 3D scanner, a laser tracker, and an industrial robot. On the basis, a novel calibration method is carried out. The remainder of the paper is structured as follows: Section 2 introduces the combined measurement method in detail. The calibration of the measurement system is described in Section 3. In Section 4, the proposed method is verified through calibration experiments and measurement experiments, and concluding remarks are provided in Section 5. Combined Measurement Method The proposed combined measurement system mainly incorporates a laser tracker, a 3D scanner, and a robot, as shown in Figure 1. Based on the fringe projection technique, the 3D scanner can capture local high-accuracy 3D shape information. The global measurement method adopts laser trackers to ensure the unity of overall and local measurement accuracy. Additionally, the industrial robot is introduced as the orienting device to improve the efficiency. The 3D scanner will be carried by the robot heading to the discrete measured regions. Then, the laser tracker measures the spherically mounted reflectors (SMRs) rigidly mounted on the base of the 3D scanner. In this way, the position and orientation of the 3D scanner can be acquired. Besides, the acquired data are unified in the world coordinate system defined by the laser tracker. Sensors 2020, 20, x FOR PEER REVIEW 3 of 19 key local features is proposed, combining a 3D scanner, a laser tracker, and an industrial robot. On the basis, a novel calibration method is carried out. The remainder of the paper is structured as follows: Section 2 introduces the combined measurement method in detail. The calibration of the measurement system is described in Section 3. In Section 4, the proposed method is verified through calibration experiments and measurement experiments, and concluding remarks are provided in Section 5. Combined Measurement Method The proposed combined measurement system mainly incorporates a laser tracker, a 3D scanner, and a robot, as shown in Figure 1. Based on the fringe projection technique, the 3D scanner can capture local high-accuracy 3D shape information. The global measurement method adopts laser trackers to ensure the unity of overall and local measurement accuracy. Additionally, the industrial robot is introduced as the orienting device to improve the efficiency. The 3D scanner will be carried by the robot heading to the discrete measured regions. Then, the laser tracker measures the spherically mounted reflectors (SMRs) rigidly mounted on the base of the 3D scanner. In this way, the position and orientation of the 3D scanner can be acquired. Besides, the acquired data are unified in the world coordinate system defined by the laser tracker. A calibration method is proposed to improve the overall measurement accuracy of the proposed system. The relationship between the intermediate coordinate system set on the 3D scanner and the 3D scanner measurement coordinate system is deduced by the method. As for improving the overall measurement accuracy, an accurate calibration method based on coordinate optimization of common points (COCP) and coordinate optimization of global control points (COGP) is proposed to determine the coordinate systems. In the calibration process, both the 3D scanner and the laser tracker are used at different positions for measuring the common points and acquiring redundant data. Firstly, the COCP is recommended. Then, the 3D data of the target are obtained by moving the multiview 3D scanner. Meanwhile, the position and orientation of the 3D scanner at each position are acquired by the laser tracker. Moreover, the COGP based on the angular constraint is proposed for controlling the L aser track er A calibration method is proposed to improve the overall measurement accuracy of the proposed system. The relationship between the intermediate coordinate system set on the 3D scanner and the 3D scanner measurement coordinate system is deduced by the method. As for improving the overall measurement accuracy, an accurate calibration method based on coordinate optimization of common points (COCP) and coordinate optimization of global control points (COGP) is proposed to determine the coordinate systems. In the calibration process, both the 3D scanner and the laser tracker are used at different positions for measuring the common points and acquiring redundant data. Firstly, the COCP is recommended. Then, the 3D data of the target are obtained by moving the multiview 3D scanner. Meanwhile, the position and orientation of the 3D scanner at each position are acquired by the laser tracker. Moreover, the COGP based on the angular constraint is proposed for controlling the measurement errors and improving the measurement accuracy for the position and orientation of the 3D scanner. Measurement Model The coordinate systems of the proposed system comprise the 3D scanner measurement coordinate system O S − X S Y S Z S (SCS), the intermediate coordinate system O I − X I Y I Z I (ICS), and world coordinate system O W − X W Y W Z W (WCS). To provide laser trackers with the position and orientation of the 3D scanner, ICS is created on the 3D scanner framework and serves as a fixed reference frame of the SCS. Besides, to get the complete data for the key local features on the support mounting surface, the measurement should be implemented multiple times at different stations. Then, the acquired multiview data can be converted into WCS. The schematic of the combined measurement model is shown in Figure 2. Sensors 2020, 20, x FOR PEER REVIEW 4 of 19 measurement errors and improving the measurement accuracy for the position and orientation of the 3D scanner. Measurement Model The coordinate systems of the proposed system comprise the 3D scanner measurement To provide laser trackers with the position and orientation of the 3D scanner, ICS is created on the 3D scanner framework and serves as a fixed reference frame of the SCS. Besides, to get the complete data for the key local features on the support mounting surface, the measurement should be implemented multiple times at different stations. Then, the acquired multiview data can be converted into WCS. The schematic of the combined measurement model is shown in Figure 2. H , as an invariable during the measurement process, reflects the coordinate transformation relationship between ICS and SCS. As the calibration of transformation for the ICS and the SCS is viewed as extrinsic parameter calibration, an accurate calibration method will be introduced in the following section. Calibration of the Measurement System The high-precision model for coordinate transformation between SCS and ICS is established by extrinsic parameter calibration. Additionally, the extrinsic parameter calibration has to be performed before the combined measurement system is applied for large-scale metrology, and there is no calibration procedure during the measurement process. Therefore, an accurate extrinsic parameter Figure 2. Schematic of the combined measurement model. With P being a visual point on the workspace, the coordinate mapping model between WCS and SCS is expressed as follows: where P W and P S are the homogeneous coordinates of the visual point P in WCS and SCS, respectively. W H I and I H S are 4 × 4 homogeneous transformation matrices. W H I denotes the transformation relationship between WCS and ICS while I H S , as an invariable during the measurement process, reflects the coordinate transformation relationship between ICS and SCS. As the calibration of transformation for the ICS and the SCS is viewed as extrinsic parameter calibration, an accurate calibration method will be introduced in the following section. Calibration of the Measurement System The high-precision model for coordinate transformation between SCS and ICS is established by extrinsic parameter calibration. Additionally, the extrinsic parameter calibration has to be performed before the combined measurement system is applied for large-scale metrology, and there is no calibration procedure during the measurement process. Therefore, an accurate extrinsic parameter calibration result is a critical factor in ensuring the overall measurement accuracy of the proposed system. As for improving the overall measurement accuracy and minimizing the measurement errors, a extrinsic parameter calibration method based on COCP and COGP optimization is proposed. Firstly, the COCP is recommended. Then, the COGP based on the angular constraint is proposed for minimizing the measurement errors and improving the measurement accuracy of the position and orientation of the 3D scanner. Calibration Principle The homogeneous coordinates of the points P i in two coordinate systems can be denoted as P i1 = (x pi1 , y pi1 , z pi1 , 1) and P i2 = (x pi2 , y pi2 , z pi2 , 1). The relationship can be expressed as follows: cos α cos β cos α sin β sin γ − sin α cos γ cos α sin β cos γ + sin α sin γ sin α cos β sin α sin β sin γ + cos α cos γ sin α sin β cos γ − cos α sin γ − sin β cos β sin γ cos β cos γ where H is the homogeneous transformation matrix. R(α β γ) is a rotation matrix, and α, β, γ are the angles of Cardan. T is a translation vector. The combined calibration system consists of a 3D scanner, a laser tracker, an industrial robot, and the calibration target. The principle of extrinsic parameter calibration is shown in Figure 3. Firstly, to establish the transformation relationship G H S between SCS and GCS, the laser tracker and the 3D scanner measure the common points arranged on the calibration target. Then, the laser tracker can locate and track the position and orientation of the 3D scanner by measuring the coordinates of the global control points, and the transformation matrix G H I is established after the process. Finally, according to the two transformation matrixes above, the extrinsic parameter matrix I H S can be calculated. calibration result is a critical factor in ensuring the overall measurement accuracy of the proposed system. As for improving the overall measurement accuracy and minimizing the measurement errors, a extrinsic parameter calibration method based on COCP and COGP optimization is proposed. Firstly, the COCP is recommended. Then, the COGP based on the angular constraint is proposed for minimizing the measurement errors and improving the measurement accuracy of the position and orientation of the 3D scanner. Calibration Principle The homogeneous coordinates of the points i P in two coordinate systems can be denoted as . The relationship can be expressed as follows: where H is the homogeneous transformation matrix. () R    is a rotation matrix, and ,,    are the angles of Cardan. T is a translation vector. The combined calibration system consists of a 3D scanner, a laser tracker, an industrial robot, and the calibration target. The principle of extrinsic parameter calibration is shown in Figure 3. Four or more noncollinear SMRs set on the 3D scanner, known as global control points, are denoted as Q ci , ci = 1, 2, · · · , cn. Besides, the homogeneous coordinates of Q ci can be denoted as X I ci , Y I ci , Z I ci , 1) in ICS. Two groups of target observation points replaced by each other are arranged on the calibration target for ensuring the accuracy of extrinsic parameter calibration. As the standard ceramic spheres take the place of SMRs, the sphere centers of the SMRs are basically in the same positions as those of Sensors 2020, 20, 4843 6 of 18 sphere centers of the standard ceramic spheres. The homogeneous coordinates of the laser tracker observation points Q gi , gi = 1, 2, · · · , gn can be denoted as P gi = (X G gi , Y G gi , Z G gi , 1) in laser tracker measurement coordinate system O G − X G Y G Z G (GCS), and the homogeneous coordinates of the 3D scanner observation points Q si , si = 1, 2, · · · , sn can be denoted as P si = (X S si , Y S si , Z S si , 1) in SCS; the following relationship between them exists: The homogeneous coordinates of Q ci can be denoted as P ci = (X G ci , Y G ci , Z G ci , 1) in GCS, and then the following relationship between Q ci in ICS and P ci in GCS is expressed as follows: where G H I is the transformation matrix between GCS and ICS. According to Equations (2)-(4), the I H S can be calculated as follows: Improving the accuracy of transformation matrixes G H S and G H I is the key to improving the accuracy of transformation matrix I H S . However, the measurement errors of the laser tracker and the 3D scanner could lead to transformation parameter errors. Therefore, to improve the accuracy of extrinsic parameter calibration by minimizing measurement errors, the coordinate optimization method of the common points is proposed in Section 3.2, and the coordinate optimization method of the global control points is proposed in Section 3.3. Optimization of the Coordinates of Common Points The common points arranged on the calibration target are measured by both the laser tracker and the 3D scanner. COCP is proposed to minimize measurement errors. In addition, it optimizes the transformation parameters of the G H S , which consists of three angle parameters in matrix G R S (α β γ) and three translation parameters in G T S ( T x T y T z ). Figure 4 shows common points that are measured at M positions. If the coordinate system of the first position is taken as the reference coordinate system, the Cartesian coordinates of the common points obtained by the laser tracker and the 3D scanner at first position can be denoted as P G0 gi = (X G0 gi , Y G0 gi , Z G0 gi ) and P S0 si = (X S0 si , Y S0 si , Z S0 si ), respectively. The Cartesian coordinates of the common points measured in other positions can be denoted as p Gm gi = (x Gm gi , y Gm gi , z Gm gi ), m = 1, · · · , M − 1; gi = 1, · · · , gn and p Sm si = (x Sm si , y Sm si , z Sm si ), m = 1, · · · , M − 1; si = 1, · · · , sn. If measurement errors are considered, Equation (2) can be rewritten as follows: represent the correction values for common points of the laser tracker and the 3D scanner, respectively. The simultaneous equation of measurement errors is as follows: Sensors 2020, 20, 4843 of 18 The rotation and translation matrix between the laser tracker and 3D scanner at all positions can be computed by the Procrustes method [27]. Based on this, the correction value of coordinates of the common points can be calculated by the rank-defect network adjustment algorithm [28]. As a result, the coordinate values of the common points are optimized. Thereby, the accuracy of the transformation parameters of G S H is improved. Optimization of the Coordinates of Global Control Points Due to environmental uncertainties and instrument instability, errors in the measurement of global control points are unavoidable. To solve the problem of unknown and uncontrollable error in measuring the global control points and optimize the transformation parameters of G I H , COGP based on the angular constraint is proposed to obtain the correction values of coordinates of the global control points. Because the angle between two vectors in Euclidean space is independent of the coordinate system [29], the geometric information of the global control points set on the 3D scanner is fully used. In four global control points, a vector is established by two target points, and the angle between the vectors 13 q and 24 q is  , as shown in Figure 5; the four target points can make up six vectors, which can form 15 angles. The rotation and translation matrix between the laser tracker and 3D scanner at all positions can be computed by the Procrustes method [27]. Based on this, the correction value of coordinates of the common points can be calculated by the rank-defect network adjustment algorithm [28]. As a result, the coordinate values of the common points are optimized. Thereby, the accuracy of the transformation parameters of G H S is improved. Optimization of the Coordinates of Global Control Points Due to environmental uncertainties and instrument instability, errors in the measurement of global control points are unavoidable. To solve the problem of unknown and uncontrollable error in measuring the global control points and optimize the transformation parameters of G H I , COGP based on the angular constraint is proposed to obtain the correction values of coordinates of the global control points. Because the angle between two vectors in Euclidean space is independent of the coordinate system [29], the geometric information of the global control points set on the 3D scanner is fully used. In four global control points, a vector is established by two target points, and the angle between the vectors q 13 and q 24 is θ, as shown in Figure 5; the four target points can make up six vectors, which can form 15 angles. CMM is used to calibrate the angle values, which are set as the nominal angles. We can obtain the angle error equation by calculating the difference between the actual angle values measured from the on-site measurement and the nominal angle values. Then, the normal equation is obtained by the least squares method. The method for finding the angle between two nonzero vectors is expressed as follows: cos θ = q 13 ·q 24 q 13 q 24 (8) Therefore, the angle θ could be obtained by the arc cosine function as follows: CMM is used to calibrate the angle values, which are set as the nominal angles. We can obtain the angle error equation by calculating the difference between the actual angle values measured from the on-site measurement and the nominal angle values. Then, the normal equation is obtained by the least squares method. The method for finding the angle between two nonzero vectors is expressed as follows: cos   13 24 13 24 qq qq (8) Therefore, the angle  could be obtained by the arc cosine function as follows: Equation (9) is expanded by Taylor's formula, and the second-order term is ignored. Therefore, the linearized equation of angular constraint is expressed as follows: where ,,    The angle error equation is expressed as follows: where  i are nominal angles, and 0  are actual angles. In an alternative way, Equation (10) can be rewritten as follows: Equation (9) is expanded by Taylor's formula, and the second-order term is ignored. Therefore, the linearized equation of angular constraint is expressed as follows: where ∆ x i , ∆ y i , ∆ z i are the optimized correction values of coordinates of the global control points, is a vector of the correction values. The angle error equation is expressed as follows: whereθ i are nominal angles, and θ 0 are actual angles. In an alternative way, Equation (10) can be rewritten as follows: where d =θ i − θ 0 i is the angle error vector. Then, the normal equation is as follows: where U is the weight matrix. The objective function for finding the best coordinate estimates can be expressed as follows: ∆ x 1 , ∆ y 1 , ∆ z 1 , · · · , ∆ x n , ∆ y n , ∆ z n = arg min Sensors 2020, 20, 4843 of 18 The angle adjustment is conducted to obtain the optimal estimate value by solving the error matrix equation in the least squares norm. However, the traditional least squares method requires the coefficient matrix to be a nonsingular or full rank matrix. The coefficient matrix B T UB in Equation (13) is an ill-conditioned matrix with maximum condition number cond(B T UB) = λ max /λ min (λ max and λ min represent the maximum and minimum eigenvalues of coefficient matrix B T UB), and the result of this solution is extremely unstable. To obtain the optimal solution, a two-objective optimization formula can be constructed as follows: According to Tikhonov's regularization method, the objective function of Equation (15) based on the ridge estimation algorithm is given as follows: where the non-negative parameter α is the ridge estimation parameter, and I is the unit matrix. Finding the conjugate gradient of ∆ X in Equation (16), we have According to the extremum condition, let Equation (17) be equal to 0. Therefore, the final solution can be expressed as follows: where the damping term αI added to the main diagonal of the coefficient matrix B T UB in Equation (18) can overcome the ill-conditioned effect of the coefficient matrix. Thus, a stable solution can be obtained. The ridge estimation method can change singular matrix B T UB into a nonsingular matrix. Besides, it ensures the stability for the solution of the ill-conditioned equation. The appropriate ridge parameter α can be solved by the L-curve method [30], which can reduce the condition number of the equation and change the ill-conditioned equation into a well-conditioned equation. As a result, the coordinate values of the global control points are optimized. Thereby, the accuracy of the transformation parameters of G H I is improved. Experiments and Discussion Based on the principle detailed in Section 3.1, the extrinsic parameter calibration was carried out. The calibration method mentioned in Section 3 can be verified through calibration experiments and measurement experiments. The experimental setup of the combined measurement system is shown in Figure 6. The 3D scanner was mounted on the 6-DOF industrial robot KR10R1420 manufactured by KUKA Corporation, and several SMRs set on the 3D scanner ensure that the position and orientation of the 3D scanner can be tracked by the laser tracker. The laser tracker was the Leica AT960, with an accuracy of 0.015 mm ± 0.006 mm/m, which can be connected to the computer by Gigabit Ethernet (GBE). The off-the-shelf visual sensor is a binocular structured light scanner with precision of 0.012 mm, resolution up to 0.020 mm, and scan range of 30 × 40 × 25 mm. Sensors 2020, 20, x FOR PEER REVIEW 10 of 19 orientation of the 3D scanner can be tracked by the laser tracker. The laser tracker was the Leica AT960, with an accuracy of 0.015 mm  0.006 mm/m, which can be connected to the computer by Gigabit Ethernet (GBE). The off-the-shelf visual sensor is a binocular structured light scanner with precision of 0.012 mm, resolution up to 0.020 mm, and scan range of 30 × 40 × 25 mm. Figure 7 shows a specific calibration target designed for extrinsic parameter calibration and experiment purposes. Six sophisticated magnetic nests (SMNs) for the SMRs of laser tracker and the standard ceramic spheres of the 3D scanner were rigidly assembled on the aluminum plate. The standard ceramic spheres and the SMRs were first inspected on a CMM, and the maximum deviation of the diameter of the spheres was found to be about 0.003 mm. Therefore, supposing that the standard ceramic spheres take the place of the SMRs, the SMR center position would be aligned with that of the standard sphere center. All experiments were performed in a stable laboratory environment. The temperature varied between 22 and 23 °C, and the relative humidity varied between 55 and 60%. Figure 7 shows a specific calibration target designed for extrinsic parameter calibration and experiment purposes. Six sophisticated magnetic nests (SMNs) for the SMRs of laser tracker and the standard ceramic spheres of the 3D scanner were rigidly assembled on the aluminum plate. The standard ceramic spheres and the SMRs were first inspected on a CMM, and the maximum deviation of the diameter of the spheres was found to be about 0.003 mm. Therefore, supposing that the standard ceramic spheres take the place of the SMRs, the SMR center position would be aligned with that of the standard sphere center. All experiments were performed in a stable laboratory environment. The temperature varied between 22 and 23 • C, and the relative humidity varied between 55 and 60%. Sensors 2020, 20, x FOR PEER REVIEW 10 of 19 orientation of the 3D scanner can be tracked by the laser tracker. The laser tracker was the Leica AT960, with an accuracy of 0.015 mm  0.006 mm/m, which can be connected to the computer by Gigabit Ethernet (GBE). The off-the-shelf visual sensor is a binocular structured light scanner with precision of 0.012 mm, resolution up to 0.020 mm, and scan range of 30 × 40 × 25 mm. Figure 7 shows a specific calibration target designed for extrinsic parameter calibration and experiment purposes. Six sophisticated magnetic nests (SMNs) for the SMRs of laser tracker and the standard ceramic spheres of the 3D scanner were rigidly assembled on the aluminum plate. The standard ceramic spheres and the SMRs were first inspected on a CMM, and the maximum deviation of the diameter of the spheres was found to be about 0.003 mm. Therefore, supposing that the standard ceramic spheres take the place of the SMRs, the SMR center position would be aligned with that of the standard sphere center. All experiments were performed in a stable laboratory environment. The temperature varied between 22 and 23 °C, and the relative humidity varied between 55 and 60%. Incidence Angle Experiment To minimize measurement errors, the influence of the incident angle error of the laser ray on the measurement accuracy of the laser tracker was analyzed under different circumstances. Figure 8a shows the SMR installed at the same height as the laser tracker. Rotating the SMR around the normal, vertical axis 1, and vertical axis 2, we collected 100 measurements for each operation to obtain the average value. The measurement results in Table 1 show that the maximum error can reach 0.008 mm with the change of SMR pose. Figure 8b also shows the SMR installed at the same height as the laser tracker. In the laser tracker measurement coordinate system, the five measuring points were distributed in a straight line and parallel to the XOY plane. The average value was also obtained by measuring 100 times within 2 m. The results of the analysis are shown in Table 2. Incidence Angle Experiment To minimize measurement errors, the influence of the incident angle error of the laser ray on the measurement accuracy of the laser tracker was analyzed under different circumstances. Figure 8a shows the SMR installed at the same height as the laser tracker. Rotating the SMR around the normal, vertical axis 1, and vertical axis 2, we collected 100 measurements for each operation to obtain the average value. The measurement results in Table 1 show that the maximum error can reach 0.008 mm with the change of SMR pose. Figure 8b also shows the SMR installed at the same height as the laser tracker. In the laser tracker measurement coordinate system, the five measuring points were distributed in a straight line and parallel to the XOY plane. The average value was also obtained by measuring 100 times within 2 m. The results of the analysis are shown in Table 2. Incidence Angle Experiment To minimize measurement errors, the influence of the incident angle error of the laser ray on the measurement accuracy of the laser tracker was analyzed under different circumstances. Figure 8a shows the SMR installed at the same height as the laser tracker. Rotating the SMR around the normal, vertical axis 1, and vertical axis 2, we collected 100 measurements for each operation to obtain the average value. The measurement results in Table 1 show that the maximum error can reach 0.008 mm with the change of SMR pose. Figure 8b also shows the SMR installed at the same height as the laser tracker. In the laser tracker measurement coordinate system, the five measuring points were distributed in a straight line and parallel to the XOY plane. The average value was also obtained by measuring 100 times within 2 m. The results of the analysis are shown in Table 2. In Table 2, the standard deviation of the X-coordinate value σ X is close to the standard deviation of the length measurement result σ r . However, σ Y (the standard deviation of the Y-coordinate value) and σ Z (the standard deviation of the Z-coordinate value) are relatively large. Because the angle measurement accuracy of the laser tracker is relatively low, larger angles result in lower measurement accuracy. Angle measurement error is the main factor affecting the measurement accuracy of the laser tracker. Therefore, the optimization methods for reducing the incident angle error can be adopted to minimize the measurement errors during the measurement process. Figure 6 shows a combined calibration system, which combines a 3D scanner, a laser tracker, an industrial robot, and the calibration target. The common points were measured by the laser tracker and the 3D scanner at eight positions. Firstly, several SMRs were mounted on the SMNs of the calibration target and the 3D scanner, and centers of SMRs were measured by the laser tracker. Then, the 3D scanner measured the standard ceramic spheres mounted on the SMNs of the calibration target. Finally, the optimization data and correction data of common points of the laser tracker and the 3D scanner were obtained according to the method introduced in Section 3.2, as shown in Tables 3 and 4. Table 3. Optimization data and correction data of common points of the laser tracker (unit: mm). Table 5 shows the optimization data and correction data of global control points obtained according to the method introduced in Section 3.3. The distance errors between the measurement values and the nominal values can be calculated by using the coordinate values of global control points optimized by COGP optimization method. The results are shown in Table 6. The max deviation and RMS errors were reduced from 0.0179 and 0.0111 mm to 0.0115 and 0.0074 mm. This demonstrates that the measurement accuracy of global control points can be improved by the angle constraint method. With the optimization data of the global control points, an accurate result of the transformation matrix G H I is given as follows: No With all data mentioned above, the accurate result of the extrinsic parameter calibration can be calculated, according to Equation (5), as follows: As shown in Figure 9, with the accurate calibration result, the max and mean errors of the coordinate transformation were reduced from 0.037 and 0.022 mm to 0.021 and 0.0122 mm. The results demonstrate that the calibration accuracy has been improved significantly, and the calibration method achieves high accuracy. As shown in Figure 9, with the accurate calibration result, the max and mean errors of the coordinate transformation were reduced from 0.037 and 0.022 mm to 0.021 and 0.0122 mm. The results demonstrate that the calibration accuracy has been improved significantly, and the calibration method achieves high accuracy. Measurement Experiment After the extrinsic parameter calibration, further experiments were needed to verify the performance and the accuracy of the proposed system. Section 4.3.1 detail how the performance of the system was assessed by the support measurement test. The measurement accuracy verification is described in Section 4.3.2. Support Measurement Test To measure key local features of the support mounting surface with the size of 300 × 300 × 100 mm, the proposed combined measurement system was applied to measure a support fixed on the experiment platform in the laboratory, as shown in Figure 10a. Firstly, the robot moved the 3D scanner to the discrete regions that needed to be measured, and then the 3D scanner scanned the region. Meanwhile, the laser tracker measured the SMRs fixed on the 3D scanner to obtain the position and orientation of 3D scanner. Finally, the 3D information of the local key features could be obtained by the 3D scanner at different robot positions. Figure 10b shows the 3D point clouds obtained at each measurement position. Furthermore, the 3D point clouds of the local key features were preprocessed and then unified into the world frame according to Equation (1), as shown in Figure 10c. This test demonstrated that the combined measurement system can be applied to measure key local features of the support mounting surface. Accuracy Verification To verify the measurement accuracy of the system, a metric tool that consisted of five homogeneous and standard ceramic spheres with approximate diameter was designed and adopted, as shown in Figure 11. The distance between the centers of two spheres was measured on the CMM, and the results are as follows: D1 = 24.7581 mm, D2 = 149.8429 mm, D3 = 269.7165 mm, and D4 = 299.5461 mm. The distance errors between the measured values and nominal values were computed on the basis of the model in Equation (1). Table 7 presents the distance errors among the centers of the five spheres, which were measured by the combined measurement system within 4 m. To better demonstrate the errors, the mean value ( μ), standard deviation value ( σ ), and root-mean-square error (RMS error) are summarized in Table 5. The proposed method can be verified by these quantitative Accuracy Verification To verify the measurement accuracy of the system, a metric tool that consisted of five homogeneous and standard ceramic spheres with approximate diameter was designed and adopted, as shown in Figure 11. The distance between the centers of two spheres was measured on the CMM, and the results are as follows: D1 = 24.7581 mm, D2 = 149.8429 mm, D3 = 269.7165 mm, and D4 = 299.5461 mm. Accuracy Verification To verify the measurement accuracy of the system, a metric tool that consisted of five homogeneous and standard ceramic spheres with approximate diameter was designed and adopted, as shown in Figure 11. The distance between the centers of two spheres was measured on the CMM, and the results are as follows: D1 = 24.7581 mm, D2 = 149.8429 mm, D3 = 269.7165 mm, and D4 = 299.5461 mm. The distance errors between the measured values and nominal values were computed on the basis of the model in Equation (1). Table 7 presents the distance errors among the centers of the five spheres, which were measured by the combined measurement system within 4 m. To better demonstrate the errors, the mean value ( μ ), standard deviation value ( σ ), and root-mean-square error (RMS error) are summarized in Table 5. The proposed method can be verified by these quantitative The distance errors between the measured values and nominal values were computed on the basis of the model in Equation (1). Table 7 presents the distance errors among the centers of the five spheres, which were measured by the combined measurement system within 4 m. To better demonstrate the errors, the mean value (µ), standard deviation value (σ), and root-mean-square error (RMS error) are summarized in Table 5. The proposed method can be verified by these quantitative statistical results; it results in relatively high accuracy in 3D measurement, with a space length measuring error of less than 0.03%. Conclusions This paper has proposed a combined measurement method for high-accuracy large-scale 3D metrology, where a 3D scanner and a laser tracker are combined to conduct highly accurate measurement of key local features. In this measurement system, the orienting device, an industrial robot, is applied. The principle of measurement has been illustrated in detail. As for improving the overall measurement accuracy, an accurate calibration method based on COCP and COGP optimization has been proposed to determine the coordinate systems. Firstly, the COCP has been recommended. Then, the COGP based on the angular constraint has been proposed for minimizing the measurement errors and improving the measurement accuracy of the position and orientation of the 3D scanner. The calibration experiment results demonstrate that the max and mean errors of the coordinate transformation have been reduced from 0.037 and 0.022 mm to 0.021 and 0.0122 mm. The application of measuring a support has also been performed to demonstrate that the measurement process is simple and efficient. Finally, a metric tool has been developed, which helped to verify the measurement accuracy of the proposed system. Future work will focus on the improvement of measuring accuracy in larger measurement range.	2020-09-02T13:06:38.613Z	2020-08-27T00:00:00.000	{ "year": 2020, "sha1": "0a2309e3f70baf91b21ab031f3ca7f664ffe2501", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1424-8220/20/17/4843/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "7f5d96d78db8acda56c6908a4674d326e10108a5", "s2fieldsofstudy": [ "Engineering" ], "extfieldsofstudy": [ "Medicine", "Computer Science" ] }
54541842	pes2o/s2orc	v3-fos-license	Impact of absorptive capacity and dominant logic on innovation performance of public sector organizations in Hefei ( Anhui Province ) , China Article history: Received: October 1, 2016 Received in revised format: November 16, 2016 Accepted: March 16, 2017 Available online: March 16, 2017 Public organization’s performance depends on multiple aspects in which there are different political and public actors involved. In this study, we discuss the innovation performance of public organizations’ in Hefei (Anhui province), China. Our targeted group for this study were public sector employees at different levels within the organizations being considered. We checked the effect of absorptive capacity (ACAP), and dominant logic on public organization’s innovation performance. We found that absorptive capacity and dominant logic had a positive relationship with firms’ performance. Furthermore, these predictors were not only improving firm’s performance, but also bring innovation into the public organizations. © 2017 Growing Science Ltd. All rights reserved. Introduction In a tumultuous and dynamic environment, knowledge creates significant resources and develops competitive advantage (Teece et al., 1997).In the last 20 years absorptive capacity (ACAP) has become one of the most important constructs, since external knowledge resources have become significant.During the management of the external knowledge ACAP helps the firms develop value and achieve competitive advantage (César Camisón, 2010).In this study our focus is on the effect of ACAP and dominant logic (Chandler & Hamks, 1994) over public service organization's performance.The objective was to combine these two intangible resources in one model to check the public sector organization's performance.In previous research absorptive capacity and dominant logic were examined in the area of private sector firms.After the work of (Cohen & Levinthal, 1989) regarding absorptive capacity, several academic and empirical studies have examined organizational capability to absorb knowledge.The concept of ACAP provides enough flexibility that it can be applied to various units of analysis, and research fields such as industry related organizations, learning in organizations, strategic management and innovation management (Zahra, 2002).DL deals with the organizations' resource allocation decisions and conceptualization.That can be in technology, development of products, distribution, advertising, or in human resources (Prahald, 2004).Moreover, DL is the DNA of the Organization (Prahald, 2004, p.172) and a key variable that is exceptional, valuable, and hard to emulate resources of the organization (Amit, 1993).Whereas, the idea of DL is academically appealing, and still empirical support is weak to date (Oblój & Pratt, 2005).Various terms are used for dominant logic in prior research, (Prahalad & Bettis, 1986), counting "mind-sets" (Nadkarni, 2007), "interconnected choices" (Siggielkov, 2002), and "strategic frames" (Huff, 1982).Each of the terms deal with the managers' ability to distinguish and adapt their firms according to the environment (Obloj, 2010).In this study, we focus on the knowledge and its positive effects on the public organizations' innovation performance.Our study is supported by absorptive capacity and dominant logic.According to Prahalad and Bettis (1986, p.491) Dominant logic is defined as an administrative tool to get future goals to make decisions about business at different levels of organizations processes.Strong test for dominant logic and its relevance are offered by transition economies.Specially, those economies, which are converted from socialist economic system to market economy for example China.In such kind of environment there are more tests offered to dominant logic to test its potential, and importance as intangible resource in limited resource environment (Kolvereid & Oblój, 1994;Bruton, 2008;Meyer & Peng, 2005).Whereas, latest research argues that for competitive advantage intangible resource are the principle resource (Amit, 2003).By definition it is hard to transfer the intangible resources (Szulanski, 2000), more difficult to replicate than tangible/physical resources (Dierickx, 1989), as well as it is harder to deal with market with these resources (Barney, 1986a).Other than the significance of the intangible resources, there has been a little empirical study and their relationship, the resulting capabilities, and the performance of the firms (Carmeli, 2004;Makadok, 2001;Newbert, 2007).This study is based on the following questions:  How absorptive capacity affects the public organization's innovation performance? What role does dominant logic play in order to improve the public organization's innovation performance? The structure of this paper is based on four parts.First part presents the hypothesis and literature review on absorptive capacity, dominant logic and their significance with public organizations' performance.Second part, presents the research method employed for analysis in this study.Third, represents the results and the relationship between variables, by using multiple regression modeling.And the last part is discussion and conclusion. Absorptive capacity and public organizations' innovation performance Prior knowledge and new knowledge may be related to extend the case that knowledge itself as a learning skill.Across the bodies of knowledge there may be transfer of learning skills that are organized and stated in similar ways.On some subsequent learning tasks effect, experience routine on one learning task may impact as well as develop performance (Ellis, 1965).This kind of progressive improvement in the learning tasks' performance is due to the transfer of knowledge that could also be called as "learning to learn" (Ellis, 1965;Estes, 1970).Estes (1970: 16), mentioned the expression "learning to learn" is a misnomer according to which previous experience would not develop recital.Since an individual must be acquainted with learning in a better way, whereas, an individual may have gathered more prior knowledge for attaining a given level of performance he or she needs to learn less (Cohen, 1990).In a variety of tasks succeeding learning tasks can be observed such as, in the case of problemsolving skills.From prior knowledge problem-solving techniques have been composed.This allows individuals to obtain capabilities that are related to problem-solving capabilities.Pirolli and Anderson (1985) worked on computer related programming skills and development and found that approximately every student developed new program by analogy-to-example programs.The accomplishment was measured by their understanding of these cases and why they worked in reality.Developing absorptive capacity in the sense of knowledge in general, problem-solving and leaning skills is inadequate to describe and pinpoint the relevant prior knowledge (Cohen, 1990).Lin et al. (2002) mentioned that it is impossible for a firm to integrate and apply the external knowledge without absorptive capacity.Moreover, they mentioned that during transfer (in technology transfer, for example) critical factors needed for absorptive capacity and found realistic relationship between absorptive capacity and factors related to the diffusion channels for external technology, organizational interactions, and R&D resources.Gupta and Govindarajan (2002) mentioned that for effective knowledge management it is important to build an environment within which individuals work during critical situations that are also challenging in there nature.Two different approaches of knowledge management have been mentioned by (Grundstein, 2007); first, technological approach which deals with the demand solution that is based on the technology.Second, approach deals with the management that is more "people-focused" (Wiig, 2003); in which knowledge is used as a resource to implement strategic vision of the organization.Hogarth (2001) mentioned the capabilities of the knowledge management that contributes to the organizational effectiveness and developed operations of the organizations by generating value.The contributions may expand ability to innovate, improve coordination and reduce redundancy of information/knowledge.Ho (2008) argued that joining knowledge management and human resource management initiatives would help to improve organization's performance.Knowledge management becomes one of the most important factors for these organizations as they identify that competitive advantage centers on effective management of their vast and diverse knowledge assets (Davenport, 1998;Wiig, 2003).Knowledge management goals deal with the development of knowledge assets of the organization to get superior knowledge practices, develop organizational learning behaviors, and improve decision making process to get better performance (King, 2009).Absorptive capacity helps to enhance the information system of the organization.In this context, Mustafa and Flanagan (2012) mentioned that the assimilation capability of absorptive capacity indicates the capability of the organization's process, analyze, explanation, and understanding the information.On the bases of this discussion we argued that: H1. Absorptive capacity is positively related to the organization's innovation performance. Absorptive capacity and dominant logic Some researchers mention that due to dominant logic there is some shared cognitive schema among top managers that are important to manage large firms.Furthermore they mentioned that among top managers dominant logic is a "missing link" between diversification and performance.Moreover, they mentioned that without dominant logic firms cannot get high performance (Charles, 2000).On natureversus-nature database it is argued that, firms are not only dependent on external environment, but also on managers' history as well as their path-dependent choices (Child, 1972;Penrose, 1959;Porter, 1991).The resource-based view (RBV) states that competitive advantage starts inside the organization and takes the form of valuable, difficult to emulate, rare, and diverse resource that builds over time in path-dependent ways (Barney, 1986a;Barney, 1986b;Barney, 1991;Peteraf, 1993;Wernerfelt, 1984;Dierieckx, 1989).It was argued that such resources are beneficial to raise superior performance (Hoopes, 2003;Lippman, 2003;Schoemaker, 1990).Teece (2000) mentioned that at different situations there are different knowledge requirements depending on the cost and demand at work, appropriate administration in which organization performs its functions, compatibility principles, characteristics of innovation, and the prosperity of the technical opportunity facing the organization.Furthermore, they mentioned that managerial challenges that are based on the centrality of information and intellectual property are far more diverse from physical assets that involve gaining competitive advantage.Combining information technology (IT) and co-aligned organizational practices can improve the ability of learning and competitive advantage.Moreover, understanding knowledge can help in knowledge transfer and sharing also helps organizations become innovative and creative.Once the knowledge is clear, it is easier to store, transfer, and share.On the other hand once the data is codified it is harder to protect, it can leak out and can get into the wrong hands quickly.However, strong intellectual property protects this information (Teece, 2000).Meyer (1979) mentioned that public organizations continuously change; due to frequent changes in political environment it is difficult to implement and maintain longterm changes in the public sector organizations.Fernandez (2007) mentioned that some of the recent studies show that a change in top-level management can be effective for organizational change.Boeker (1997) discussed the top managers across organizations and their influence over the organization's decision making process to enter into new markets.The new executives can promote changes in organizations by promoting knowledge transfer, organizational learning, new cognitive models and assumptions, and the displacement of present organizational values (Fernandez, 2007).Simon (1995) mentioned that managers are rationally bounded, they have to depend on simplified world representations to process information.These insufficient representations start from the basis of mental development model and strategic beliefs that constrain managerial decisions.Managerial cognitive capabilities are dependent on the basis of past experience as contrast to current knowledge, that is available in the environment (Kiesler, 1982).For example, over time as managers work together they build up a set of beliefs for the organization that are based on their past experience.In the reference of historical environment regarding development of beliefs, top managers repeatedly have difficulty while changing environment, adapting mental models that also have negative influence over organizational performance (TRIPSAS, 2000).From the above we argue that: H2. Absorptive capacity has a positive influence over dominant logic. Dominant logic and innovation performance Numerous public management studies show that managerial leadership has significant influence on organizational change (Fernandez, 2007).For example, the study of Hennessey (1998) shows the impact of top management on the outcome of public sector organizations.Furthermore, the author mentions the effect of leadership on changes in culture, organizational climate as envisioned by the reinventers, by these changes superior organizational performance can be achieved.Through dynamic capabilities of managers alter their resource base acquire an important resource, combine them together, and recombine them (Eisenhardt, 2000).They integrate sets of particular, identified processes that are commonly accepted as "best practices" which might examine, compare and sometimes transfer from one organization to another organization.This dynamic capability activates and maintains inter-firm performance (Henderson, 1994).Latest theoretical studies focus on the development of managerial decisions as well as flexible strategic choices as important factors in the development of capabilities (Helfat, 2003;Zahra, 2002).Connections concerning strategies and performance have received a huge interest from researchers and produced regular set of findings (Branzei, 2006).Depending on the characteristics of environment and organization (Covin, 1990;Covin & Slevin, 1990;McDougall, 1992;Kelley, 2002), tremendous focus on the strategic choices (Sandberg, 1986), also over time better constancy of these choices.Freeser and Willard (1990) developed survival chances of firms McCann (1991), developed new venture growth, furthermore active advance market as well as financial performance (Schroeder, 2002).Some of these studies describe the relationship of strategies and performance that depend on the resource-based capabilities (Chandler & Hanks, 1994), and fit between firms' strategies and their existing capabilities (Fingenbaum & Karnani, 1991).The above discussion shows that top management is key factor for firms' performance and that their dynamic capabilities have positive effect on the development of the firms' performance.From the discussion above we develop the following hypothesis: H3. Dominant logic is positively related to the organizations' innovation performance. Methodology For this study we collect data from public sector organizations located in Hefei (Anhui province) of China.Different levels of employees were questioned using survey questionnaire that was adopted from Zahra (2002), Obloj (2010), and Alegre and Chiva (2013).We translated the questionnaire into Chinese for the convenience of the target sample (Leung, 2009).In order to have a certainty that the meaning of the translated questionnaire did not change we translated it back into English to verify and concluded that the items had the same meaning. Innovation Performance The ability of organizational learning is related to innovation.Calantone et al. (2002) and Alegre et al. (2008) mentioned that organizational learning is the fact that precedes the innovation.Jiménez-Jiménez and Sanz-Valle (2011) mentioned that innovation, organizational learning and performance are interconnected factors.Innovation may affect directly the organizational performance or work with the innovative performance and increase organizational overall innovativeness (Alegre, 2006).Innovation is the most important element of today's knowledge-based economy in the world for gaining competitive advantage under rapidly changing environment (Chen, 2009).Innovation is a core element of strategy and performance studies along with knowledge and capabilities (Zehir, 2016).We used seven point Likert scale in this study to measure innovation performance from strongly disagree = 1 to strongly agree = 7. Independent variable 4.2.1.Absorptive capacity Cohen and Levinthal (1989) mentioned that ACAP is the ability of learning from environment by multiple processes, for instance, attainment, absorption and exploitation.Cohen (1990) redefined ACAP Construct as the capability of organization to value, absorb and routine knowledge that is absorbed from external sources.According to this new approach ACAP is considered as a by-product not only for activities of R&D, but also of the breath of the organization's knowledge base, learning experience, collective language, the existence of cross-functional interfaces plus problem solving and mental models, capability of the organizations' members.In this study to measure the dimensions of ACAP we used seven Likert scale from strongly disagree = 1 to strongly agree = 7. Dominant Logic As mentioned earlier that dominant logic is the DNA of the organization (Prahalad, 2004, p. 172) without dominant logic it is impossible to manage the tangible and intangible resources of the organizations.Moreover, DL deals with the resource allocation and making important decisions (Prahalad, 2004, p. 172) that may affect the firms' performance.In emerging economies, dominant logic plays an important role, where limited resources are available in the environment (Amit & Schoemaker, 1993;Barney, 1991).To measure the dimensions of dominant logic we used seven point Likert scale from strongly disagree = 1 to strongly agree = 7. Results and Discussion Table 1 shows the mean, standard deviation and correlations between variables.Table 1 describes that absorptive capacity is positively and significantly related to dominant logic (r = .541,p = .001),and positively related to innovation performance (r = .716,p = .001).In addition, dominant logic is positively and significantly related to innovation performance (r = .735,p = .001).These results show that absorptive capacity and dominant logic are the important predictors of innovation performance. Furthermore, regression analysis is performed to check the strength of relationship among variables.A significant relationship was found to check the relationship between ACAP, DL and IP (F (2, 292) = 308.85,p = .000),with R 2 of .679.These results prove hypotheses 1,2 and 3 in which absorptive capacity has positive relationship with dominant logic and innovation performance, and dominant logic has positive relationship with innovation performance. Conclusion One key aspect in which the public sector organizations need to be more focused on creativity and innovation is the fact that the private enterprises have been rigorously innovating in terms of products and systems, whereas the public sector or rather the government organizations have not been developing at the same pace as their counterparts.This makes it difficult for organizations to survive in such a dynamic fast paced market place.The relationship proven in this study further attests to the fact that innovation without knowledge and leadership is impossible.The focus in this study was on leadership and knowledge, and their impacts on innovation performance.In the past literature absorptive capacity, dominant logic and innovation performance were discussed in different organizations and separately.This study combined these variables to check the public organizations performance.We have found a positive relationship among these variables which have indicated that without leadership and knowledge it is impossible for public organizations to improve their performance.This paper has also provided an insight regarding knowledge absorption for top management in public sector organizations and how the top management could use their dynamic capabilities to improve performance and foster an innovative environment. Table 3 Regression analysis between ACAP, DL and IP Table 4 Coefficients of ACAP, DL and IP a. Dependent Variable: IP1	2018-12-02T20:22:18.552Z	2017-01-01T00:00:00.000	{ "year": 2017, "sha1": "9069cf434537b916a0c6194c99fbbec03f28fa18", "oa_license": "CCBY", "oa_url": "https://doi.org/10.5267/j.msl.2017.3.004", "oa_status": "GOLD", "pdf_src": "ScienceParseMerged", "pdf_hash": "9069cf434537b916a0c6194c99fbbec03f28fa18", "s2fieldsofstudy": [ "Business" ], "extfieldsofstudy": [ "Business" ] }
249056161	pes2o/s2orc	v3-fos-license	Concurrent Lead and Noise Exposure Effects on Testicular Tissue of Rat: An Experimental Study Objective: Environmental stressors such as physical and chemical factors can have a destructive testicular function. The present study aimed to determine the effect of concomitant exposure to lead and noise on testicular tissue in male rats. Materials and methods : Twenty male Wistar rats (250-300 g, 12-13wk) were divided into four groups (n=5/each): 1) Control group, 2) Lead group treated with 4 mg/kg lead acetate by gavage for 30 days, 3) Noise group exposed to 4 kHz octave band at 105 dBA for eight hrs./ day for 30 days, and 4) the exposed group to lead plus noise concurrently. The testes' weight was measured, and testes abnormalities were assessed after staining with Hematoxylin and Eosin. Results: The results showed that the weight of testes in experimental groups was significantly decreased compared with the control group (p<0.0001). Also indicated edema, degeneration and necrotic cell debris in the lumen, congestion and atrophy of seminiferous tubule in rat testes tissue due to sub-acute exposure to lead and noise. Conclusion: Exposure to 105 dB and lead can cause damage to the seminiferous tubules, intubation edema, and testicular weight loss compared to control. We also found that simultaneous exposure to noise and lead could have more detrimental effects on testicular histology and weight than others. physical factors in the workplace, and according to the Ministry of Health's Center for Health and Workplace, more than 2 million workers in Iran are exposed to harmful occupational noise. Noise stress has adverse effects on the human body, such as cardiovascular disorders, hearing loss, reduced mental efficiency, and can affect the hormonal system (4,5). Some studies by evaluating testosterone and spermatogenesis level, Luteinizing hormone (LH) and follicle stimulating hormone (FSH), sperm count, morphology, viability have shown that noise-induced stress could affect the structure of testicular tissue (2,6,7). Long-term exposure to noise-induced stress seems to have adverse effects on male fertility based on increased apoptosis due to pathogenesis stress (noise and lead) and suppression of spermatogenesis kinetics (7,8). Among toxic chemicals, metals are considered important environmental toxins due to their biodegradability and non-degradability properties and are harmful to environmental systems (9). Lead is one of the most persistent environmental pollutants, and due to its everyday use in various industrial products, it is one of the essential heavy metals and a serious occupational hazard worldwide (10,11). Humans can be exposed to metal by eating contaminated food, water, or soil (12). Exposure to lead causes a wide range of physiological, biochemical, and behavioral disorders and testicular dysfunction (for example, affects the testes' weight, sperm concentration, morphology, and motility) (13)(14)(15)(16). Production of reactive oxygen species (ROS) is a probable mechanism that was considered for the detrimental effect of lead. Lead can accumulate in organs and change their biological activity via oxidative stress. The testis is very sensitive to oxidative stress due to its unsaturated fatty acids (17). Noise and lead appear to cause oxidative stress in various tissues, including the testes, by disturbing the balance of the prooxidant/antioxidant system (18,19). Given that so far, few studies conducted to investigate the cumulative effects of noise and lead on testicular tissue, and most studies have examined chronic and sub-chronic effects, this study aimed to investigate the Concurrent effect of sub-acute exposure to noise and lead on testicular tissue in male rats. Materials and methods Study design and animals: Twenty healthy male Wistar rats (weight 250-300 gr, 12-13 wk) were used for the experiment and obtained from the Experimental and Comparative studies center, Iran University of Medical Science, Tehran, Iran. The animals were housed in the temperature and lightcontrolled room (23±1 ˚C, and 12-hr dark/ light cycle) and 40-50 % relative humidity with free access to standard food and water and accustomed for one week before the study. Lead acetate powder was purchased from MERCK Co. Germany (CAS: 107375). Rats were randomly divided into four groups (n=5 in each group) as below: Group 1: control; the animals received anything. Group 2: rats were treated with lead acetate (4 mg/kg; lead acetate solution made by distilled water) by gavage for 30 days (20). Group 3: rats were exposed to a 4 kHz octave band at 105 dBA for eight hrs/ day (occupational exposure) for 30 days. Group 4: the animals were exposed to lead by gavage (4 mg/kg) plus noise (4 kHz octave band at 105 dBA for eight hrs/ day) for 30 days. Noise exposure: Noise exposure room was applied in dimensions of 808095, and ten rats could concurrently live inside it comfortably. The chamber temperature was adjusted to 22 o C using a temperature control device. Noise 105 ± 1 dB with 4 kHz frequency generated by the signal software and played by cool edit pro V.21 and animals in groups 3 and 4 were exposed to this noise for eight hrs/day, six days/wk, for four next wks (groups 1 and 2 were kept in similar conditions with a difference that the noise did not exceed 60 dB). The noise was amplified and played by two speakers (Venous-SU132) located in the room's ceiling at equal distances four holes, placed on the center of sidewalls to monitor the chamber condition (21). The noise level was measured using a sound level meter model CEL-400 through holes made on every side of the chamber. Sample preparation: Rats were sacrificed after 30 days. The weight of testes was measured by digital balance (Shimadzu, Japan, LIBROR AEL-200), dissected the left testes tissue, and fixed with 10% formalin. Samples were then provided using a graded ethanol series and embedded in paraffin. Sections Statistical analysis: All the statistical analyzes were carried out using SPSS (Version 22.0, SPSS Inc., Chicago, IL, USA). Kolmogorov-Smirnov test was used to assay the normal distribution of data. One-way ANOVA and the Tukey test were applied for comparisons among the groups. P-value ˂0.05 was considered significant. The results were presented as mean± SD. Results There were considerable differences in the weight of testes between all experimental groups. The findings of the present study indicated the weight of testes in the noise plus lead group (p-value< 0.0001) was significantly less than in the lead and noise groups ( Figure 1). Figure 1: Comparison of the effect of sub-acute lead plus noise exposure on testis weight (N= 5 each group). Data presented as mean±SD. Oneway ANOVA and the Tukey test were applied. *p-value < 0.0001 Microscopic Description: Figure 2 reveals the histopathological changes of the testes in experimental groups. The control group is shown to have a well-preserved lobules structure in which each lobule contains four seminiferous tubules embedded in a connective tissue stroma (A). There are mild to moderate congestion, intertubular edema, and a few degenerations and necrotic cell, debris in the lumen (B & C) in the lead, and noise groups. Also, there are extensive damages to testicular tissue, such as moderate or severe congestion, atrophy of seminiferous tubule necrotic cells debris in the lumen, and intertubular edema (D) in the Noise plus Lead group (Table 1). Edema, degeneration and necrotic cell debris in the lumen Noise Edema, degeneration and necrotic cell debris in the lumen Lead plus noise Severe congestion, atrophy of seminiferous tubule, necrotic cell debris in lumen and edema These pathologic changes were diagnosed by hyperemia in blood vessels (congestion), increased distance between tubules (intertubular edema), structural abnormality (degenerations), cells without nuclei (necrotic cell), fragments of a necrotic cell (debris), and thinning of seminiferous tubule walls (atrophy). Discussion The testis is a dual organ and the main reproductive organ in men with endocrine function (22,23). Environmental factors such as physical and chemical agents can have a destructive testicular function (2, 3). Among environmental pollutants, noise is one of the most common natural pollutants and an inevitable factor at home and the workplace (24). Several studies have reported that vocal stress can have structural and functional adverse effects on testicular tissue (25,26). Due to the industrialization and development of urbanization, there is a growing concern about the adverse effects of metals on the male reproductive system (27). Lead is an abundant toxic metal in the environment (28), and exposure to it has increased due to its widespread use in industry, cosmetics, and medicine (29). Lead has no recognizable biologically beneficial role, but on the contrary, its destructive effects on humans and animals have been proven by several researchers (30,31). Even at low levels, exposure to lead can affect male fertility (8). Reports suggest that lead may damage testicular tissue by producing free radicals and thus increasing lipid peroxidation (32). This study investigated the simultaneous effects of noise and lead exposure on the testis tissues. The present study results showed that testis weight in the noise exposure group was significantly reduced compared to the control group (Figure 1), which is in line with the previous studies (33,34). It is determined that androgen hormones regulate the weight of genital organs, and it seems noise-induced stress by impaired hormones regulation could cause loss of weight (2,7). There are conflicting results about the effects of lead on testicular weight. One study reported a significant increase in testicular weight (35); on the other hand, the other studies reported a significant decrease (36,37) or no change in weight (38). This difference may be due to differences in the exposure pathway, lead dose, age, etc. According to the results of this study, exposure to lead acetate caused a significant reduction in testicular weight in the lead exposure group compared to the control group (Figure 1), the results of which are in line with the study by Dorostghoal et al. (39). Also, in this study, we found that testicular weight in the group exposed to noise plus lead has a significant decrease compared to the other groups ( Figure 1). One of the probable reasons for more decreasing testis weight in this group is the additive effect of lead and noise. Usually, antioxidant defense systems in reproductive tissues prevent oxidative damage in spermatogenesis (38). Studies have shown that noise and lead increase the production of free radicals by reducing or inhibiting the activity of antioxidant enzymes, thereby destroying the layers of spermproducing cells (4,40). Histological examinations of testicular specimens are shown in Figure 2 and Table 1. The control group (A) shows the typical organization of the seminiferous tubules associated with arranged germinal lineage and the spermatids inside the tubes. Spermatogenic cells are characterized by dark nuclei, while serotonin cells are characterized by pale cells. Connective tissue and interstitial cells are also commonly seen between tubes. Some pathological changes are observed in the experimental groups. Histological assessments show that exposed groups lack natural structure in seminiferous tubules. The thickness of the allogeneic tonic and the diameter of the seminiferous tubules decreased significantly in the study groups. Interstitial connective tissue and the number of Leydig cells are significantly reduced, indicating severe tissue damage (arrows). In agreement with the present study results, two studies have shown that exposure to lead causes damages to the seminiferous tubules and interstitial connective tissue (36,41). In addition, a study on http://jfrh.tums.ac.ir Journal of Family and Reproductive Health  lead-exposed mice showed degeneration of the seminiferous tubules (42). In line with this study, a study also showed that the number of Leydig cells decreased in lead-exposed mice (36). In the other study Damage to the seminiferous tubules was observed in mice exposed to sound, which is in line with the present study (43). Another study also showed damage to the seminiferous tubules and interstitial space in fetuses of mice exposed to vocal stress (7). Other degenerative effects on sperm tubes include the destruction of the germination layer in the experimental groups (shown with a star). In addition, moderate to severe damage, deficiency, or reduction in spermatid count were observed in some tubes in group D (shown with a circle). Atrophy of the seminiferous tubules is also seen in exposure to noise and lead (D) (white star). One study showed that exposure to lead for 5 and 10 wk causes irregularity, degeneration, and atrophy in most seminiferous tubules, increases interstitial space and reduces Leydig cells, which agrees with the findings of the present study (37). Also, in line with the results of this study, the other study showed that exposure to 115 dB noise causes damage to the seminiferous tubules, atrophy of the germinal epithelium, and increased connective tissue between the tubes (2). There are some limitations in our study such as failure of the noise generation system, animal death, shortage of examination equipment and financial support, to examine the male hormones, sperm count, and motility. Conclusion That said, exposure to 105 dB and lead could cause damage to the seminiferous tubules, intubation edema, and testicular weight loss compared to controls. We also found that simultaneous exposure to noise and lead could have more detrimental effects on testicular histology and weight than other groups. These findings could be helpful for people who are exposed to noise and lead at work. It is recommended to use different sound levels and doses of lead in future studies. Also, examination of tissue changes in the fetus and female reproductive organs, examining the male hormones, sperm count, and motility is suggested. Conflict of Interests Authors have no conflict of interests.	2022-05-26T15:10:05.882Z	2022-05-24T00:00:00.000	{ "year": 2022, "sha1": "bb36738e8232e12dbde70005d8f9df7079883403", "oa_license": "CCBYNC", "oa_url": "https://doi.org/10.18502/jfrh.v16i2.9483", "oa_status": "GOLD", "pdf_src": "MergedPDFExtraction", "pdf_hash": "b4abb30048e065f387b385fd39c4e1e331d962fd", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
43900079	pes2o/s2orc	v3-fos-license	Bronchoconstrictor effect of histamine in cats. It has been reported that bronchoconstrictor effect of a certain spasmogen is modified by release of catecholamines. According to McCulloch et al. (1), these catecholamines are released in guinea pigs by histamine as a result of a sympathetic bronchodilator reflex. Collier et al. (2) assumed that bradykinin or angiotensin directly liberates catecholamines from the adrenal glands of guinea pigs. In cats, Colebatch (3) proposed similar mechanism to that of Collier et al. (2) on the constriction of alveolar ducts produced by histamine. Thus it seems that the source of catecholamines, which affect the bronchoconstriction, varies with sort of spasmogens or with species of animals. The present experiments were undertaken to examine the role of autonomic nervous system or humoral regulation involved in the bronchoconstriction induced in cats by histamine, which was administered intravenously (4, 5). Besides histamine, acetylcholine and serotonin were also used as bronchoconstrictor agents. In addition, bronchoconstriction induced in guinea pigs by histamine was studied and the results were compared with that of cats. T he present experiments were undertaken to examine the role of autonomic nervous system or humoral regulation involved in the bronchoconstriction induced in cats by his tamine, which was administered intravenously (4,5). Besides histamine , acetylcholine and serotonin were also used as bronchoconstrictor agents. In addition, bronchocon striction induced in guinea pigs by histamine was studied and the results were compared with that of cats. MATERIALS AND METHODS Bronchoconstrictions in cats Adults cats of either sex, weighing 1.8 to 4 kg, were anesthetized with sodium pento barbital (35 mg/kg, i.p.). Artificial ventilation was carried out with a constant volume respiratory pump through a tracheal cannula. The ventilation rate was 26 strokes/min and the stroke volume was about 15 ml/kg. Animals were then immobilized with intra venous injection of gallamine triethiodide (5 mg/kg). Intratracheal pressure , the increase of which denotes the bronchoconstriction, was measured by means of a pressure transducer connected to the side arm of tracheal cannula. The bronchoconstriction was induced by intravenous injection of histamine, acetylcholine or serotonin . Blood pressure was recorded from the femoral artery by means of a pressure trans ducer. Heart rate was measured by cardiotachography , triggered by the arterial pulse. All responses were recorded simultaneously on the ink-writing oscillograph. Pithed and adrenalectomized cats Pithed cats were prepared according to the method of Pulchino and Trenderenberg (6). Either in normal or in pithed cat, bilateral adrenalectomy was performed during the experiments under artificial respiration. Contraction of nictitating incinbrane The contraction of nictitating membrane in pithed cats was measured by means of a strain gauge transducer. The unilateral common carotid artery supplying the nictitat ing membrane was left intact. The initial tension of 2 g was applied to the membrane. Blood pressure, heart rate and intratracheal pressure were recorded simultaneously. Bronchoconstr'ictions i11 guinea pigs Guinea pigs (about 300 g), anesthetized with intraperitoneal injection of urethane 1.25 g/kg, were artificially ventilated with a respiratory pump delivering 72 strokes/min at a constant stroke volume of 2-4 ml and then immobilized with gallamine triethiodide Intratracheal pressure was measured with a pressure transducer by the same pro cedure mentioned above. Bilateral adrenalectomy was carried out during the experi ments. In some cases, guinea pigs were pithed with a wire, passed through the orbital fossa. In each experiment, at least five animals were used. 2) Effect of propranolol As shown in Fig. 1, the histamine induced bronchoconstriction was poten tiated by pretreatment with propranolol. The potentiation was more pronounced in the duration than in the maximum response to histamine. It was also ob served that propranolol had no influence on the basal intratracheal pressure without histamine. 4 shows the effect of acute bilateral adrenalectomy on the response to histamine. After adrenalectomy, the duration of histamine-induced bronchoconstriction was con spicuously prolonged, whereas the maximum response to histamine was little affected. Unlike in intact or pithed cat, propranolol did not potentiate the response to histamine in adrenalectomized cats (Table 1). These results suggest that catecholamines, released by histamine from the adrenal medulla, counteract the bronchoconstriction induced by histamine. The possibility that histamine would liberate catecholamines from the adrenal medulla was also examined in the following experiments on heart rate, blood pressure and nictitating membrane. Positive chronotropic action of histamine The intravenous administration of histamine caused positive chronotropic effect in the anesthetized cat. As shown in Fig. 3 and Table 2, the increase in heart rate produced by histamine was not affected by pithing, whereas it was inhibited by bilateral adrena lectomy. Propranolol markedly depressed the positive chronotropic action of histamine in both pithed and adrenalectomized cats ( Table 2). When bilateral adrenalectomy was carried out in pithed cats, the positive chronotropic action of histamine was completely inhibited (Fig. 6). 3. Secondary pressor affect of histamine As shown it Fig. 3, the secondary pressor effect of histamine became distinct by pith ing the cat. This pressor response was not antagonized by pretreatment with propranolol. On the other hand, phentolamine or adrenalectomy caused inhibition of the secondary pressor effect (Figs. 5, 6). Contraction of n ctitatina membrane The contraction of nictitating membrane produced by histamine or adrenaline was inhibited by pretreatment with phentolamine (Fig. 5). As shown in Fig. 5, phentolamine did not influence the effects of histamine on the heart rate and intratracheal pressure. Just as the case of pressor response or positive chronotropic effect, the contraction of nictitating membrane of pithed cat induced by histamine was markedly suppressed by adrenalectomy (Fig. 6). When propranolol was administered to the pithed guinea pig, no potentiation was observed (Fig. 9). In addition, it was found that the bronchoconstrictor effect of histamine was intensified by pithing. In the present experiment, it was demonstrated that the histamine-induced bronchocon• striction, which was measured as an increase in intratracheal pressure, and, which involve! the reduction of lung compliance as well as the decrease in airway conductance, was no affected by pithing the cat or by intravenous administration of atropine. In addition it was found that hexamethonium had no influence on the bronchoconstriction inducec by histamine. According to Trendelenberg (8), the pressor response of cats to histaminf was not abolished by hexamethonium. Staszewska-Barczak et al. (9) showed that hexa• methonium did not inhibit the liberation of catecholamines by histamine from adrena medulla of cats. From these facts, it may be concluded that the effect of vagus or sym. Acetylcholinc and serotonin-inchrced bronclroconstrictions in cats pathetic nerve was not involved in the bronchoconstrictor response to histamine in cats In cats, histamine-induced bronchoconstriction was potentiated by pretreatment witl propranolol. This potentiation was more pronounced in the duration than in the maximurr resnonse to histamine. Like propranolol. bilateral adrenalectomy strongly prolongec the duration of histamine-induced bronchoconstriction. When administered to adrena lectomized cats, propranolol could not potentiate the constriction produced by histamine. On the contrary, pithing the cat had no influence on the response to histamine, while pro pranolol considerably prolonged the duration of the constriction in pithed cats. These results may indicate that histamine directly releases catecholamines from ad renal medulla and that these catecholamines tend to reduce the bronchoconstrictor action of histamine. The conclusion supports the view of Colebatch (3). In this connection, it has been reported by Stazewska-Barczak et al. (9) and Trende lenberg (10) that histamine liberated adrenaline from adrenal medulla in cats. In the present study, it was found that the contraction of nictitating membrane and secondary pressor response induced by intravenous injection of histamine, at a dose which caused bronchoconstrictor effect, were inhibited or abolished by phentolamine or bilateral adre nalectomy in pithed cats. The increase in heart rate produced by histamine was also in hibited by adrenalectomy after pithing. These facts suggest that catecholamines released by histamine from adrenal medulla are involved in pharmacological effects. Unlike the effect of histamine, acetylcholine and serotonin-induced bronchocon strictions in cats were not affected by adrenalectomy. Propranolol, administered after adrenalectomy, produced hardly any effect on these constrictions. Therefore, it seems likely that in cats humoral effect is not concerned in bronchospasm induced by acetyl choline or serotonin. In guinea pigs, propranolol potentiated the histamine-induced bronchoconstriction especially in its maximum response. The potentiation of propranolol was still observed in adrenalectomized guinea pigs but not occurred in pithed preparations. It was further observed that adrenalectomy had little influence on the bronchoconstrictor response of histamine. In isolated tracheal preparation of guinea pigs, propranolol did not potenti ate the histamine-induced constriction. These results are basically the same as that of McCulloch et al. (I) who concluded that propranolol antagonized bronchodilator effect of catecholamines which were released from sympathetic nerve endings by homeostatic reflex. To summarize the data obtained so far, it can be said that histamine administered intravenously liberates catecholamines, which tend to reduce the bronchoconstrictor action of histamine. These catecholamines are released directly from adrenal medulla in cats, while they are liberated reflexly from sympathetic nerve ending in guinea pigs. On the other hand, bronchoconstriction induced by acetylcholine or serotonin seems not to be modified by endogenous catecholamines in cats. SUMMARY The hronchoconstrictor effect of histamine in cats was studied and compared with that in guinea pigs. The actions of acetylcholine and serotonin were also examined in cats. 1. The bronchoconstriction induced by histamine is modified and reduced by release of catecholamines.	2018-04-03T05:41:59.715Z	1971-08-01T00:00:00.000	{ "year": 1971, "sha1": "5f3ebd54b4f27ce514ceb517312d1264a00031ea", "oa_license": null, "oa_url": "https://www.jstage.jst.go.jp/article/jphs1951/21/4/21_4_467/_pdf", "oa_status": "GOLD", "pdf_src": "Adhoc", "pdf_hash": "a726c30ad10f287eef2c9af851f6ca9f23c9c9ba", "s2fieldsofstudy": [ "Medicine", "Chemistry", "Biology" ], "extfieldsofstudy": [ "Medicine", "Chemistry" ] }
87283478	pes2o/s2orc	v3-fos-license	Studies of coprophilous ascomycetes in Kenya . Podospora species from wildlife dung Moist chamber cultures were made from wildlife dung obtained from national parks in Kenya. Ten dung types produced 28 specimens of Podospora. Five species, Podospora anserina, P. argentinensis, P. australis, P. communis and P. minor are described using their morphological features. Podospora minor seems to be a rare species and is recorded for the first time in Kenya. Podospora communis, P. anserina and P. australis are the most common species on dung types examined. Introduction A survey of coprophilous fungi in Kenya that we started three years ago has so far produced four publications (Mungai et al. 2012a(Mungai et al. , 2012b(Mungai et al. , 2012c(Mungai et al. , 2012d) ) on some interesting ascomycetes groups.We have examined Ascobolus Pers., Saccobolus Boud., Schizothecium Corda., and an assortment of less common Sordariales, followed now by this work on Podospora Ces., one of the commonest genera on dung (Lundqvist 1972, Doveri 2004a, 2008, Bell 2005).Podospora is predominantly coprophilous, cosmopolitan and normally perithecioid.This genus is strikingly similar to Schizothecium Corda, Arnium Nitschke ex G. Winter, Zygopleurage Boedijn and Cercophora Fuckel.It has been treated in great detail by Cain (1934), Mirza & Cain (1969), Lundqvist (1972), Bell & Mahoney (1995, 1997) Bell (2005) and Doveri (2004aDoveri ( , 2008)).Objectives of this survey are to: i) study the taxonomy of Podospora from Kenyan wildlife dung, ii) document the ecology and biodiversity of Podospora species on different wildlife dung types and, iii) create awareness on the importance of dung fungi in biodiversity conservation and management. In this study, we have used morphological characters to identify the Podospora species isolated from wildlife dung in Kenya. Material examined -14 isolates from dung incubated between 12 and 64 days -KENYA, Nairobi National Park, Nairobi Province, UTM37 02537515 M 9849130, Notes -The Kenyan collections have typical features for the species and compare well with descriptions of same species in previous examinations (Lundqvist 1972, Doveri 2004, Bell 2005).Podospora communis belongs to the sect.Malinvernia noted for having clavate or dumb-bell shaped immature ascospores with complex gelatinous equipment at maturity, and peridium lacking a middle gelatinous layer.The main diagnostic characters for P. communis include its long, cylindrical pedicel equipped with four short, claw-like, independent gelatinous caudae on the lower end and the presence of four similar caudae on the apex of the spore head (dark cell) (Lundqvist 1972).However, our Kenyan collections have some slight variance with the Australian collection which had three basal caudae (Bell 2005).P. spinulosa R.S. Khan & Cain, in the same section, is differentiated from P. communis by having short spinules on perithecial neck and more caudae at the base of the pedicel; P. hyalopilosa (R. Stratton) Cain has hyaline perithecial neck hairs and a pedicel with only a single apical cauda; P. multicaudiculata Cailleux has a slightly hairy neck, shorter asci and multiple but simpler structured caudae; P. austrohemisphaerica N. Lundq.has rigid neck hairs, larger spores, multiple caudae, usually four at each pole, additonal shorter caudae at base of pedicel and a sheath covering the pedicel and spore head; the neck of P. deropodalis R.S. Khan & Cain is covered with short spinules, it has smaller spores but with longer pecidels and variable number of upper caudae and fewer lower caudae; P. alexandri Doveri is similar but can be differentiated by its larger ascospores and the inconspicuous caudae when mounted in water (Doveri 2004a(Doveri , 2004b(Doveri , 2008)).Podospora communis is the most widespread and commonest species occurring on several wildlife dung types in Kenya (Table 1).Ellis & Everh., Amer. Nat. 31: 341, 1897.(Figs.7A-F, 8A-D, 9A-H Notes -Podospora minor is in the section Podospora.The ascospores of P. minor are morphologically similar to those of P. fimiseda (Ces.& De Not.) Niessl.belonging to the same section but are slightly smaller.The characters of P. minor seem to be intermediate between P. appendiculata (Auswer.)Niessl and P. fimiseda.This Kenyan collection has broader asci and larger ascospores in comparison to the specimen examined by Mirza & Cain (1969).This species appears to be rare in wildlife dung and we make the first record of P. minor in Kenya. Type, structure, texture, moisture contents and age of dung collected are important variables for coprophilous Podospora.The structure and composition of dung varies according to the animal species voiding it.The dung of African elephant is usually voided as one or two very coarse large randomly dispersed piles; black rhinoceros and hippopotamus dung are also large, less coarse and look very similar.However, black rhinoceros dung was always voided at one point. Hippopotamus is territorial and thus usually marks its territory by splashing dung on vegetation within the territory and therefore its dung is always a mixture of several days voided biomass.The texture of Cape buffalo dung was the finest for the large mammals sampled.The rest of the wild animals sampled void dung in form of pellets of various shapes and sizes.Zebra has the largest pellets followed by giraffe while dikdik has the smallest dung pellets sampled. Moisture content of the dung has a great influence on Podospora sporulation.Extremely wet dung samples such as those from the very rainy mountainous Aberdares National Park needs thorough air-drying after sampling.This was noted to have some influence on Podospora species occurrence. The time the dung is exposed since voiding and the time it is sampled is also a crucial variable.Very old and dry dung does not yield as many isolates as freshly voided dung. Dung sampled from grassland habitats produced the most Podospora isolates in this study (Table 1). Conclusion Coprophilous Podospora species diversity in Kenya is high and seems to be influenced by a number of factors both biotic and abiotic.The correlation observed between the structure of dung and type on one hand and Podospora on the other hand need further elucidation.This study has helped create awareness on the need to understand dung fungi and its ecological roles.It is hoped that there will be a shift in policy to embrace fauna, Table 1 Occurrence and distribution of Podospora spp. in study area	2018-12-29T10:17:51.183Z	2012-01-01T00:00:00.000	{ "year": 2012, "sha1": "89dac4d25fd515c5ba1755fe405b39633fe5bfc4", "oa_license": "CCBYNCSA", "oa_url": "https://doi.org/10.5943/mycosphere/3/6/12", "oa_status": "GOLD", "pdf_src": "ScienceParseMerged", "pdf_hash": "89dac4d25fd515c5ba1755fe405b39633fe5bfc4", "s2fieldsofstudy": [ "Environmental Science", "Biology" ], "extfieldsofstudy": [ "Biology" ] }
249859526	pes2o/s2orc	v3-fos-license	Exploration of environmental noise in Saharan oases on the basis of urban configurations: City of Biskra datasets The scope of this dataset is to explore the acoustic environment on the basis of spatial configurations. To fulfill this objective, two ranges of data were involved. Firstly, urban acoustic data was obtained by setting up 240 stations of measurements based on equivalent continuous A-weighted sound pressure level, across 16 urban zones in Biskra City, Algeria, during working hours. The data was analyzed using geostatistical and interpolation models on a Geographic Information System platform in order to enhance the observations and examine their distribution within the urban area. Secondly, spatial configuration data is based on two indicators: Normalized Angular Integration and Normalized Angular choice. Integration refers to the degree of accessibility ``to-movement'', whereas Choice values depict the urban route hierarchy, or the ``through-movement'' potential. The data includes different values of the global and local scale of the urban structure using several metric radii of 400m, 800m, 1200m, 1600m, 2000m, 2400m and 3200m. A correlation scheme is required in order to validate the effectiveness of this methodological approach. The dataset serves as an insightful reference for further inquiry and policymakers in dealing with sustainable concerns related to soundscape, noise pollution, and urban planning. a b s t r a c t The scope of this dataset is to explore the acoustic environment on the basis of spatial configurations. To fulfill this objective, two ranges of data were involved. Firstly, urban acoustic data was obtained by setting up 240 stations of measurements based on equivalent continuous A-weighted sound pressure level, across 16 urban zones in Biskra City, Algeria, during working hours. The data was analyzed using geostatistical and interpolation models on a Geographic Information System platform in order to enhance the observations and examine their distribution within the urban area. Secondly, spatial configuration data is based on two indicators: Normalized Angular Integration and Normalized Angular choice. Integration refers to the degree of accessibility "to-movement", whereas Choice values depict the urban route hierarchy, or the "through-movement" potential. The data includes different values of the global and local scale of the urban structure using several metric radii of 400m, 80 0m, 120 0m, 160 0m, 20 0 0m, 240 0m and 3200m. A correlation scheme is required in order to validate the effectiveness of this methodological approach. The dataset serves as an insightful reference for further inquiry and policymakers in dealing with sustainable concerns related to soundscape, noise pollution, and urban planning. © 2022 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ) Table Subject Environmental Science Environmental Engineering Specific subject area Urban Acoustics, Spatial analysis Type of data Value of the Data • The systematic review of these data yields a broad sense assessment of the relationship between the acoustic environment and the functional state of the urban system. • A starting point for synchronized or diachronic research in the same Saharan urban area, as well as for comparisons with other contexts. • These data may be used by researchers in a range of domains, including soundscape, noise pollution, urban planning, computational and applied sciences, social and psychological sciences, inhabitants' health and well-being, and sustainability challenges. • Due to the lack of data in such regions, this article assists policymakers throughout the urban planning process and acts as a diagnostic aid for soundscape management. Data Description The urban soundscape is a complex and multifaceted topic that has been widely addressed across a broad array of disciplines using different methods and multidimensional data [1 , 2] . However, only a few studies were based on urban structure components related to people's movement and activities. This article provides in-site measurements of the acoustic environment and spatial configuration analysis data series. The meteorological conditions during data collection are illustrated in Table 1 which highlights different indicators such as temperature, wind speed, precipitation, humidity and cloud cover ratio. Table 2 summarizes the contextual and physical characteristics of the case study region's urban sectors, including zone identifiers and names for 16 urban zones, land use, building typologies, and area. The raw data shared in the repository (in particular in File: A. Level Sound Measured Data.csv) contains for each measurement the following information: sector ID, sector name, stations ID, values of equivalent continuous A-weighted sound pressure level (LeqA), date, time and the geographic coordinates (Longitude and Latitude) of each measurement station. Fig. 1 depicts the propagation of measured data according to geospatial interpolation methods such as Inverse Distance Weighted Gaussian, IDW Exponential, IDW k2, Ordinary Kriging and Universal Kriging. A statistical description and validation of the preceding interpolation models are highlighted in Table 3 . Fig. 2 represents zonal statistics for the validated model IDW k2 raster, as well as some useful statistics to describe each zone, based on the clustered data spread. The spatial database extracted from depthMapX via angular segment analysis of the entire region is stored in a second file (File: B. Extracted DATA from depthMapX.csv), that further includes global measures of Normalized Angular CHoice (NACH) and Normalized Angular INtegration (NAIN), along with local measures at various metric radii of 40 0m, 80 0m, 120 0m, 160 0m, Table 4 includes a validation test between the observed sound-level data and the spatial database at global and local scales. Detailed associated observations are shared in the third file (File: C. Measurements associated to ASA.csv) using the ID and geolocation fields. Figs. 7 and 8 provide visual representations of modeled data explored by NACH and NAIN, respectively, with the same color legend display for both datasets. Experimental Design, Materials and Methods The methodological approach is based on two main dimensions: the acoustic environment and space syntax analysis. The acoustic data was collected by establishing 240 stations of measurements using a calibrated sound level meter (Model SL-586P of Merit-mi Brand). The instrument was fixed at one-meter level from the ground for a duration of 5 minutes for each station location, and the sound equivalent continuous weighted A levels (LeqA) were acquired. The temporal trends of the LeqA parameter for each station location were acquired with a frequency of 2 data per second, obtaining 600 data for each station location. These measurements were performed in the city of Biskra, Algeria, during the working days in November 2020 and January 2021. The meteorological conditions should be assessed as indicated in Table 1 . The street network map was imported from the open street map database, edited and prepared to explore the urban system that compromises 16 official divisions according to the Municipal Master Plan. Contextual and physical properties are required for a further diagnosis (see Table 2 ). Therefore, the mean values of the performed measurements were modeled on open source software QGIS [3] in order to obtain the acoustic profile for the extent region using Inverse Distance Weighted (IDW) and Kriging interpolation which have proven their potential in acoustic environment studies [4][5][6] . For soundscape profiling, several models are used, including Invert Distance Weighting via different weighted functions (Gaussian, Exponential, and Power 2 functions), as well as Kriging spatial interpolation (Ordinary and Universal) resampling by a B-spline Interpolation ( Fig. 1 ). A validation process of the modeled observations was required to verify data accuracy, as mentioned in Table 3 . The model that accurately represents the acoustic data has been explored by a zonal statistical raster function on QGIS in order to explore data propagation through the city urban system (see Fig. 2 ). Space syntax theory and angular segment analysis were used to analyze the georeferenced map of the city. Two key values were the subject of the second data range whereas, Normalized Angular Integration and Normalized Angular Choice, which represent o the degree of accessibility "to-movement potential", whereas Choice values depict the urban route hierarchy, or the "through-movement potential" [7] . This analysis was applied at global and local scales of the urban system applying various metric radii for NACH and NAIN values [8][9][10][11] . These values were extracted after a simulation run on depthMapX [12] and visualized on QGIS (see Figs. 5 and 6 ). To emphasize the novelty of this methodological approach, which promotes a sense of collecting and pairing the various data presented in this article, a Pearson correlation analysis was established, see Table 4 . The correlation analysis can be explained by graph visualizations on QGIS, as shown in Figs. 5 and 6 . 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.	2022-06-20T15:05:08.951Z	2022-06-01T00:00:00.000	{ "year": 2022, "sha1": "7cb861486b1afa8e86becd5b3cc72c30129c1ad1", "oa_license": "CCBYNCND", "oa_url": "https://doi.org/10.1016/j.dib.2022.108392", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "1f035b30ea197c5fede6362abb131f6bb1abeab9", "s2fieldsofstudy": [], "extfieldsofstudy": [] }
202731507	pes2o/s2orc	v3-fos-license	MicroRNA-133b targets TGFβ receptor I to inhibit TGF-β-induced epithelial-to-mesenchymal transition and metastasis by suppressing the TGF-β/SMAD pathway in breast cancer Breast cancer (BC) is one of the most common types of cancer and the leading cause of cancer-associated mortality among women worldwide. Accumulating evidence indicates that microRNA (miR)-133b inhibits the proliferation and invasion of cancer cells. Considering that transforming growth factor (TGF)-β signaling plays a key role in cellular epithelial-to-mesenchymal transition (EMT) and cancer metastasis, it is crucial to explore the roles and underlying molecular mechanisms of miR-133b in regulating TGF-β-induced EMT during progression of BC. In the present study, an inverse correlation was observed between the expression of miR-133b and TGFβ receptor I (TGFβR1) mRNA in BC cells and tissues. Furthermore, miR-133b expression was found to be decreased in the BC tissues of patients with lymph node metastasis and advanced tumor-node-metastasis stage, while the expression of TGFβR1 was upregulated. Overexpression of miR-133b significantly decreased the expression of TGFβR1, an indispensable receptor of TGF-β/SMAD signaling, and suppressed TGF-β-induced EMT and BC cell invasion in vitro, whereas miR-133b knockdown exerted the opposite effects. Mechanistically, TGFβR1 was verified as a direct target of miR-133b as determined by bioinformatics analysis and a dual-luciferase reporter assay. In addition, small interfering RNA-mediated knockdown of TGFβR1 mimicked the phenotype of miR-133b overexpression in BC cells. Furthermore, miR-133b overexpression suppressed BC cell invasion in vivo. Collectively, the findings of the present study indicated that miR-133b acts as a tumor suppressor, inhibiting TGF-β-induced EMT and metastasis by directly targeting TGFβR1, and suppressing the TGF-β/SMAD pathway. Therefore, miR-133b may be of value as a diagnostic biomarker of BC. Introduction Breast cancer (BC) is the most frequently diagnosed cancer and the second leading cause of cancer-associated mortality among women worldwide (1). It was estimated that there were ~2.1 million newly diagnosed BC cases among women worldwide in 2018 (2). Despite significant improvements in early diagnosis and therapeutic strategies, patients with BC have a poor prognosis (3), which is predominantly due to tumor metastasis (4). Therefore, understanding the mechanisms underlying BC metastasis is crucial for developing novel therapeutic strategies for BC. Epithelial-to-mesenchymal transition (EMT), as a vital process for tumor metastasis, is essential for the transformation of early tumors into aggressive malignancies (5). During EMT, the cells lose their epithelial characteristics, including cell polarity and specialized cell-cell adhesions, and exhibit migratory and invasive behavior (6). During the process of EMT, the expression levels of the epithelial marker E-cadherin are significantly downregulated, while those of mesenchymal cell markers, such as Snai1 and N-cadherin, are upregulated (7)(8)(9). Accumulating evidence has demonstrated that the transforming growth factor (TGF)-β/SMAD signaling pathway plays a key role in EMT (10,11). TGF-β/SMAD signaling is initiated by the binding of the TGF-β1/2 to the TGF-β receptor (TGFβR) II. The ligand-receptor complex then recruits TGFβR1 to activate downstream SMAD2/3. Phosphorylated (p-)SMAD2/3 further forms trimers with SMAD4, which translocate to the nucleus, subsequently leading to the activation of crucial transcription factors regulating EMT (12). Of note, previous studies have indicated that numerous microRNAs (miRNAs/miRs) play pivotal roles in TGF-β/SMAD signaling-mediated EMT and tumor metastasis (13)(14)(15). miRs are a class of endogenous, highly conserved, small non-coding RNAs that are ~22 nucleotides in length, and negatively regulate gene expression by binding to the 3'-untranslated region (UTR) of target mRNAs, suppressing translation and inducing mRNA degradation (16)(17)(18). Recent studies demonstrated that miR-133b mainly acts as a tumor suppressor in numerous types of human cancer (19)(20)(21). In colorectal cancer, it has been reported that patients with low levels of miR-133b expression have a poor prognosis in terms of overall survival and positive lymph node metastasis (22). Overexpression of miR-133b inhibits the clonogenicity and invasion of BC cells (23); however, whether miR-133b affects TGF-β-induced EMT and metastasis of BC cells and the underlying mechanism remain unknown. The aim of the present study was to investigate the association of miR-133b expression with BC metastasis and clinical stage, as well as the correlation between the expression of miR-133b and TGFβR1 in BC, hoping that the findings will provide further insight into the mechanisms by which miRNA regulates TGF-β-induced EMT and BC metastasis, and to determine whether miR-133b may be considered as a potential prognostic predictor and therapeutic target for the clinical diagnosis and treatment of BC. Materials and methods Patients and tissue samples. A total of 78 paired BC and adjacent normal tissues (≥2 cm away from the edge of cancer tissue, with no cancer cells identified via microscopy) were obtained from The First Affiliated Hospital of Nantong University between June 2013 and December 2018. Among the 78 patients enrolled, 42 had BC (mean age 60.9 years; range, 32-77 years), while 36 had metastatic BC (mean age, 66.3 years; range, 48-87 years). Pathological analysis of patients with BC was performed according to the Revised International System for Staging Breast Cancer (24). None of the patients had been treated with chemotherapy, radiotherapy, or any other therapy, prior to surgical intervention. All tissues were immediately frozen in liquid nitrogen following surgery, and were then stored at -80˚C until use. RNA extraction, cDNA synthesis, and reverse transcriptionquantitative polymerase chain reaction (RT-qPCR) analysis. Cells (2x10 6 ) and tissues (200 mg) were employed for total RNA extraction using TRIzol ® reagent (Thermo Fisher Scientific, Inc.), according to the manufacturer's protocols. The concentration and purity of total RNA were detected using a NanoDrop2000 spectrophotometer (NanoDrop Technologies; Thermo Fisher Scientific, Inc.). Pre-designed stem-loop RT primers were employed to synthesize the cDNA of miR-133b and U6 (25), and random primers were used for the cDNA synthesis of TGFβR1. The synthesis of cDNA using reverse transcriptase was performed using the M-MLV First Strand kit (Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Briefly, the reaction conditions were as follows: 25˚C for 10 min, 42˚C for 45 min, and 85˚C for 5 min. qPCR was performed using SYBR Green qPCR SuperMix-UDG kits (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The thermocycling conditions were as follows: Initial preincubation at 95˚C for 10 min, followed by 40 cycles of amplification at 95˚C for 5 sec and at 60˚C for 30 sec. qPCR was performed on an ABI Prism 7500 Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). The primer sequences used for RT and qPCR were pre-synthesized (Genewiz Inc.) and are summarized in Table I. U6 was used for the normalization of miR-133b expression, whereas β-actin served as an endogenous control for TGFβR1 mRNA expression. RT-qPCR analysis was conducted in triplicate; relative expression levels were evaluated using the 2 -ΔΔCq method (26). Establishment of stable cell lines overexpressing miR-133b. To stably overexpress miR-133b (LV-miR-133b) in MDA-MB-231 cells, the miR-133b sequence was cloned into the pLKO.1-purovector (InovoGen Tech. Co.) via the AgeI and EcoRI (Fermentas; Thermo Fisher Scientific, Inc.) sites to generate the pLKO.1-puro miR-133b plasmid. Then, the aforementioned construct or control vector with the packaging plasmids psPAX2 and pMD2.G (Geneseed Biotech) were co-transfected into 293T cells to produce the lentivirus. MDA-MB-231 cells were infected with the lentivirus and selected with 1 µg/ml puromycin (Sigma-Aldrich; Merck KGaA) to obtain stable cell lines (LV-miR-133b). Finally, the stable cell lines were used for in vivo experiments of metastasis. The sequence of the miR-133b segment was 5'-CCG GUC AGA AGA AAG AUG CCC CCU GCU CUG GCU GGU CAA ACG GAA CCA AGU CCG UCU UCC UGA GAG GUU UGG UCC CCU UCA ACC AGC UAC AGC AGG GCU GGC AAU GCC CAG UCC UUG GAG A-3' . RNA oligonucleotides and cell transfection. In the present study, two small interfering (si)RNAs) were employed, which were designed using BLOCK-iT™ RNAi Designer (https://rnaidesigner.thermofisher.com/rnaiexpress/design.do) and synthesized by Shanghai GenePharma Co., Ltd., to specifically target TGFβR1 (si-TGFβR1-1 and si-TGFβR1-2). Scramble siRNA (si-NC) was used as a negative control. miR-133b mimic, inhibitor (anti-miR-133b) and negative controls (NCs) were obtained from Shanghai GenePharma Co., Ltd. MCF-7 and MDA-MB-231 cells were cultured on 6-well plates and transiently transfected using Lipofectamine ® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) when 70-80% confluence was attained, according to the Dual-luciferase reporter assay. The bioinformatic tools TargetScan (http://www.targetscan.org) and miRPathDB (https://mpd.bioinf.uni-sb.de/mirnas.html) were used to predict whether miR-133b targets TGFβR1. The psiCHECK-2 dual-luciferase vector (Promega Corporation) was used to generate a construct containing the TGFβR1 3'-UTR fused to the 3'-end of the luciferase reporter. The wild-type (WT) fragment containing predicted miR-133b target sites (positions 2,161-2,167) and mutant (MUT) fragment were directly synthesized (Genewiz, Inc.) and respectively subcloned into the psiCHECK-2 vector to create corresponding constructs. To investigate the effects of miR-133b on luciferase activity, MCF-7 and MDA-MB-231 cells were inoculated in 24-well plates, and were then co-transfected with WT or MUT 3'-UTR luciferase reporter plasmids, and miR-133b mimics or miR-NC using Lipofectamine 3000. Following transfection for 48 h, the cells were harvested, and the relative luciferase activity was determined using a Dual-Luciferase Reporter Assay System (Promega Corporation) on a TD20/20 Luminometer (Turner Designs) according to the manufacturer's protocols. The relative luciferase activity was presented as the ratio of Renilla/firefly in accordance with the Dual-Luciferase Assay Manual (27). Each experiment was repeated in triplicate. Transwell migration and invasion assays. For the migration assay, 8-µm Transwell chambers (Corning Incorporated) were prepared, whereas the membranes of chambers coated with Matrigel (200 ng/ml; BD Biosciences) were used for the invasion experiments. According to the manufacturer's protocols, at 48 h following transfection, MCF-7 and MDA-MB-231 cells were trypsinized and counted. A total of 5x10 4 cells in 100 µl serum-free RPMI-1640 medium were added to the upper chamber, and 600 µl RPMI-1640 medium supplemented with 20% fetal bovine serum was added to the lower chamber. After 6 h, TGF-β1 (5 ng/ml) was added to the lower chambers. After culturing for 24 h (migration) or 36 h (invasion) in a 5% CO 2 incubator at 37˚C, the upper chamber was removed, and the non-migrating or non-invading cells were wiped away using a cotton swab. The cells in the lower chamber were fixed with 100% methanol for 20 min and stained with 0.1% crystal violet solution for 1 h at room temperature. Cells in five random fields (magnification, x100) were imaged and counted under a light microscope. Experiments were performed in triplicate. In vivo metastasis assays. To assess lung metastasis, 5-week-old female BALB/c nude mice were purchased from the Laboratory Animal Center of Nantong University. The mice were divided into two groups (5 mice per group), including the LV-miR-133b and control (LV-miR-NC) groups. The mice were bred and housed under specific pathogen-free conditions. LV-miR-133b and LV-miR-NC MDA-MB-231 cells (3x10 5 cells/mouse) in PBS were injected into the tail vein of mice. At 8 weeks following the injection of MDA-MB-231 cells, the mice were euthanized; the lungs were collected and fixed in Bouin's solution for metastatic nodule analysis, or fixed with 4% polyoxymethylene for H&E staining. H&E staining. According to the H&E staining protocols (15), the lung tissues were fixed in 4% paraformaldehyde overnight at room temperature, and dehydrated through an ethanol gradient [60, 70, 80, 95, 100 (I) and 100% (II)] for 20 min per concentration. Then, the samples were embedded in paraffin for 40 min at 60˚C and cut into 5-µm sections. The lung tissue sections were dewaxed with xylene (I) and xylene (II) for 10 min, followed by rehydration via an ethanol gradient (100, 95, 80, 70 and 60%) for 5 min per concentration. Subsequently, the tissue sections were washed with distilled water for 2 min, followed by hematoxylin staining for 5 min, and washing in distilled water for 30 sec. Subsequently, the samples were treated with 1% hydrochloric acid ethanol for 10 sec and washed for 15 min. Next, the lung tissue sections were stained with 1% eosin for 2 min and washed for 2 min. The tissue sections were dehydrated via an ethanol gradient (95, 95, 100 and 100%), and then cleared with xylene (I) and xylene (II) for 10 min. The sections were sealed with neutral gum and viewed under a light microscope (Olympus Corporation). Statistical analysis. SPSS 23.0 (IBM Corp.) and GraphPad Prism 7.01 (GraphPad Software, Inc.) were used for all statistical analyses. In clinical samples of BC, differences between two groups were assessed using paired or unpaired Student's t-tests (two-tailed). The association between the expression levels of miR-133b or TGFβR1, and the clinicopathological parameters of patients with BC were analyzed by non-parametric tests (Mann-Whitney U test for two groups, and a Kruskal-Wallis test for multiple groups). The correlation between the miR-133b and TGFβR1 mRNA expression level ratios (T/N) in the BC tissues was analyzed by Spearman's rank correlation analysis. Differences among multiple groups were determined by one-way ANOVA followed by Tukey's post hoc test. Prognostic analysis was performed using the Kaplan-Meier method with log-rank test. The results from tissues are presented as box and whiskers with minimum and maximal values, while other data are expressed as mean ± standard deviation. All data were generated from three independent experiments. P<0.05 was considered to indicate a statistically significant difference. Results miR-133b expression is downregulated and is inversely associated with TGFβR1 expression in BC cells and tissues. In order to investigate the association between miR-133b and TGFβR1, RT-qPCR was performed using four human BC cell lines. As shown in Fig. 1A, miR-133b expression levels were notably decreased in the MCF-7, MDA-MB-231, MDA-MB-453 and MDA-MB-468 cell lines compared with those in the immortalized normal breast epithelial cell line MCF-10A. Additionally, the four BC cell lines exhibited significantly increased TGFβR1 mRNA expression levels (P<0.01; Fig. 1B). To further confirm whether there is an inverse association between miR-133b and TGFβR1 mRNA expression in patients with BC, and whether it affects the development of BC, RT-qPCR was performed. From the analysis of 78 BC tissues and paired non-cancerous tissues, a significant reduction in miR-133b expression (P<0.001; 61/78, 78.2%) and a significant increase in TGFβR1 mRNA expression (P<0.001; 59/78, 75.6%) were detected in BC tissues when compared with non-cancerous breast tissues ( Fig. 1C and D). Furthermore, clinicopathological analysis revealed that miR-133b expression (T/N) was markedly reduced in BC tissues from patients with advanced tumor-node-metastasis (TNM) stage and lymph node metastasis compared with those with earlier TNM stage and no lymph node metastasis (P<0.05; Table II). On the contrary, the mRNA expression levels of TGFβR1 (T/N) were higher in BC tissues from patients with advanced TNM stage and lymph node metastasis compared with those with earlier TNM stage and no lymph node metastasis (P<0.05; Table II). Of note, a significant inverse correlation was observed between miR-133b expression (T/N) and TGFβR1 mRNA expression (T/N) in 78 paired tissues (r=-0.495, P<0.001; Fig. 1E). Additionally, in order to verify these findings, datasets containing the expression data of BC tissues of the Gene Expression Omnibus and The Cancer Genome Atlas (TCGA) were analyzed. Data of the GSE19536 dataset revealed that the expression of miR-133b is downregulated in BC tissues (P<0.001; Fig. 1F); TCGA (NM_004612_1_2228) indicated that TGFβR1 mRNA expression was upregulated in invasive BC (P<0.001; Fig. 1G). These results support the findings of the present study on expression, and the association between miR-133b and TGFβR1 expression in BC cells and tissues. Collectively, these results suggest an inverse association between the expression of miR-133b and TGFβR1, which indicates a potential role for miR-133b in regulating TGFβR1 expression in BC. miR-133b suppresses TGFβR1 expression by directly targeting TGFβR1 3'-UTR in BC cells. miRs play an important role in negatively regulating gene expression by binding to the 3'-UTR of target mRNAs (16). Therefore, it was hypothesized that miR-133b suppresses TGFβR1 expression by directly binding to the 3'-UTR of TGFβR1. To verify this hypothesis, two mRNA target-predicting tools (TargetScan 7.1 and miRPathDB) were used to identify the candidate target genes of miR-133b. As expected, TGFβR1 was validated as a putative target gene of miR-133b, as the 3'-UTR of TGFβR1 was observed to contain one conserved binding site for miR-133b ( Fig. 2A). To further confirm the prediction, a dual-luciferase reporter assay was conducted ( Fig. 2A) to determine whether miR-133b can recognize and target the seed region in the 3'-UTR of TGFβR1. The results revealed that miR-133b significantly inhibited the luciferase activity in MCF-7 (P<0.01; Fig. 2B) Fig. 2B and C). Additionally, overexpression of miR-133b ( Fig. 2D and G) markedly decreased the expression levels of TGFβR1 mRNA (Fig. 2E and H) and protein ( Fig. 2F and I) in MCF-7 and MDA-MB-231 cells, whereas cells transfected with miR-133b inhibitor ( Fig. 2D and G) exhibited increased TGFβR1 expression at the mRNA ( Fig. 2E and H) and protein ( Fig. 2F and I) levels. In conclusion, these findings indicated that TGFβR1 is a direct target gene of miR-133b in BC cells. miR-133b inactivates the TGF-β/SMAD pathway, and suppresses TGF-β-induced EMT and cell invasion. TGFβR1 is an important receptor of the TGF-β/SMAD signaling axis, and plays a key role in TGF-β-induced EMT and cancer metastasis (28)(29)(30). It was observed that miR-133b negatively regulates the expression of TGFβR1 (Fig. 2F and I) in BC cells. In order to determine the biological effects of decreased miR-133b expression on the progression and metastasis of BC, western blot and Transwell assays were performed to analyze the effects of miR-133b on TGF-β-induced EMT and BC cell migration and invasion. As shown in Fig. 3A and B, overexpression of miR-133b significantly inhibited the protein expression of TGFβR1 in MCF-7 and MDA-MB-231 cells. Furthermore, upon treatment with TGF-β1, BC cells transfected with miR-133b exhibited reduced expression levels of p-SMAD3, which is an indispensable downstream effector in canonical TGF-β/SMAD signaling; Snail and N-cadherin expression was reduced compared with that in miR-NC-transfected cells. Of note, the protein expression levels of SMAD3 were markedly unchanged. Conversely, the ectopic expression of miR-133b significantly suppressed the migration and invasion abilities of MCF-7 (P<0.01; Fig. 3C and D) and MDA-MB-231 (P<0.01; Fig. 3E and F) cells induced by TGF-β in vitro. Collectively, these results indicate that miR-133b overexpression may inhibit the expression of TGFβR1, and inhibit TGF-β1-induced EMT and cell migration and invasion via suppression of the TGF-β/SMAD signaling pathway in BC cells. Knockdown of TGFβR1 suppresses TGF-β-induced EMT and BC cell invasion. To determine the functional role of TGFβR1, and examine whether TGFβR1 knockdown can mimic the phenotype generated by miR-133b overexpression, two siRNAs (si-TGFβR1-1 and si-TGFβR1-2) were used to specifically silence TGFβR1 expression. The results demonstrated that TGFβR1 expression was markedly downregulated in the MCF-7 and MDA-MB-231 cells transfected with TGFβR1 siRNA (Fig. 4A-C). Subsequently, the protein expression levels of EMT markers were detected in the presence or absence of TGF-β1. As shown in Fig. 4B and C, knockdown of TGFβR1 significantly inhibited TGF-β-induced upregulation of p-SMAD3, Snail and N-cadherin in MCF-7 (Fig. 4B) and MDA-MB-231 (Fig. 4C) cells. On the contrary, the protein expression levels of SMAD3 were markedly unaltered. Additionally, the results of the Transwell assays revealed that knockdown of TGFβR1 suppressed the migration (P<0.01, Fig. 4D and F) and invasion (P<0.01, Fig. 4E and G) abilities of the cells. Our results indicated that TGFβR1 knockdown mimicked the phenotype induced by miR-133b overexpression, which further suggested that miR-133b inhibits TGF-β-induced EMT, cell migration and invasion by directly targeting TGFβR1 in BC cells. miR-133b attenuates BC cell metastasis in vivo. To further investigate the role of miR-133b overexpression in BC cell metastasis in vivo, LV-miR-133b MDA-MB-231 cells were established. A ~50-fold increased expression of miR-133b in LV-miR-133b MDA-MB-231 cells was observed (P<0.001; Fig. 5A). Western blot analysis confirmed the protein expression levels of TGFβR1 to be significantly downregulated in MDA-MB-231 cells stably overexpressing miR-133b compared with control cells (Fig. 5A). Then, LV-miR-133b and LV-miR-NC MDA-MB-231 cells were intravenously injected into BALB/c nude mice via the tail vein (Fig. 5B). At 8 weeks post-inoculation, the mice were euthanized and the lungs were collected for the analysis of metastases and histological examination. As expected, mice injected with LV-miR-133b MDA-MB-231 cells developed fewer metastatic nodules in the lungs (Fig. 5C and D). Consistently, H&E staining of the lungs demonstrated that the number and size of micrometastases were markedly deceased in LV-miR-133b mice compared with control mice (Fig. 5C and E). In addition, Kaplan-Meier survival analysis (31) revealed that patients with BC and low levels of miR-133b expression had significantly shorter overall survival compared with those with high miR-133b expression levels (Fig. 5F). Conversely, the expression levels of TGFβR1 were negatively associated with overall survival in patients with BC (Fig. 5G). Based on these findings, it may be concluded that low expression levels of miR-133b and high expression levels of TGFβR1 may be associated with poor prognosis in patients with BC. Collectively, these data indicate that the ectopic expression of miR-133b is associated with downregulated TGFβR1 expression, which may suppress the invasion and metastasis of BC cells in vivo. Discussion Despite advances in clinical therapy, metastasis remains the leading cause of mortality among patients with BC (32). It has been well documented that the TGF-β/SMAD signaling pathway plays an important role in EMT and tumor metastasis. Furthermore, the crucial upstream receptor of the TGF-β/SMAD pathway, TGFβR1, plays a pivotal role in TGF-β-induced EMT and tumor metastasis (28)(29)(30). miR-133b was recently reported to be involved in the development and progression of BC (23). The effects of the crosstalk between miR-133b and the TGF-β/SMAD pathway on tumor metastasis and the underlying mechanisms remain unclear, particularly in BC. To the best of our knowledge, the present study is the first to demonstrate that miR-133b directly targets TGFβR1 to inhibit TGF-β-induced EMT and metastasis via suppression of TGF-β/SMAD signaling. Accumulating evidence has demonstrated that miRNAs play a pivotal regulatory role in the initiation and development of various cancers (33). Therefore, an in-depth understanding of the biological function of specific miRNAs in cancer may aid with the evaluation of potential therapeutic targets. Recently, dysregulated miR-133b was reported to be involved in the initiation and progression of BC (23,34); however, the effects of miR-133b on TGF-β-induced EMT and tumor metastasis and the underlying mechanisms require further investigation, particularly in BC. The aim of the present study was to investigate the biological behavior of miR-133b and the possible mechanism of action in BC. As demonstrated by the findings, miR-133b expression was found to be downregulated in BC cell lines and tissues; this decrease was significantly associated with poor prognosis of patients with BC. Furthermore, the expression of miR-133b was inversely associated with lymph node metastasis and TNM stage in patients with BC. Functional analysis indicated that miR-133b significantly inhibited TGF-β-induced EMT and BC cell invasion in vitro and in vivo. The findings of the clinical analysis revealed that miR-133b may serve as a prognostic marker in BC patients; however, whether miR-133b expression levels are correlated with BC recurrence and whether they may serve as a predictor of clinical outcome in BC is yet to be determined. Consistent with our findings, recent studies have suggested that miR-133b is also downregulated in gastric, lung and colorectal cancer (35)(36)(37). On the contrary, Qin et al (38) reported that miR-133b acted as an oncogene, promoting the progression of cervical carcinoma. This discrepancy may be attributed to differences in the tumor microenvironment, tumor heterogeneity or target genes. miRNAs may act as oncogenes or tumor-suppressor genes, which is largely dependent on their targets in a variety of human cancers (39). Bioinformatic algorithms were utilized to predict potential targets, and ultimately identified a novel miR-133b-targeted gene, TGFβR1, which is a critical receptor in the TGFβ1/SMAD signaling pathway. TGFβR1 has been confirmed to play a pivotal role in TGF-β-induced EMT and tumor metastasis (28)(29)(30). It has been well-documented that TGFβ1/SMAD signaling plays a dual role in cancer development and progression. Of note, activated TGF-β/SMAD signaling serves as a tumor suppressor in early-stage tumors, but acts as a tumor promoter in late-stage tumors (40,41). Consistently, Fukai et al (42) demonstrated that the expression of TGFRβ1 was downregulated in early-stage human esophageal squamous cell carcinoma; however, our results were consistent with those of a recent study by Li et al (43) reporting that TGFβR1 mRNA expression was upregulated in hepatocellular carcinoma, with high mRNA levels of TGFβR1 being associated with advanced TNM stage (43). Furthermore, in silico analysis demonstrated that increased TGFβR1 expression was associated with poor prognosis of BC patients. Most importantly, we observed that siRNA-mediated knockdown of TGFβR1 suppressed TGF-β-induced EMT and BC cell invasion, whereas TGFβR1 silencing mimicked the effects and phenotype of miR-133b overexpression. The findings of the present study may provide insight into the association among miRNAs, the TGF-β/SMAD pathway and cancer metastasis. However, there were certain limitations to the present study, including insufficient materials, such as the small number of BC cell lines and BALB/c nude mice used. Moreover, the present study focused on the function of miR-133b in TGF-β-induced EMT, which is an important event promoting metastasis in advanced-stage cancers. Since TGF-β/SMAD signaling plays a key role in regulating cell proliferation in terms of cell cycle progression during the early stages of BC, it would be interesting to address the issue In summary, to the best of our knowledge, the present study was the first to identify that miR-133b inhibits TGF-β-induced EMT and BC metastasis via suppression of TGFβR1 expression in vitro and in vivo. These findings are further supported by the inverse association observed between miR-133b and TGFβR1 expression in BC cell lines and tissues. The results of the present study suggest a novel mechanism by which miRNA regulates TGF-β-mediated EMT and tumor metastasis, and indicate that overexpression of miR-133b may be considered as a promising strategy for treating advanced BC in the future.	2019-09-19T09:09:39.062Z	2019-09-18T00:00:00.000	{ "year": 2019, "sha1": "e0149cc50d9db3cb44a1becf683fc8ee6a5468f4", "oa_license": "CCBYNCND", "oa_url": "https://www.spandidos-publications.com/10.3892/ijo.2019.4879/download", "oa_status": "HYBRID", "pdf_src": "Anansi", "pdf_hash": "7e0c599431f12d28643593c5c2f42160874f4ec3", "s2fieldsofstudy": [ "Biology", "Chemistry" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
16557908	pes2o/s2orc	v3-fos-license	Bacterial strains isolated from river water having the ability to split alcohol ethoxylates by central fission Alcohol ethoxylates (AE) are a major component of the surfactant stream discharged into surface water. The “central fission” of AE with the formation of poly(ethylene glycols) (PEG) is considered to be the dominant biodegradation pathway. However, information as to which bacterial strains are able to perform this reaction is very limited. The aim of this work was to establish whether such an ability is unique or common, and which bacterial strains are able to split AE used as a sole source of organic carbon. Four bacterial strains were isolated from river water and were identified on the basis of phylogenetic trees as Enterobacter strain Z2, Enterobacter strain Z3, Citrobacter freundii strain Z4, and Stenotrophomonas strain Z5. Sterilized river water and “artificial sewage” were used for augmentation of the isolated bacteria. The test was performed in bottles filled with a mineral salt medium spiked with surfactant C12E10 (10 mg L−1) and an inoculating suspension of the investigated bacterial strain. Sequential extraction of the tested samples by ethyl acetate and chloroform was used for separation of PEG from the water matrix. LC–MS was used for PEG determination on the basis of single-ion chromatograms. All four selected and investigated bacterial strains exhibit the ability to split fatty alcohol ethoxylates with the production of PEG, which is evidence that this property is a common one rather than specific to certain bacterial strains. However, this ability increases in the sequence: Stenotrophomonas strain Z5 < Enterobacter strain Z2 < Enterobacter strain Z3 = Citrobacter freundii strain Z4. Graphical Abstract Biodegradation by central fission of alcohol ethoxylates by bacterial strains isolated from river water Introduction Surfactants are a major source of anthropogenic carbon in the aquatic environment, due to their use in a huge number of household and industrial products, often typified by downthe-drain disposal. Non-ionic surfactants (NS) are a major component of the surfactant stream discharged into surface water (Zoller 1994;Fuerhacker et al. 2001;Traczyk et al. 2006;Nowicka et al. 2013a;Kopiec et al. 2014Kopiec et al. , 2015Zembrzuska et al. 2014). 1397 million tonnes of NS were manufactured in the EC in 2012, which is more than the quantity of anionic surfactants (1201 million tonnes) (CESIO Statistics 2012). Alcohol ethoxylates (AE) are the major component of NS flux (Traverso-Soto et al. 2016). Their production in the EC in 2012 was approximately 1 million tonnes. The preferential use of AE is due to their relatively easy biodegradation under aerobic conditions. Their Bcentral fissionŵ ith the formation of poly(ethylene glycols) (PEG) is considered to be the dominant biodegradation pathway (Fig. 1) (Patterson et al. 1967;Swisher 1987;Szymanski et al. 2000;Franska et al. 2003a). This is confirmed by identification of the oligomeric distribution and quantification of PEGs released during biodegradation (Sparham et al. 2008). PEGs are present both in water and in bottom sediments, including estuary sediments (Lara-Martín et al. 2011Traverso-Soto et al. 2016). It should be noted that PEGs are also a technological impurity of AEs, which are a source of these compounds in the environment (0.6-10 %) (Lee et al. 2009). The splitting of AE by bacterial consortia such as activated sludge and river water is well-documented (Patterson et al. 1967;Zanette et al. 1996;Szymanski et al. 2000;Szymański et al. 2001Szymański et al. , 2002aMarcomini et al. 2000;Eadsforth et al. 2006;Morrall et al. 2006). However, information as to which bacterial strains are able to perform this reaction is very limited. Bacteria of Pseudomonas sp. strain TR01 isolated from activated sludge at a municipal sewage treatment plant biodegrade AEs with various numbers of EO units, but not polyethylene glycols (Maki et al. 1994); however, the pathway of this degradation is not given. The gram-positive Microbacterium strain E19, isolated from soil, is able to split fatty alcohol ethoxylates with PEG formation (Nowicka et al. 2013b). This is the only paper concerning the central fission of AEs by an isolated bacterial strain. Investigation of this stage of AE biotransformation is significant for better understanding of the self-cleaning of the aquatic environment. Therefore, more information is required concerning which aquatic bacteria strains are able to split AEs with PEG formation. It should be remembered that several hundreds of families of microorganisms have been identified in river water (Staley et al. 2014). Thus, the question is whether such an ability is common or more specific. The aim of this study was to find an answer to this question. River water from the River Warta (Poznan, Poland) was selected as a source of bacterial strains. This medium is in continuous contact with AEs, and bacterial consortia of river water are able to split AEs (Szymański et al. 2002a). Four bacterial strains were isolated, and all exhibited the ability to form PEG in preliminary experiments. The sample was obtained from seventh day of biodegradation test of surfactant C 12 E 10 by bacteria Enterobacter strain Z3. The bacteria was augmented on an artificial sewage Sequential extraction of tested samples by ethyl acetate and chloroform was used for separation of PEG from the water matrix. This procedure has been successfully used for such a purpose (Szymański et al. 2002a, b;Nowicka et al. 2013a, b). The ethyl acetate phase contains AE, while the chloroform phase contains PEG. The LC-MS technique was selected for PEG determination on the basis of single-ion chromatograms of particular PEG homologues. Two media were selected for augmentation of the isolated bacteria: sterilized river water and Bartificial sewage^used in continuous flow activated sludge tests (CFAS) (Szymanski et al. 2000). River water is a native environment for the isolated bacterial strains in terms of nutritional components. The artificial sewage contains typical components used in bacterial augmentation. Microorganisms Bacterial strains were isolated from river water obtained from the River Warta in Poznan, Poland. A river water sample (100 mL) was filtered through a sterile 0.2 μm filter. The filter was incubated in agar medium at 30°C for 24 h. Separation of pure cultures was performed using the Lisner method. Pure cultures of isolated strains were maintained on an agar medium. Four bacterial strains were isolated and were identified on the basis of phylogenetic trees as Enterobacter strain Z2, Enterobacter strain Z3, Citrobacter freundii strain Z4 (all three from the family Enterobacteriaceae), and Stenotrophomonas strain Z5 (family Xanthomonadaceae). Bacterial strains belonging to these families have been identified in many different rivers throughout the world (Ogbulie et al. 2008;Ganesan and Muthuchelian 2009;Li et al. 2014;Obukhova and Lartseva 2015). An inoculum was prepared by suspending a loopfull from a 3 day culture in a 5 mL of a sterile enrichment broth. The solution was incubated at 30°C for 24 h. The obtained solution was introduced into 45 mL of the sterile enrichment broth and the mixture was incubated at 30°C for 24 h and then was introduced into 450 mL of the sterile enrichment broth and incubated at 30°C for 72 h. Reagents and chemicals Polydisperse surfactant C 12 E 10 (Sigma-Aldrich, St. Louis, MO, USA) was used as received. It is a mixture Fig. 4 Phylogenetic tree of bacteria Enterobacter strain Z2 (a). Concentration of particular PEG homologues (represented as the peak area) at day seven and at the beginning of the biodegradation test of surfactant C 12 E 10 with bacteria Enterobacter strain Z2. Inoculum was augmented on sterilized river water (b). As Fig. 4c with inoculum augmented on an artificial sewage of homologues having an average of 10 oxyethylene subunits. C h l o r o f o r m , s o d i u m c h l o r i d e , a n d s o d i u m hydrocarbonate of analytical grade were obtained from POCh (Gliwice, Poland). All reagents used for preparation of inoculum and mineral salt medium were from POCh, Poland, with the exception of meat extract and peptone, which were from BTL, Poland. MS-grade acetonitrile used for LC-MS measurements and the ammonium acetate used as a mobile phase additive was purchased from Sigma-Aldrich (St. Louis, MO, USA). The HPLC-grade water was prepared by reverse osmosis in a Demiwa system from Watek (Ledec nad Sazavou, Czech Republic), followed by double distillation with a quartz apparatus. Only freshly distilled water was used. All sterile solutions were obtained by treatment in an autoclave at 121°C for 30 min. Biodegradation test The test was performed in bottles filled with 200 mL of medium consisting of a mineral salt medium, spiked with surfactant C 12 E 10 (10 mg L −1 ) and an inoculating suspension of investigated bacterial strain (5 mL Concentration of particular PEG homologues (represented as the peak area) at day seven and at the beginning of the biodegradation test of surfactant C 12 E 10 with bacteria Enterobacter strain Z3. Inoculum was augmented on sterilized river water (b). As Fig. 5c with inoculum augmented on an artificial sewage a rotary shaker to provide oxygen. The samples were preserved with 1 % formaldehyde on selected days of the biodegradation test. Liquid-liquid sequential extraction of substrate and PEG 15 g sodium chloride and 0.1 g sodium hydrogen carbonate were added to 50 mL of the sample. The sample was sequentially extracted with three portions of ethyl acetate, to remove NS, and with three portions of chloroform (10, 10, and 5 mL respectively) (Zgola-Grzeskowiak et al. 2008). An aliquot of the combined chloroform extracts, containing PEG, was taken for LC-MS determination, while the ethyl acetate extracts were discarded. The aliquot of chloroform extract was gently evaporated and reconstituted in the mobile phase. LC-MS LC analysis was performed using the UltiMate 3000 RSLC chromatographic system from Dionex (Sunnyvale, CA, USA). 5 μL samples were injected into a 150 mm × 2.0 mm I.D. analytical column packed with 3 μm TSK Gel Amide 80 from TOSOH Bioscience (Stuttgart, Germany). The column was kept at 35°C. The mobile phase considered of 10 mmol L −1 ammonium acetate in water (A) and acetonitrile (B) at a flow rate of 0.1 mL min −1 . The following gradient was used: 0 min 99 % B, 0.5 min 50 % B, 4 min 5 % B, 7 min 1 % B, 10 min 1 % B. A pre-run time of 9 min was left before the next injection. The LC system was connected to an API 4000 QTRAP triple quadrupole mass spectrometer from AB Sciex (Foster City, CA, USA). The Turbo Ion Spray source was operating in Fig. 6 Phylogenetic tree of bacteria Citrobacter freundii strain Z4 (a). Concentration of particular PEG homologues (represented as the peak area) at day seven and at the beginning of the biodegradation test of surfactant C 12 E 10 with bacteria Citrobacter freundii strain Z4. Inoculum was augmented on sterilized river water (b). As Fig. 6c with inoculum augmented on an artificial sewage positive ion mode. Analytes were detected using the following settings for the ion source and mass spectrometer: curtain gas 10 psi, nebulizer gas 45 psi, auxiliary gas 45 psi, temperature 350°C, ion spray voltage 5500 V, declustering potential 36 V, and collision gas set to medium. Total ion chromatograms within an m/z range between 100 and 1200 and mass spectra corresponding to the obtained chromatographic peaks were recorded. An example of a total ion chromatogram of chloroform extract is shown in Fig. 2a. The peak at 7.09 min corresponds to PEGs, being identified on the basis of the mass spectrum shown in Fig. 2b. PEGs are identified as a series of mass peaks which differ by m/z = 44 Da (Franska et al. 2003b), i.e., the mass of a single oxyethylene unit. A series of mass peaks having m/z = 212 + n44 corresponds to ammonium complexes of PEGs having 4 + n oxyethylene subunits. Apart from this series, another series having m/z = 658 + n44 appears, which corresponds to ammonium complexes of carboxylated surfactant C 12 E x ; there is also a series m/ z = 404 + n22 (double ammonium complexes of carboxylated surfactant C 12 E x ). Apart from ammonium complexes, weaker signals of complexes with Na(I) and K(I) appear, which form a Bforest^of lower mass peaks. Single-ion chromatograms at particular m/z values were recorded, and the areas of the peaks appearing were used as the analytical signals of determined individual PEGs. The example of a series of single-ion chromatograms of PEGs having 4-26 oxyethylene subunits is shown in Fig. 3. Areas of peaks were presumed to be equivalent to PEG concentration. Results and discussion Eight parallel flask biodegradation tests were performed with bacterial strains isolated from river water. Each of the four Fig. 7 Phylogenetic tree of bacteria Stenotrphomonas strain Z5 (a). Concentration of particular PEG homologues (represented as the peak area) at day seven and at the beginning of the biodegradation test of surfactant C 12 E 10 with bacteria Stenotrphomonas strain Z5. Inoculum was augmented on sterilized river water (b). As Fig. 7c with inoculum augmented on an artificial sewage isolated stains was tested twice: (i) with bacteria augmented in sterilized river water, and (ii) with bacteria augmented in artificial sewage. Tests were performed in aerobic conditions with surfactant C 12 E 10 as the sole source of organic carbon. Determination of PEG concentration was performed at the very beginning of the tests and on the seventh day of the test. The samples were sequentially extracted with ethyl acetate (to remove residual C 12 E 10 ) and with chloroform. Aliquots of chloroform extracts were determined for PEG concentration. The results for Enterobacter strain Z2 are shown in Fig. 4a, together with its phylogenetic tree, being the Bproof of identity^of the strain. The concentration of a particular homologue is represented by a peak area on a single-ion chromatogram. The concentrations of homologues having 4-26 oxyethylene units was determined. The polydispersity of PEG obtained after the central fission of surfactant C 12 E 10 is an obvious consequence of the polydispersity of surfactant C 12 E 10 . A strong increase in the concentration of homologues having 5-18 oxyethylene subunits is apparent in Fig. 4c. The results at the seventh day of the experiment are approximately 4 times higher than at the beginning of the test. This increase is obvious evidence of the central fission of surfactant C 12 E 10 by bacteria of Enterobacter strain Z2. The presence of PEG homologues at the beginning of the test is due to their presence in surfactant C 12 E 10 as a technological impurity. A lower increase in PEG concentration is also observed in Fig. 4b, i.e., in the test with Enterobacter strain Z2 augmented in sterilized river water. It seems that this medium is less favorable to the growth of Enterobacter strain Z2 than the artificial sewage. The results concerning Enterobacter strain Z3, Citrobacter freundi strain Z4, and Stenotrophomonas strain Z5, together with their phylogenetic trees, are shown in Figs. 5, 6, and 7 respectively. All these bacterial strains causes an increase in PEG concentration, although to different degrees. Enterobacter strain Z3 (Fig. 5) and Citrobacter freundii strain Z4 (Fig. 6) show a stronger ability to split surfactant C 12 E 10 with PEGs formation. In both cases the PEG concentration is 5 times higher than that at the start of the biodegradation tests. No significant difference is observed between the experiments with bacteria augmented with sterilized river water and with the artificial sewage. The production of PEG by Stenotrophomonas strain Z5 is much weaker (Fig. 7). The results on the seventh day of the test are only 2-2.5 times higher than at the start of the biodegradation tests. The fact that all investigated bacterial strains exhibit the ability to split fatty alcohol ethoxylate with the production of PEGs is evidence that such a property is a common one rather than specific to certain bacterial strains. This conclusion may be supported by the splitting of fatty alcohol ethoxylates with PEG formation by bacteria of Microbacterium strain E19 (Nowicka et al. 2013b). This bacterium is gram-positive (belonging to the phylum Actinobacteria), while all four strains investigated in the present work are gram-negative (belonging to the phylum Proteabacteria). Moreover, they belong to two different families: Enterobacteriaceae (Enterobacter strain Z2, Enterobacter strain Z3, Citrobacter freundi strain Z4) and Xanthomonadaceae (Stenotrophomonas strain Z5). Thus, the ability to split AEs in accordance with the central fission pathway is a common property of the kingdom Bacteria. However, different strains exhibit variable PEG productivity; even two strains belonging to the same genus Enterobacter (strains Z2 and Z3) show significantly different PEG productivity. Moreover, the conditions of bacterial augmentation have a strong effect on the AE central fission. Finally, the presence of carboxylated AEs in mass spectra (see Fig. 2b) may be evidence of the existence of an alternative AE biodegradation pathway, parallel to central fission. Conclusions The ability to split alcohol ethoxylates in accordance with the central fission pathway with the formation of poly(ethylene glycols) is a common property of microorganisms belonging to the kingdom Bacteria. Polydisperse PEGs are produced from polydisperse AEs. This ability is subject to variation among strains belonging to various families. Even strains belonging to the same family exhibit differences in their PEG productivity. The conditions of bacterial augmentation are also significant for the PEG productivity of particular strains.	2018-04-03T04:55:59.481Z	2016-04-07T00:00:00.000	{ "year": 2016, "sha1": "b81b5fdd268cd788a77c666ad77fe8eb4b282868", "oa_license": "CCBY", "oa_url": "https://europepmc.org/articles/pmc4943993?pdf=render", "oa_status": "GREEN", "pdf_src": "PubMedCentral", "pdf_hash": "b81b5fdd268cd788a77c666ad77fe8eb4b282868", "s2fieldsofstudy": [ "Engineering" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
348207	pes2o/s2orc	v3-fos-license	iPSC-Based Cell Therapy: An Important Step Forward Morizane et al. (2013) show that donor-matched differentiated derivatives of induced pluripotent stem cells (iPSC) do not cause an immune response after transplantation, whereas transplantation of HLA-mismatched iPSC derivatives to the same site clearly does. The importance of these results is discussed in this commentary as we assess how best to move forward with iPSC-based cell therapy. show that donor-matched differentiated derivatives of induced pluripotent stem cells (iPSC) do not cause an immune response after transplantation, whereas transplantation of HLA-mismatched iPSC derivatives to the same site clearly does. The importance of these results is discussed in this commentary as we assess how best to move forward with iPSC-based cell therapy. One promise of IPSC-based cell therapy is that personalized medicine using cells from the patient may be feasible. Somatic cells could be harvested and reprogrammed with integration-free methods to produce a stock of clinically compliant pluripotent cells ready for use at any time in later life. To ensure that this may really become possible, various groups have been working on the two major hurdles: (1) reduce the cost of development of iPSC and (2) develop clinical grade processes to generate iPSC lines. Other groups have approached the problem by suggesting that, although not fully personalized, one could perhaps develop HLA-matched banks of lines; indeed, the Japanese government and others have proposed building a collection of diverse HLA types that could be then matched between donor and recipient, akin to the current practice using bone marrow and tissue transplants (Taylor et al., 2012;Cyranoski, 2012). Commercial entities have approached this by offering to develop personalized iPSC banks, rather like the private cord blood banks that will store individual samples in anticipation of future need in ''for-profit'' models (Cave, 2013). In principle this process seems reasonable; however, several questions need to be addressed. Perhaps the most important is whether the cost of making iPSC lines for a single individual is worth the time and effort or whether cells matched at the major loci of the histocompatibility complex may be adequate. Some experiments have even suggested that mismatched cells may also be an option. The reverse issue also needs to be considered: it is not unreasonable to suppose that autologous iPSC may generate an immune response because of their exposure to foreign antigens during the culture period or because of their expression of embryonic antigens not normally present in the adult. Likewise, viral agents used for reprogramming can activate the innate immunity pathway and thus cause an immune response (Zhao et al., 2011;Guha et al., 2013). These issues are critical, and adequate experimental models to assess these issues directly need to be developed. Unambiguous answers will dictate where resources and time should be allocated to develop iPS cell based therapy. In this issue of Stem Cell Reports, Morizane and colleagues (Morizane et al., 2013) have developed an elegant system to address these issues. They have taken advantage of a primate colony to generate matched, related, and unmatched iPSC, which they have then differentiated into dopaminergic neurons. These neurons were then transplanted without immune suppression into the monkey striatum as autografts or HLA-mismatched allografts to evaluate immune responses (Figure 1). The animals were followed over a sufficiently long time period to allow examination of acute, hyperacute, and chronic immune responses. The authors were able to make two important observations: (1) mismatched cells generated an immune responsenot the acute or hyperacute type of rejection response seen in mismatched organs but a slow chronic rejection phenotype mediated by macrophages and T cells; and (2) no rejection was seen in the matched transplants, suggesting a clear, albeit small, short-term and larger long-term benefit of developing matched iPSC lines. Of importance was the observation that such rejection can occur even in the brain, which is widely considered as somewhat immune privileged. The authors have established a system in which they can begin to dissect out even more subtle issues related to immune response in a species that is closely related to humans and has a life span that will allow longer term studies to be performed. These experiments compliment the elegant work of Araki et al. (2013), which showed that autologous or matched ESC and iPSC are equivalent in that they lack of an immunogenic response when differentiated cell products are transplanted and provide an opportunity to perform similar experiments in a primate model. The same models will also allow us to evaluate which immune suppressive regime is best in the long run and when and if any such regime can be discontinued. Two other advantages of this model system are that we now have a primate model and corresponding pluripotent cells from the same species that can be differentiated into many of the cell types that are being investigated for cell based treatment. Thus, both human and primate iPSC derivatives can be compared in the same animal species. This should make it possible to determine whether the experimental paradigm of the xenograft (human into primate, human into mouse, human into porcine, etc.) and its associated immune suppressive regime has any value in predicting what we really want to know: will human pluripotent stem cell derivatives have therapeutic value in humans and what will be the requirements for immune matching or (long-term) immune suppression. Embryonic stem cell (ESC) lines can be derived from primates, and were, in fact, derived before human ESC lines (Thomson et al., 1995), and iPSC could be made from differentiated derivatives of primate ESC, as has been shown for rodent and human lines (Teichroeb et al., 2011). One could therefore extend these experiments to study subtle differences between ESC and iPSC cells over a long time period in a model that is as close to human as is possible. ACKNOWLEDGMENTS This work was supported by the NIH Center of Regenerative Medicine, a NIH common fund initiative. The opinions expressed in this article represent the views of the author and in no way can be construed to reflect the opinion or policy of the NIH. Figure 1. A Schematic Outline of the Experiment iPSC were made using viral vectors from animals that were maintained in a colony. Differentiation was performed using standard published protocols and transplants were placed within the brain.	2016-05-04T20:20:58.661Z	2013-10-15T00:00:00.000	{ "year": 2013, "sha1": "6d3ea293f91f1f286a98e116428570a2d2005926", "oa_license": "CCBYNCND", "oa_url": "https://doi.org/10.1016/j.stemcr.2013.10.002", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "6d3ea293f91f1f286a98e116428570a2d2005926", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
215773246	pes2o/s2orc	v3-fos-license	Racial/ethnic disparities in colorectal cancer treatment utilization and phase-specific costs, 2000-2014 Background Our study analyzed disparities in utilization and phase-specific costs of care among older colorectal cancer patients in the United States. We also estimated the phase-specific costs by cancer type, stage at diagnosis, and treatment modality. Methods We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare database to identify patients aged 66 or older diagnosed with colon or rectal cancer between 2000–2013, with follow-up to death or December 31, 2014. We divided the patient’s experience into separate phases of care: staging or surgery, initial, continuing, and terminal. We calculated total, cancer-attributable, and patient-liability costs. We fit logistic regression models to determine predictors of treatment receipt and fit linear regression models to determine relative costs. All costs are reported in 2019 US dollars. Results Our cohort included 90,023 colon cancer patients and 25,581 rectal cancer patients. After controlling for patient and clinical characteristics, Non-Hispanic Blacks were less likely to receive treatment but were more likely to have higher cancer-attributable costs within different phases of care. Overall, in both the colon and rectal cancer cohorts, mean monthly cost estimates were highest in the terminal phase, next highest in the staging phase, decreased in the initial phase, and were lowest in the continuing phase. Conclusions Racial/ethnic disparities in treatment utilization and costs persist among colorectal cancer patients. Additionally, colorectal cancer costs are substantial and vary widely among stages and treatment modalities. This study provides information regarding cost and treatment disparities that can be used to guide clinical interventions and future resource allocation to reduce colorectal cancer burden. Introduction Colorectal cancer (CRC) is the third most common cancer among both men and women in the United States and the third leading cause of cancer death. Approximately 145,600 new cases and 51,020 deaths are expected in 2019. [1] Survival rates have been increasing over time, with the five-year survival rate currently around 64%. [2] CRC nevertheless remains financially burdensome, with an estimated annual cost exceeding $17 billion, [3] and has long been known to disproportionately afflict certain racial and ethnic minority groups in the U.S. [4][5][6] A particularly pronounced and long-standing trend in the U.S. concerns the lower colorectal cancer survival rate among Black patients as compared to White patients. Disparate cancer outcomes can result in part from measurable inequities in screening rates and access to other early detection measures, as well as differences in treatment. [7][8][9] Indeed, Black colorectal cancer patients are more likely to be diagnosed at later stages and less likely to have potentially curative treatment such as surgery. [4][5][6]10] Multiple studies have shown, however, that previously investigated, measurable factors such as stage at diagnosis and treatment receipt are not solely responsible for the observed survival difference between Blacks and Whites in the U.S. [5,[11][12][13] A recent study suggests that the stage disparity may be abating while incidence and mortality remains disproportionately high among Black patients, underscoring the need for further research into drivers of the survival disparity. [14] The extent to which these trends translate into cancer-associated cost disparities is unknown. These racial/ethnic disparities in colorectal cancer outcomes pose challenges to health equity among Americans. In a previous study, we determined lung cancer-related costs in the last month of life for White, Black, Asian and Hispanic patients separately and identified significant racial/ethnic cost disparities. [15] Given the known differences in colorectal cancer treatment and outcomes by race/ethnicity, we were interested in seeing whether cost disparities persisted in this context. To pinpoint where disparities might be most pronounced throughout the entire course of care, we used a large, population-based database to estimate CRC care costs by treatment phase. We determined whether cost differences existed between racial/ethnic groups within each phase. We also assessed differences in treatment receipt by race/ethnicity after controlling for potential confounders. An additional focus of our study was to use recent data to update the currently available phase-specific cost estimates for CRC, irrespective of race/ethnicity. [16,17] These accurate cost estimates are needed in cost-effectiveness analyses evaluating the relative benefits of investing in new treatments. Detailed estimates may also be used to calculate future total costs of CRC care when new screening guidelines or treatment modalities change treatment patterns. Cohort inclusion and exclusion We calculated mean (95% CI) monthly costs of colorectal cancer by phase of care using the Surveillance, Epidemiology and End Results (SEER)-Medicare database. SEER contains demographic and clinical information on cancer patients collected from 17 United States registries representing about 28% of the population. [18] The Medicare dataset includes health insurance enrollment information and detailed claims for 97% of the population aged 65 and older. [19] SEER-Medicare is a linkage of these two datasets and includes approximately 95% of the SEER registry population aged 65 and older. [19] We included patients aged 66 and older who were diagnosed with colorectal cancer as their first and only cancer from 2000-2013 and were continuously enrolled in Medicare Parts A and B coverage from 15 months prior to diagnosis until death or December 31, 2014. We excluded patients enrolled in a Health Maintenance Organization (HMO) during this period because these claims are not available in the SEER-Medicare database. This was to ensure that we had complete claims information for care for all patients. We also excluded patients whose Medicare enrollment was not due to age, who were diagnosed at autopsy only, or who had an unknown cancer stage. Finally, we excluded patients who had an unknown month of cancer diagnosis, a date of death recorded in the Medicare database that differed from that recorded in the SEER database by more than three months, no costs recorded post-diagnosis, costs for claims with unknown dates, or any costs post-death to reduce the potential for including incorrect claims or those due to data entry errors. [20] An inclusion and exclusion criteria flowchart is reported in Fig A S1 File. Cost-estimation methods We based our cost estimation methods on previous studies. [20][21][22][23] We defined treatment modalities for patients with stages I-III CRC based on treatment(s) initiated two months prior to cancer diagnosis through six months after diagnosis. The two months prior to diagnosis were included to account for any treatment given to a symptomatic cancer patient who had not yet been diagnosed, as well as possible errors in dates recorded in the claims. Treatment modalities for stage IV CRC patients were defined as treatment(s) ever received. Patients were defined as having treatment if there was a claim in any of the SEER-Medicare claims files with a code pertaining to that treatment. A full list of codes used to define treatment modalities is in Table A S1 File. [24,25] Patients who had no treatment claims were considered to be not actively treated with surgery, chemotherapy, or radiation and were defined as having received best supportive care. Those who only had a surgery date within the two months prior to diagnosis were not counted as having received surgery. Patients remained in their treatment modality subgroup throughout the analysis. Each patient's costs were divided into separate phases of care-staging (or surgery), initial, continuing, and terminal (Fig 1). [16,20,21,23] The staging phase was defined as a one-month stage beginning on the date of diagnosis for nonsurgical patients. Patients who received surgery were given a one-month surgery phase, beginning on the date of major surgery, to allow us to isolate the cost of surgery from any post-operative care. [20] In addition, surgical patients were given a variable presurgery phase beginning on the date of diagnosis. Patients had an initial phase with a maximum length of six months starting after the staging or surgery phase, followed by a continuing phase varying in length between patients depending on survival time. Those patients who died on or before December 31, 2014 had a six-month terminal phase that ended on the date of death. We chose to definite the initial and terminal phases as six months to allow for a more accurate cost estimation of these distinct phases. [20,23] We allocated costs first to the terminal phase (when applicable), then the staging or surgery phase, the initial phase and, lastly, the continuing phase. For example, a patient who lived 16 months would have a six-month terminal phase, one-month staging phase, six-month initial phase, and three-month continuing phase. CRC patients were only factored into cost averages for phases for which they had at least one month. Matched control cohort We created a control subject cohort from the random sample of 5% of all Medicare enrollees aged 65 years and older who were continuously enrolled in Medicare Part A and B through the study period, were not enrolled in an HMO, and had no cancer diagnosis. This control cohort was matched to CRC cancer patients on an individual (1:1) level within each phase by 5-year age group, sex, and SEER registry region (Northeast, South, Midwest, West). [21] Because control subjects did not have cancer, each was randomly assigned a "pseudodiagnosis" date matching the cancer diagnosis date of a randomly chosen CRC patient. [21] Care costs for control subjects were allocated to either the continuing or the terminal phase. The terminal phase was defined as the subject's last six months of life, and the continuing phase was defined as all months between the "pseudodiagnosis" date and the start of the terminal phase. In addition to being matched by 5-year age group, sex, and SEER registry region, cancer patients and control subjects were also matched by phase of care. To determine cancer-attributable costs during both the initial and the continuing phase, patients were matched to control subjects in the continuing phase of care. To determine cancer-attributable costs in the terminal phase, cancer patients who died of their cancer were matched to the control subjects in the continuing phase, while cancer patients who died from other causes were matched to control subjects in the terminal phase. [21] This is to account for the fact that end-of-life care costs are typically high regardless of the cause of death. [16,26] Cancer-attributable costs were not calculated for the staging, presurgery, or surgery phases. Cost analysis We calculated costs as the sum of Medicare reimbursements, co-insurance reimbursements, and any co-payments and deductibles billed to patients. [21] We calculated total and patientliability costs for each patient, which may include costs paid by a purchased Medigap policy to help cover a patient's coinsurance, copayment, and deductible costs. [20,27] Cancer-attributable costs were estimated by subtracting the matched noncancer subject's mean monthly phase costs from the cancer patient's mean monthly phase costs for the initial, continuing, and terminal phases. We report treatment modality costs if at least 10% of patients within a stage received that treatment, except for best supportive care, which is reported for all stage and phase subgroups. Costs were converted to constant 2019 U.S. dollars by adjusting Part A claims using the CMS Prospective Payment System Hospital Price Index and Part B claims using the Medicare economic index. [28,29] All costs are reported in 2019 U.S. dollars. Statistical analysis Logistic regressions. We fit logistic regression models to determine predictors of treatment receipt (surgery, radiation, or chemotherapy) for colon and rectal cancer. We included the following variables in the models: sex, age at diagnosis, race/ethnicity, marital status, ecological socioeconomic status (SES) quintile, SEER region, urban/rural status, year of diagnosis, AJCC stage at diagnosis, and Charlson comorbidity score (0, 1, 2+). We defined patient race/ ethnicity as Non-Hispanic (NH) White (White), NH Black (Black), Hispanic, NH Asian including Pacific Islander (Asian), and "Other" (defined as NH Native American, Native Alaskan, Other, or Unknown in SEER-Medicare). Since the SEER-Medicare database does not include income data at the individual level, we created an ecological SES index using quintiles of ZIP code-level median household incomes from the provided U.S. census data. [30] We categorized SEER regions into Northeast, South, Midwest, and West according to the U.S. Census Bureau's definition. [31] Comorbidity scores were calculated using the Deyo adaptation of the Charlson comorbidity index for Medicare claims during the year prior to cancer diagnosis. [32][33][34] Linear regressions. We constructed linear regression models for each phase to examine the association between log-transformed costs and the above patient and clinical characteristics. We used the log-transformed total costs for the staging and surgery phase and the logtransformed cancer-attributable costs for the initial, continuing, and terminal phases. We included treatment modality as an additional covariate. Results are reported as relative cost ratios compared to the reference group. Statistical significance was defined as P < 0.05 in a two-sided test. We performed all statistical analyses using SAS software, version 9.4 (SAS Institute, Inc., Cary, NC). Ethics statement This study was approved by the Institutional Review Board at Massachusetts General Hospital. The SEER-Medicare data is a limited data set without personal identifiers. The Institutional Review Board waived the requirement for individual informed consent for use of this data. Patient cohort characteristics Our cohort included 90,023 colon cancer patients and 25,581 rectal cancer patients; their characteristics are reported in Table 1. Among colon cancer patients, 43.1% were male, 82.6% were White, and 19.7% were diagnosed with stage IV disease. The median (25 th , 75 th percentile) age at diagnosis was 78 (73, 84). Among rectal cancer patients, 51.2% were male, 82.3% were White, 33.0%, 17.9% were diagnosed with stage IV disease, and the median (25 th , 75 th percentile) age of diagnosis was 76 (71, 82). Racial/ethnic disparities in treatment receipt We considered whether there were racial/ethnic differences in treatment receipt after controlling for known patient and clinical characteristics. As shown in Table 2, treatment receipt was significantly lower for Blacks compared to Whites, after controlling for other covariates. Specifically, Black patients were less likely to receive surgery (OR: 0.76; 95% CI: 0.62-0.72; p<0.0001), radiation (OR: 0.76; 95% CI: 0.65-0.89; p = 0.0005), or chemotherapy (OR: 0.798; 95% CI: 0.74-0.84; p<0.0001) when compared to White patients. Relative ratios in monthly cancer-attributable costs by initial, continuing, and terminal phase for rectal cancer patients are reported in Table 5. There were no significant racial/ethnic relative cost ratios reported during the initial phase. In the continuing phase, however, Blacks had a significantly higher relative cost when compared to White Detailed phase-specific cost estimates For comparison among cancer types, stages, and phases of care, we report total, cancer-attributable, and patient-liability cost estimates by stage at diagnosis and treatment phase in Table 6. Additionally, mean (95% CI) monthly total cost estimates for each cancer stage at diagnosis and treatment phase are shown in We further estimated these costs based on specific treatment modalities within each stage. These detailed phase-specific costs by stage and treatment modality are reported in Text A-B S1 File and Tables D-H S1 File. Discussion Our study analyzed treatment utilization and costs among SEER-Medicare patients diagnosed with colon or rectal cancer and identified several racial/ethnic disparities in rates of treatment receipt and relative phase-specific costs. We also estimated these phase-specific costs by stage at diagnosis and treatment subgroup and found that cancer-attributable costs varied widely between subcategories. Disparities in treatment receipt and relative cost ratios After controlling for all other characteristics, we found that Black colon cancer patients were significantly less likely than Whites to receive treatment with surgery, radiation, or chemotherapy. Black rectal cancer patients were also significantly less likely to receive treatment with surgery or chemotherapy. Additionally, when controlling for factors including treatment type, Black colon and rectal cancer patients incurred statistically significantly greater costs in every phase of care, compared to White patients, except for the staging and initial phases of rectal cancer treatment. These trends were observed to a lesser extent when comparing Hispanic and Asian patients with Whites. Significant differences most often revealed Hispanic and Asian patients to have lower rates of treatment utilization and higher surgery costs. Notably, colon and rectal cancer patients in all three racial/ethnic groups had higher costs in the terminal phase than did White patients. Factors such as stage at diagnosis, educational level, SES, and personal beliefs can contribute to observed racial/ethnic disparities in treatment utilization and cost in the U.S. [35,36] SES is, for example, known to exacerbate inequalities in treatment receipt; however, the disparities we identified are not attributable to SES, as we controlled for this variable. Compared to Whites, certain racial/ethnic groups have historically been more likely to be diagnosed with stage III-IV colorectal cancer regardless of SES. [11,14,37,38] We therefore controlled for stage at diagnosis as well as SES. Our results suggest that factors independently attributable to race/ethnicity may contribute to care utilization and cost disparities, even after controlling for known potential confounders. No conclusive evidence exists for any sort of racial/ethnic biological or genetic determinant of CRC outcomes. [39][40][41][42] On the other hand, cultural and personal beliefs, though difficult to measure, may substantially impact health outcomes. Indeed, studies have shown that even when rates of specialist consultations are similar between Black and White patients diagnosed with advanced stage disease, Black patients are less likely to undergo treatment in the U.S. [43][44][45] Patients also may not seek out appropriate care when a language barrier exists. The need to improve language services in healthcare settings is well-established, but the issue is difficult to resolve given the perceived costs and impracticalities of using professional interpreters throughout the course of care; furthermore, there are generally few resources available to physicians to address language barriers in the absence of more favorable alternatives. [46][47][48][49] Poor communication between patients and healthcare providers for other reasons may deleteriously affect health outcomes in ways that are difficult to measure. Prior studies have found that Black, Asian, and Hispanic cancer patients are more likely to have lower quality communication as measured by access to clear information on treatment pros and cons and prognosis; time devoted to patient-centered communication and relationship-building between patients and healthcare providers; responsiveness to patient requests; patients' willingness to ask questions related to their care and condition; and knowledge about who to go to for desired information. [50][51][52][53][54] General mistrust of the healthcare system among Black Americans is also well-established and can impede patients from seeking out needed care. [55][56][57] It is challenging to overcome these disparities within the American healthcare system. A greater comorbid disease burden among Black patients may influence observed disparities. Most studies on the subject have found that cancer patients with comorbidity are less likely to receive treatment with curative intent. Physicians may not deem the benefits of such therapy worth the risks of toxicity and of exacerbating a patient's pre-existing conditions. [58] These findings may validate our results suggesting racial/ethnic disparities in cancer treatment utilization. The causal connection between comorbidity and cost disparity by race/ethnicity is, however, unknown. In order to investigate this connection, a causal inference analysis is needed, which is beyond the scope of our study. [59] More generally, our cost estimates potentially reflect the financial effects of suboptimal disease management. While less likely to receive surgery, radiation, and chemotherapy, certain minority groups also had higher phase and cancer-specific costs even after SES, geographic region, stage of diagnosis, treatment type, and comorbidity were controlled for. Earlier studies have shown that racial/ethnic minority groups are less likely to receive appropriate cancer care at the optimal time. [60][61][62][63] Failure to alleviate the disease burden in its early stages or to address accompanying comorbidities can culminate in more intensive and costly end-of-life treatment, with higher rates of emergency visits and hospitalizations. [64][65][66][67] Of note, racial/ethnic minorities in the U.S. are also generally less likely to use palliative care and hospice services, compounding high end-of-life expenditures as hospital-based end-of-life care is more likely to be aggressive, and therefore costly, in comparison. [68][69][70][71] Our high terminal phase cost estimates for racial/ethnic minorities may reflect these realities. The end-of life period, however, is not the only point at which low treatment utilization may translate to higher cancer-attributable costs. This may occur as early as the initial phase, as our results reveal higher initial and continuing phase costs for Black patients than for Whites. Further research is needed into the specific aspects of care that may be driving these earlier cost trends. More broadly, the influence of patient beliefs, preferences, and provider interactions on treatment decisions and disease management must be better understood in order to reduce disparities in colorectal cancer outcomes. Stage and phase-specific cost estimates Overall, regardless of stage at diagnosis and cancer type, mean monthly total costs were highest in the terminal phase. Costs were lower in the staging phase, followed by the initial phase, and lowest in the continuing phase. Mean monthly total costs in the surgery phase were high, regardless of stage at diagnosis and cancer type. Given that most colorectal cancer patients in our sample received surgery, most treatment-related costs are attributable to the surgery phase rather than the initial phase, especially among patients presenting with stage I disease. Radiation and chemotherapy treatment rates increased as stage at diagnosis increased; therefore, initial phase cost estimates increased in tandem, as patients diagnosed at later stages typically receive these treatments over the course of several months. Colorectal cancer treatment cost estimates have been previously published, but these rely on SEER-Medicare data prior to 2007. Our estimates use more recent data. [16,17] Compared to our figures for overall cancer-attributable costs, those reported in prior studies are lower; Brown et al., for example, estimated a mean annual terminal phase cost of $26,200 (or $2,183/month) in 1994 U.S. dollars for stage IV CRC patients, [16] and Lang et al reported a mean annual terminal phase cost of $27,898 (or $2,324/month) in 2006 dollars for stage IV colon cancer patients. [17] In comparison, we estimated a mean monthly cancer-attributable cost of $13,316 in 2019 dollars for stage IV colon cancer patients. Together these results suggest that cancer costs have been rising over time. We also define the initial and terminal phases as six months each, with the initial phase beginning one month after diagnosis, rather than adopting the 12-month definition used in the previous studies. We believe our methods more precisely circumscribe the financially distinct phases of CRC care and, thus, allow us to more accurately determine costs. Strengths and limitations Our study contributes to existing literature by demonstrating the persistence of potential differences in treatment utilization and cancer-attributable costs among racial/ethnic subgroups throughout the course of care. We also provide updated, detailed cost estimates using more recent data, which are a better reflection of current treatment patterns. However, there are several limitations inherent in claims data analyses. Our study was limited to patients over the age of 65 who were diagnosed in a SEER region. Therefore, our results may not be generalizable to younger patients or the entire U.S. population. Our study cohort is limited to those patients who were continuously enrolled in Medicare Parts A and B and who had no HMO for the entire study period. This requirement could potentially bias the analysis by excluding those with noncontinuous Medicare enrollment, or who had their care covered by other forms of insurance. We are unable to determine whether patient-liability costs were paid out-of-pocket or covered by a Medigap program. Since we are not able to determine individual SES from SEER-Medicare, we used an aggregate variable, which many not reflect true SES. However, this method has been determined to be effective for this database. We were not able to determine which treatments received by stage IV patients were palliative. Finally, some costs may have been misclassified. For example, since our study end date was December 31, 2014, patients who died in early 2015 do not have terminal phase costs. In conclusion, racial/ethnic disparities in CRC treatment receipt persist, with some groups incurring higher care costs during specific phases of care. In addition, the updated cost estimates for CRC care remain substantial and vary widely by phase, cancer site, stage at diagnosis, and treatment modality. These findings are useful for health care professionals seeking to identify potential disparities in care. Our cost estimates may also help to guide future resource allocation and reduce CRC burden.	2020-04-16T09:07:26.712Z	2020-04-14T00:00:00.000	{ "year": 2020, "sha1": "d4568c744ed36b5fd485bccff6e755c751637bf3", "oa_license": "CCBY", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0231599&type=printable", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "440326b431d9118d1f4ed0b64541287ff1acdba9", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
214160540	pes2o/s2orc	v3-fos-license	Use of an integrated pan-cancer oncology enrichment NGS assay to measure tumour mutational burden and detect clinically actionable variants The identification of tumour mutational burden (TMB) as a biomarker of response to PD-1 immunotherapy has necessitated the development of genomic assays to measure this. We carried out comprehensive molecular profiling of cancers using the Illumina TruSight Oncology panel (TSO500) and compared to whole genome sequencing. Methods Cancer samples derived from formalin fixed material were profiled on the TSO500 panel, sequenced on an Illumina NextSeq 500 instrument and processed through the TSO500 Docker Pipeline. Either FASTQ files (PierianDx) or VCF files (OncoKDM) were processed to understand clinical actionability In total, 108 samples (a mixture of colorectal, lung, oesophageal and control samples) were processed via the DNA panel. There was good correlation between TMB, SNV, indels and CNV as predicted by TSO500 and WGS (R2>0.9) and good reproducibility, with less than 5% variability between repeated controls. For the RNA panel, 13 samples were processed, with all known fusions observed via orthogonal techniques detected. For clinical actionability 72 Tier 1 variants and 297 Tier 2 variants were identified with clinical trials identified for all patients. The TruSight Oncology 500 assay accurately measures TMB, MSI, single nucleotide variants, indels, copy number/structural variation and gene fusions when compared to whole genome sequencing and orthogonal technologies. Coupled with a clinical annotation pipeline this provides a powerful methodology for identification of clinically actionable variants. Introduction Recent developments in next-generation sequencing and tumour immunology have allowed the discovery that targeting the CTLA4, PD-1 and PD-L1 receptors using therapeutic monoclonal antibodies (1) can unmask cancer to the immune system, facilitating its immunemediated destruction. Although initial trials of PD-1 inhibitors had mixed results (2,3), as with previous targeted therapies, it was determined that a specific tumour genotype was required in order for these inhibitors to be effective, leading to the finding that dramatic regression of tumours could occur with the correct genotype. In order for tumours to become immunogenic, a high neoepitope load must be generated via hypermutation (4)(5)(6), ideally indel/frameshifts or non-synonymous mutations that generate novel proteins that can be recognised by the immune system. These neoepitopes can then be presented via MHC in order to aid immune killing (7). The CHECKMATE (8-10) series of trials have suggest that a specific threshold of "tumour mutational burden" (TMB) must be reached in order for PD-1 blockade to become effective. Although TMB has variable definitions, it is broadly accepted (9) as the number of missense mutations in the tumour genome, either divided by the size of the exome panel (35)(36)(37)(38)(39)(40)(41)(42)(43)(44)(45) or via the size of the human genome for whole genome sequencing (3.3 Gb). Based on the CHECKMATE trials, the suggested TMB threshold is greater than 10 mutations/mb, based on the objective response rates of the tumours in these studies not improving much beyond this threshold. Initially, TMB was measured using whole genome and whole exome sequencing (11), however these technologies are not cost effective currently for routine use in the clinic. Despite the falling cost of next-generation sequencing reagents, the volume of data required for sufficient coverage of either whole genome (200-300 Gb for 60X read depth) or whole exome (4-5 Gb for 100X read depth) sequencing make it impractical except for dedicated sequencing cores. Secondly, even with high read depth, sufficiently deep coverage in order to identify rare subclonal (12) mutations that may contribute to the neoantigen load is required, of the order of 500X. Thus, whole genome/exome coverage is not cost effective (13). Rizvi et al (13) demonstrated that in order to accurately measure TMB using a NGS-based assay, a panel size of at least 1.5 megabases is required. This panel size offers opportunities for a pan-cancer assay, as a panel of this size could cover the majority of known driver genes across multiple cancer types. In designing an oncology assay, ideally other types of variations would be included. Recent studies (14,15) have shown the . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint potential utility of selecting targeted therapies using large gene panels and therefore a panel should include mutations associated with targeting therapies. An additional advantage of panel-based designs is the ability to enrich RNA targets. Recent studies have shown the importance of RNA fusions such as the TMPRSS-ERG fusion in prostate cancer (16), the FGFR2 fusion in cholangiocarcinoma (17), and the NTRK fusion in lung and other cancers (18). These fusions are either targetable with molecularly targeted agents (e.g. larotectinib (19) or pemigatinib (20)) or are prognostically relevant (i.e. TMPRSS-ERG). An ideal oncology panel-based assay would have several characteristics (21): enrichment chemistry rather than PCR chemistry for identification of rare alleles with straightforward library preparation; a broad panel that targets the majority of DNA & RNA alterations in cancer; rapid run time; prediction of novel biomarkers such as TMB; and a standardised, reproducible analysis pipeline that can be used in a clinical setting. In this study, we present our initial results using the Illumina TruSight Oncology 500 assay across a range of cancer types. We benchmarked it against whole genome and exome sequencing, as well as determining its ability to detect RNA fusions and copy number variants. . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Bioinformatics Raw sequencing output was transferred from the sequencing instrument to a bioinformatics server running Ubuntu 18.04LTS. A pre-supplied Docker image (The TruSight Oncology 500 pipeline; Illumina, San Diego, CA, USA) was used to generate TMB and Microsatellite Instability (MSI) calls. The pipeline consists of several steps. Initially, raw bcl files are converted to sample-specific FASTQ files as specified by the sample index. FASTQ files were then aligned against the hg19 reference genome using Isaac 4, local realignment to indels was performed, paired-end reads were stitched together, followed by variant calling with the somatic sample caller Pisces. Germline variants were filtered using a proprietary database, then the called variants were annotated to identify synonymous and nonsynonymous variants. Actual coverage of the panel compared to the reference coverage was computed and TMB was calculated based on the number of synonymous and nonsynonymous mutations detected divided by the size of the panel successfully sequenced. Small variants were exported from the TSO500 pipeline and annotated using VEP, then converted using vcf2maf and imported into the maftools module of R/Bioconductor. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint TMB calls for whole genome sequenced control data were carried out using the Genomics England v3 pipeline for calling tumour-normal pairs and used to compare to calls from the TSO-500 pipeline. In brief, this pipeline utilised Isaac v3 to align sequence data to the hg19 genome, followed by copy number variant calling using Canvas and structural variant calling using Manta. CNV calls for the TSO500 files were obtained using the Craft copy number caller set in somatic tumour only mode. Overlaps were computed using bedtools. SV calls for the TSO500 files were obtained using the Manta structural variant caller set in tumour only mode with a custom modification to the C++ code of the Manta SV caller to enable detection with less read support and on amplicon sequencing data. SV overlaps were computed using bedtools. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint DNAquality metrics In total, 108 samples were profiled using the assay, with a median sample age of 2 years (range 4 months-10 years). All samples were from FFPE blocks. This input for all assays is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint Precision of control calls In order to understand the ability of the assay to detect low variant allele frequencies (VAF), assessment of the VAF was performed for two known low VAF mutations in the HD753 cell line (Figure 4). This cell line has validated mutations in AKT1 (E17K, chr14: 105246551C>T) with a VAF of 0.05 and in PIK3CA (E545K, chr3: 178936091G>A) with a VAF of 0.056. The same control was run across 12 runs, with AKT1 median VAF=0.059 (IQR 0.037-0.072) and PIK3CA median VAF=0.036 (IQR 0.033-0.0493). Copy number calls A subset of 24 sample underwent copy number calling with the Craft pipeline. A variety of copy number gains and losses were detected in the 520 genes profiled on the TSO500 panel. The HD753 control was used to determine whether the observed copy number calls Structural variant calls The HD753 control is known to have a variety of structural variants including a SLC34A2/ROS fusion (VAF = 5.6%) and CCDC6/RET fusion (VAF = 5.0%). With use of a custom pipeline there was evidence for detection of both fusions: 7/506 reads supported the SLC34A2/ROS fusion and 5/498 reads supported the CCDC6/RET fusion. A variety of structural variants were observed in the tumour cohort. In addition, long indels were successfully detected by the Manta pipeline, specifically a 14bp deletion in EGFR (NM_005228.5:c.2235_2249del) known to be present in the HD753 control. Tumour mutational burden (TMB) & Microsatellite instability (MSI) TMB calling was successfully performed in 107/108 samples. The one failure was a very poor sample quality that failed hybridisation. There was good correlation between TMB determined by TSO500 and WGS (R 2 =0.9, Figure 6). The median TMB was 8. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint 311 muts/mb (range 289-325) in the HD753 control, a variance in TMB score of +/-5%. Comparison to The Cancer Genome Atlas (TCGA) tumour cohorts was performed in mafTools and is shown in Figure 8. For microsatellite instability (Figure 9), the threshold for classification as MSI-H was >10% of microsatellite sites being unstable. Using this threshold, both known somatic mismatch repair mutant (MLH1 and MSH6) cancers were MSI-H with 55% and 67% of sites being unstable respectively. Reassuringly, the POLE mutant cancer had 2% of MSI sites being unstable, meaning it was microsatellite stable (MSS) as is typical in POLE mutant cancer. RNA fusions RNA fusion analysis was carried out on 13 samples, of which 6 had known fusions. Fusions were detected between ETV6/NTRK3 (3 samples), RBPMS/NTRK3 (1 sample), EML4/ALK (1 sample), and TG/RET fusion (1 sample). All fusions that had previously been identified by FISH were detected using this methodology. A fusion was detected in one sample between ETV6/NTRK3 that had not been identified via FISH, however the fusion was supported by 12,627 reads in the sequencing run, which we felt was unlikely to be a false positive and therefore labelled it as a true fusion. Clinical actionability In order to recover as many clinically actionable variants as possible, mutational calls were is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint PierianDx and OncoKDM pipelines were not directly comparable in this study because of the differing inputs (FASTQ for PierianDx, VCF for OncoKDM) but the combination of both platforms provided a comprehensive variant overview. Conclusions We utilised the Trusight Oncology 500 assay in order to understand its utility and accuracy in determining both the tumour mutational burden and druggable mutation calls in cancer. One of the key challenges with patient testing (23) is the ability to take a patient biopsy sample, with limited input material and produce sequencing data and mutational calls of sufficient quality in order to make decisions on target selection and drug therapy (24). The assay was designed in its first iteration to measure tumour mutational burden as a surrogate marker for response to anti-PD1 immunotherapy as multiple studies have shown a correlation between TMB and response to this type of therapy (8,13). The TSO500 assay performs well in this respect with accurate measurement of TMB when compared to whole genome sequencing. Taking a threshold of 10 mut/mb as "TMB-high" (i.e. that which would have benefit for immunotherapy), we found that the TSO500 assay was able to classify samples with 100% accuracy. The precision of the calls varies at the extremes of TMB value, undoubtedly as a factor of panel size in calling TMB at extremely high levels. We conclude that the TSO500 pipeline is usable in clinical determination of TMB status across a range of clinical sample types and DNA inputs. We successfully detected microsatellite instability in all samples that were known to be MSI-H using TSO500. MSI detection using NGS has been shown to be feasible (25) previously using a variety of software solutions, usually relying on off-target reads (26), but other assays have used dedicated MSI probes (like the TSO500). We have found that the performance of this approach is variable, as the probes are vulnerable to drop out in FFPE samples. We propose that TMB instead may be a good surrogate biomarker for MSI, as a range of 30-80 mut/mb is typically seen in MSI tumours as opposed to MSS POLE/POLD1 tumours which typically have greater than 150 mut/mb. A key requirement for clinical specimens is the ability to process low-input specimens as well as the ability to detect the low variant allele frequencies (VAF) associated with these specimens (27). Reassuringly, we found that the TSO500 assay performed well at its recommended input concentration and also below these levels. Within our control samples with known VAF (of approx. VAF=0.05), we determined that there was good precision and . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint reproducibility with minimal variability. Another advantage to tolerance of low sample input is the possibility of using input levels seen in circulating tumour DNA (ctDNA) which are typically 1 ng/ml plasma in most cancers. This would allow derivation of blood TMB (28), which has been shown to be a better biomarker of response in PD-1/PD-L1 inhibitor therapy. The assay is also performed at sufficiently high read depth to allow calculation of clonal TMB (29,30), another marker associated with more accurate identification of potential response to immunotherapy. In terms of identifying druggable mutations for targeted therapy selection, the TSO pipeline presents an attractive platform especially when coupled with a clinical annotation engine such as the two used here (OncoKDM and PierianDx CGW (31)). We found good correlation between mutations detected in whole genome sequencing experiments, and the identification of druggable mutations was made straightforward by the use of integrated clinical pipelines to produce reproducible data. Copy number variations, especially amplifications, represent important therapeutic targets. The TSO500 assay detected the known amplifications in a control sample meaning that patients can potentially undergo therapeutic targeting. A unique advantage of the TSO500 system is the ability of a partner targeted RNA-seq assay that can detect RNA fusions. We found that the assay reliably detected NTRK (32), ALK (33), and RET (34) fusions that had previously been identified by FISH, as well as a novel fusion not previously detected using other technologies. Intriguingly, we also successfully detected known fusions at the DNA level de novo in the HD753 control sample, suggesting that this methodology may also be valid for future use, although DNA-based fusion calling has a high false negative rate. Fusion genes represent good drug targets, and a number of novel agents (19,32) have been shown to be active against fusion genes. Detection of circulating RNA for these fusion genes may also be possible (35) using this assay and could be explored further. The UK 100,000 Genome project has recently completed, and analysis and reporting is ongoing. The use of whole genome sequencing for tumour-normal pairs using fresh frozen material still has significant challenges from a cost perspective as well as the practicalities of obtaining fresh-frozen tissue over readily available paraffin embedded material. The TSO500 assay costs approximately one third the price of a whole genome sequencing assay, requires no germline DNA control, allows RNA fusion detection, and can be implemented on benchtop sequencers. Its main disadvantages include more laborious library preparation and enrichment chemistry that is vulnerable to drop out. In conclusion, we believe that the TruSight Oncology 500 assay offers a cost-effective, accurate, pan-cancer assay that can derive SNP, CNV, and gene fusion information across . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint the majority of cancers using a standardised pipeline and therefore is suitable for routine use in precision oncology as a comprehensive genomic profiling solution. . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted February 4, 2020. ; https://doi.org/10.1101/2020.02.01.20019992 doi: medRxiv preprint	2020-03-19T19:41:27.782Z	2020-02-04T00:00:00.000	{ "year": 2020, "sha1": "d0e209e6490d617ce5bfd44acb7a52d08bce3f9e", "oa_license": "CCBY", "oa_url": "https://link.springer.com/content/pdf/10.1007/s40291-020-00462-x.pdf", "oa_status": "GREEN", "pdf_src": "ScienceParsePlus", "pdf_hash": "d65626e462ae59b5925538c5a25837ade4cb566c", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Biology" ] }
271742372	pes2o/s2orc	v3-fos-license	A View in the Dark: Two Cases of Acute Esophageal Necrosis in the Setting of Diabetic Ketoacidosis Acute esophageal necrosis (AEN), also known as Gurvits syndrome, is a rare and potentially life-threatening condition characterized by necrosis of the esophageal mucosa. Acute esophageal necrosis is often associated with critical conditions, such as myocardial infarction, diabetic ketoacidosis (DKA), coronavirus disease 2019 (COVID-19) infection, or post-surgical complications. Patients typically present with nausea, hematemesis, acute dysphagia, and melena. Given its high mortality rate, prompt detection with upper endoscopy and early initiation of treatment are crucial. Most cases of Gurvits syndrome are managed conservatively using intravenous fluids, proton pump inhibitors, and antibiotics. Herein, we present a case series of AEN in the setting of DKA. Both patients received supportive care and were discharged in a stable condition. Introduction Acute esophageal necrosis (AEN), also known as "black esophagus," Gurvits syndrome, or acute necrotizing esophagitis, is a rare condition characterized by necrosis of the esophageal mucosa, leading to a blackened appearance of the esophagus. 1,2Typically, AEN is associated with poor prognosis, with mortality rates ranging from 32% to 36% due to comorbidities.4][5] Although the exact cause of AEN remains unclear, it is believed to be multifactorial and primarily managed medically.][8] Patients with AEN may present with symptoms such as upper gastrointestinal bleeding (GIB), abdominal pain, dysphagia, nausea, vomiting, fever, and syncope. 2,6,9,10Early recognition and appropriate management are crucial for improving the outcomes of AEN patients.Supportive care is the mainstay of treatment for most AEN cases, with surgery reserved for acute complications such as esophageal perforation and mediastinitis. 6Herein, we describe a case series of AEN in the setting of DKA.Both patients were successfully treated with proton pump inhibitors (PPIs), carafate suspension, intravenous (IV) fluids, and insulin infusion.Acute esophageal necrosis should be considered in the differential diagnosis of upper GIB in patients with DKA. Case 1 A 52-year-old male with a medical history of insulin-dependent diabetes mellitus (IDDM) and hypertension was brought to the emergency department (ED) for evaluation of bizarre behavior and vomiting.The patient stated that he was released from jail earlier in the day and that he missed his insulin doses.He denied having fever, chills, dysphagia, odynophagia, acid reflux, chest pain, shortness of breath, palpitations, abdominal pain, melena, hematochezia, recent weight loss, polyuria, polyphagia, polydipsia, or dysuria.In the ED, the patient appeared confused but followed verbal commands.The initial vital signs were notable for elevated blood pressure (140/72 mm Hg), tachycardia (125 beats per minute), and tachypnea (28 breaths per minute).On further examination, the patient was alert and oriented to self and place only.There was no facial asymmetry or tongue laceration, and the pupils were dilated but reactive to light and accommodation.The remainder of the examination was unremarkable. Triage blood tests were significant for hyperkalemia, hyperglycemia, elevated creatinine and blood urea nitrogen, elevated alkaline phosphatase, elevated serum lipase, and leukocytosis (Table 1).Arterial blood gas analysis revealed a severe metabolic acidosis and elevated lactic acid.Additional tests revealed moderate ketones with an anion gap of 27.5.An electrocardiogram (EKG) revealed atrial fibrillation with a rapid ventricular rate but no acute ischemic pattern.The patient was treated for hyperkalemia and admitted to the medical intensive care unit (ICU) for the management of DKA.The heart rhythm converted to sinus rhythm after 10 mg of dose of IV diltiazem and 150 mg of amiodarone.The patient was managed with a DKA protocol comprising insulin infusions, IV fluids, and electrolyte fluid solutions. The patient's ICU course was complicated by melena and acute blood loss anemia, necessitating a gastroenterology consultation.Bedside esophagogastroduodenoscopy (EGD) revealed diffuse esophageal plaques concerning for candidiasis, ulcerated esophageal necrosis, and a small hiatal hernia (Figure 1).The histopathology of random esophageal biopsies revealed necroinflammatory material and ulcer debris associated with abundant fungal microorganisms, which were morphologically compatible with Candida species.Overall, the findings were consistent with AEN in the setting of DKA and Candida esophagitis.Immunohistochemical staining for cytomegalovirus (CMV) and herpes simplex virus (HSV-1 and HSV-2) was negative.The patient received supportive care with buprenorphine, proton pump inhibitors (PPIs), carafate suspension, fluconazole, IV fluids, and insulin infusion.A relook endoscopy after 6 months showed normal esophageal mucosa (Figure 2). Case 2 A 60-year-old male with a history of type 2 diabetes mellitus, hypertension, chronic constipation, and Parkinson's disease (on carbidopa/levodopa and pramipexole) presented to the ED complaining of acute-onset abdominal pain and distension.The abdominal pain was vaguely localized, sharp in nature, and rated 9/10 at its peak.It was associated with nausea, constipation, and shortness of breath.The patient denied recent illness, melena, hematochezia, chest pain, palpitations, polyuria, polyphagia, or recent hypoglycemic events.He denied alcohol, tobacco, or illicit substance use.The patient also reported compliance with diabetes mellitus medications (linagliptin/metformin and glipizide) and a hypocaloric diet. In the ED, the patient was lethargic and in acute distress due to pain.The vital signs were significant for transient hypotension (99/57 mm Hg) and tachycardia (103 beats per minute).He was afebrile and saturated 97% in room air.Upon examination, the patient spoke in short sentences with Kussmaul breathing, but the lungs were clear on auscultation.The abdomen was distended and diffusely tender on palpation.The patient had diminished bowel sounds in the lower abdominal quadrants, but there were no signs of trauma, peritonism, flank ecchymosis, or palpable organomegaly.The patient was alert and oriented to self, time, and place.He had a resting tremor in the right upper extremity and cogwheel rigidity.The rest of the examination was unremarkable. Owing to concerns for DKA, a venous blood gas was ordered, which showed low pH, elevated lactic acid, low bicarbonate level, and an anion gap of 24.Triage blood tests revealed leukocytosis with bandemia, thrombocytopenia, hyperglycemia, low serum bicarbonate level, moderate ketones, elevated creatinine and blood urea nitrogen, and hyperbilirubinemia (Table 2).The EKG showed sinus tachycardia and left axis deviation, with no evidence of acute ischemic changes.Contrast-enhanced computed tomography (CT) of the abdomen and pelvis revealed hepatic steatosis, fecal impaction in the rectosigmoid region, and focal colitis in the sigmoid colon.The patient was administered 3 L of IV fluids and the serum lactic level decreased to 6 mmol/L.The patient was admitted to the ICU for septic shock and DKA.The patient was treated with vasopressor support, insulin infusion, electrolyte solutions, and broad-spectrum antibiotics. Overnight, the patient developed peritonitis and melenic stools.A nasogastric tube (NGT) was placed on low intermittent suction for gastric decompression, and a rectal tube was inserted because of persistent melenic stools.A repeat abdominal CT scan showed worsening colitis without signs of bowel perforation.The surgical service was consulted due to concerns for acute mesenteric ischemia, but they found no objective evidence of ischemia on imaging.They disimpacted the patient at the bedside and recommended an aggressive bowel regimen.Bedside EGD revealed a "black esophagus," consistent with AEN (Figure 3), and the patient was managed conservatively with PPIs, carafate suspension, and insulin infusion.On day 4 of admission, the patient's stool became yellowish, and the hemoglobin level remained stable.A clear liquid diet was initiated and advanced as tolerated, and the patient was discharged on day 11 of admission in stable condition.A relook endoscopy after 8 months showed normal esophageal mucosa (Figure 4). Discussion Acute esophageal necrosis is a rare condition characterized by a striking black discoloration of the esophageal mucosa, typically seen during EGD.The necrosis usually starts in the distal esophagus and abruptly transitions to normal mucosa at the gastroesophageal junction (GEJ). 1,2][10][11] The cause of AEN is multifactorial, with risk factors including DKA, sepsis, malnutrition, chronic renal disease, cardiovascular disease, alcohol abuse, and male gender. 2,9Acute esophageal necrosis is thought to result from a combination of esophageal ischemia, impaired mucosal barrier function, and reflux of gastric contents, leading to necrosis of the esophageal mucosa. 2 In this series, we share 2 cases of AEN in the setting of DKA managed with supportive care.Acute esophageal necrosis is an extremely rare complication of DKA that may result from poor tissue perfusion, mucosal barrier dysfunction, and corrosive injury.The profound osmotic diuresis in DKA limits the blood supply to watershed areas such as the distal esophagus, predisposing them to ischemia and necrosis. 11In addition, diabetes mellitus promotes atherosclerosis, which potentiates tissue ischemia in affected patients. 6Signs and symptoms include nausea, vomiting, coffee-ground emesis, acute dysphagia, abdominal pain, and melena. 2,6,9,11Patients may also experience odynophagia, eructation, pyrosis, and regurgitation. 2,9,12riage blood tests may show acute blood loss anemia, leukocytosis, and hyperlactatemia. 5,13In severe cases, physical examination may show signs of sepsis or shock, characterized by tachycardia, tachypnea, hypotension, and fever. 12bdominal guarding and tenderness may also be observed on examination.Complications of AEN include esophageal stenosis, strictures, perforation, mediastinitis, peristaltic abnormalities, and abscess formation. 7,14he diagnosis of AEN is usually based on endoscopic findings of circumferential black esophageal mucosa that abruptly terminates at the GEJ. 11,13,15Other possible endoscopic findings include different bleeding sites and presence of gastric and duodenal ulcers. 13The CT scans will show thickening of the distal esophagus with or without esophageal perforation secondary to the associated inflammation. 11owever, it is important to note that CT may show nonspecific changes depending on the symptom onset, disease course, and involvement. 11An important diagnostic finding to be aware of is signs of occlusions of the splenic or left gastric arteries which can also impact the blood supply to the distal esophagus. 8,11,15Although the causes of AEN vary, the fundamental pathophysiology and histological appearance are the same.The foundation is based on the hemodynamic instability that is caused by the underlying insult to the body, which compromises the blood flow to the distal esophagus. 1istologically, AEN presents with mucosal necrosis with the absence of viable epithelium and a brisk increase in inflammatory infiltrate. 1 Depending on the insult and the time frame of hypoperfusion, the necrosis can extend into the muscularis propria.Although helpful at times, a biopsy is not essential for diagnosing AEN. The management of AEN is directed at correcting the underlying insult and hemodynamic resuscitation to ensure adequate blood flow to the distal esophagus. 8,11,15upportive measures include nil per os (NPO) status, IV hydration, IV PPIs, and short-term parenteral nutrition. 16,17n addition, sucralfate reduces local insult to the esophagus from gastric acid, thus aiding the healing process. 16Acute esophageal necrosis associated with DKA also requires electrolyte repletions and insulin infusions to achieve euglycemia.A prolonged course of oral PPIs even after symptom resolution prevents esophageal stricture and stenosis. 12nteral tube placement should be avoided, given the increased chances of perforation. 10In case 2, an enteral tube was inadvertently placed and should have been avoided.Prophylactic antibiotics should be limited to patients with signs of sepsis, septic shock, esophageal perforation, or mediastinitis. 7,16In cases of co-existing infectious esophagitis, a specific antimicrobial therapy should be started to hasten healing. 6The patient described in case 1 had comorbid Candida esophagitis, which was effectively treated with IV fluconazole.Surgical intervention is usually reserved for patients with esophageal perforation, or mediastinitis. 10In these cases, self-expandable stents (SEMS) or intrathoracic drainage systems have proved effective. 6Adrenaline sclerotherapy can be employed in patients with active bleeding. 12Balloon tamponade with a Sengstaken-Blakemore tube is contraindicated due to the increased risk of esophageal perforation. 6,10,12Repeat EGD is warranted to monitor for improvement within 1 month as the esophageal mucosal recovery occurs over a period. 6,12ur patients underwent relook endoscopies a few months after their hospitalization, and both studies were unremarkable.Endoscopic balloon dilation is the primary treatment for strictures, although some patients may require multiple dilation sessions and, in severe cases, esophagectomy and esophageal bypass may be necessary. 6 Conclusion Acute esophageal necrosis is an infrequent complication of DKA and is characterized by extensive black discoloration of the esophageal mucosa.Acute esophageal necrosis is an endoscopic diagnosis, and prompt treatment of the underlying etiology can avert further complications such as esophageal perforation or mediastinitis.The patient described in case 1 was admitted for DKA requiring insulin therapy and aggressive hydration.This condition along with the atrial fibrillation potentially predisposed the patient to developing AEN.Similarly, the patient in case 2 was admitted for DKA and septic shock.Although DKA is the likely cause of his AEN, sepsis-induced hypotension may have also contributed to the limited blood supply to the distal esophagus. Figure 2 . Figure 2. Endoscopic images for case 1 (6 months after discharge) showing normal esophageal mucosa and a small food bolus in the distal esophagus. Figure 3 . Figure 3. Endoscopic images for case 2 (during hospitalization) showing diffuse dark discoloration and ulceration of the upper esophagus.The cardia/fundus, duodenal bulb, and second portion of duodenum were normal. Table 1 . Pertinent Admission Laboratory Values for Case 1 Compared With the Reference Ranges. Figure 1.Endoscopic images for case 1 (during hospitalization) showing diffuse esophageal plaques, concerning for candidiasis, and ulcerated esophageal necrosis.The stomach and duodenum appeared normal. Table 2 . A View in the Dark: Acute Esophageal Necrosis in the Setting of Diabetic Ketoacidosis.	2024-08-08T06:17:44.112Z	2024-01-01T00:00:00.000	{ "year": 2024, "sha1": "8c1c2040f2e4a19f30a178f75b09e07ac9d05ade", "oa_license": "CCBYNC", "oa_url": null, "oa_status": null, "pdf_src": "PubMedCentral", "pdf_hash": "102c7be776ea11c58b1128e279e4a40c90063bc4", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
44844773	pes2o/s2orc	v3-fos-license	Matrix Metalloproteinase 1 Interacts with Neuronal Integrins and Stimulates Dephosphorylation of Akt* Several studies have demonstrated that matrix metalloproteinases (MMPs) are cytotoxic. The responsible mechanisms, however, are not well understood. MMPs may promote cytotoxicity through their ability to disrupt or degrade matrix proteins that support cell survival, and MMPs may also cleave substrates to generate molecules that stimulate cell death. In addition, MMPs may themselves act on cell surface receptors that affect cell survival. Among such receptors is the α2β1 integrin, a complex that has previously been linked to leukocyte death. In the present study we show that human neurons express α2β1 and that pro-MMP-1 interacts with this integrin complex. We also show that stimulation of neuronal cultures with MMP-1 is associated with a rapid reduction in the phosphorylation of Akt, a kinase that can influence caspase activity and cell survival. Moreover, MMP-1-associated dephosphorylation of Akt is inhibited by a blocking antibody to the α2 integrin, but not by batimastat, an inhibitor of MMP-1 enzymatic activity. Such dephosphorylation is also stimulated by a catalytic mutant of pro-MMP-1. Additional studies show that MMP-1 causes neuronal death, which is significantly diminished by both a general caspase inhibitor and anti-α2 but not by batimastat. Together, these results suggest that MMP-1 can stimulate dephosphorylation of Akt and neuronal death through a non-proteolytic mechanism that involves changes in integrin signaling. The matrix metalloproteinases (MMPs) 1 represent a family of endopeptidases named for their ability to degrade extracellular matrix components. These proteinases play a role in tissue remodeling associated with both development and disease (1)(2)(3)(4)(5)(6)(7). In the central nervous system MMPs are released by cells including activated astrocytes, microglia, and neurons (8,9). Previously, we and others have shown that MMPs can be toxic to neurons in vitro (10 -13). Although such toxicity may follow from extracellular matrix destruction (14), it is also reasonable to consider the non-mutually exclusive possibility that other mechanisms are involved. One extracellular matrixindependent mechanism by which MMPs may function involves their ability to cleave non-matrix proteins and thereby generate potential cell surface receptor signaling ligands (15)(16)(17)(18)(19). Another mechanism by which MMPs could affect cell survival might include direct effects on cell surface receptors. For example, MMP-9 binds to CD44 (20), MMP-2 binds to integrin ␣ v ␤ 3 (21), and pro-MMP-1 binds to integrins ␣ 2 ␤ 1 and ␣ 1 ␤ 1 (22). Such binding interactions may facilitate activation of the proenzyme, localize enzyme activity, disrupt cell matrix interactions to promote cell motility, and/or mediate internalization of the protease. Cell surface receptor binding by an MMP may also alter intracellular signaling. It has been shown that the snake venom metalloproteinase jararhagin can signal via ␣ 2 ␤ 1 on fibroblasts in a manner that is insensitive to inhibition of proteinase activity (23). In addition, we have shown that MMP-1 can stimulate protein release from neurons and monocytes in a manner that is insensitive to inhibition of its enzymatic activity (24). These results fit into a growing literature suggesting that enzymes can stimulate biological effects in a manner that is independent of their catalytic activity (25)(26)(27). Although protein release has been examined as an end point of proteinase-associated ␣ 2 ␤ 1 signaling, other potential endpoints have not been investigated. Of interest to the question of cytotoxicity, ␣ 2 ␤ 1 signaling has been linked to G i protein-dependent cell death (28), presumably via a seven transmembrane CD47-␣ 2 ␤ 1 complex (29). In addition, stimulation of ␣ 2 ␤ 1 with bovine collagen has been linked to protein phosphatase activation and the dephosphorylation of Akt (30). Akt is a critical effector of cell survival with targets that include Bad, caspase-9, and GSK-3␤ (31,32). Reduced Akt activity has also been implicated in apoptosis mediated by diverse agents (33). In the present study we have investigated the possibility that MMP-1stimulation of neuronal cultures leads to the dephosphorylation of Akt. We have also examined the possibility that MMP-1 is linked to caspase inhibitor-sensitive cell death, a potential consequence of Akt dephosphorylation (34,35). were released through gentle shaking and then plated onto flat-bottomed plates, in which they were maintained for 5 days before experiments were conducted. The purity of cultures was established by immunostaining for microtubule-associated protein-2 (Dako, Carpinteria, CA). Immunostaining-Neuronal cultures were grown on glass coverslips and fixed in 4% paraformaldehyde for 10 min at room temperature. The cells were washed three times with PBS, 5 min each, then blocked for 1 h in a PBS solution containing 4% donkey serum (Sigma) and 0.3% Triton X-100. This was followed by 4°C overnight incubation with 1:200 dilution of the primary antibody (mouse anti-human ␣ 2 ␤ 1 or rabbit anti-human ␣ 2 , both from Chemicon) diluted in 3% donkey serum. Normal mouse IgG was used as a negative control for ␣ 2 ␤ 1 staining, and secondary antibody alone was used as a control for both ␣ 2 ␤ 1 and ␣ 2 staining. Cells were washed 3 times in PBS for 5 min each and incubated for 2 h at room temperature in secondary antibodies diluted 1:500 in PBS (Alexa 488-conjugated anti-mouse from the Jackson Laboratory or goat anti-rabbit from Molecular Probes). Cells were again washed three times in PBS before mounting onto slides with Mowoil (Calbiochem). Slides were then analyzed using confocal (␣ 2 ␤ 1 ) or fluorescent (␣ 2 ) microscopy. MMPs and Inhibitors-Purified MMP-1 was purchased from Oncogene Research Products (San Diego, CA) and Chemicon International (Temecula, CA). The proteinase was separated in aliquots and stored at Ϫ70°C upon its arrival. MMP-1 from Oncogene Research Products was purified from human rheumatoid synovial fibroblasts by affinity chromatography, ion exchange chromatography, and gel filtration, and MMP-1 from Chemicon was purified from transfected P2AH2A cells and also by ion exchange and affinity chromatography. The inactive pro-MMP-1 mutant, in which Glu-200 was changed to Ala (pro-MMP-1(E200A)), was overproduced in and purified from Escherichia coli according to methods previously described for the wild type enzyme (37). Except where indicated MMP-1 from Oncogene Research Products was used in the experiments shown. As determined by Western blot, MMP-1 preparations from Chemicon contained both pro-and activated MMP-1, whereas that from Oncogene typically contained a majority of the pro-enzyme. Of note, 50 -100 ng/ml MMP-1 (ϳ1-2 nM) was used in experiments since previous studies showed that interleukin-1␤-stimulated astrocyte supernatants contained amounts in this range and that neurons were responsive to these concentrations of the proteinase (12,24). Purified active MMP-9 was purchased from Oncogene Research Products and was a recombinant preparation made in mammalian cells. The ␣ 2 integrin blocking antibody, which has previously been shown to block signaling through ␣ 2 ␤ 1 (30), was obtained from Serotec (MCA743). Okadaic acid was obtained from Sigma. The metalloproteinase inhibitor batimastat was obtained from British Biotech, and the broad spectrum caspase inhibitor ZVAD was obtained from Biomol Research Laboratories. Preparation of Cell Extracts-Neuronal extracts were prepared from cells that had been cultured on 12-well plates as follows. Cells were washed in cold phosphate-buffered saline, and 200 l of lysis buffer (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.1%SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.2 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol, and 1 mM sodium vanadate) was then added to each well (ϳ10 5 cells). After 5 min the cell suspension was scraped and placed into an Eppendorf tube. The suspensions were then sonicated for 10 s and spun at 4°C in a desk top Eppendorf centrifuge at 14,000 rpm for 20 min. The supernatants were then saved for analysis by Western blot or ELISA. Protein concentrations were determined using the BCA protein assay reagent kit (Pierce). Immunoprecipitation Experiments-After a 1-h incubation with MMP-1 that, as determined by Western blot, contained both pro-and active length MMP-1, cultured cells were washed twice in PBS and then lysed in a detergent-containing buffer (50 mM Tris, 150 mM NaCl, 0.1% SDS, 1% IGEPAL, 1 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride). Lysates were spun, and supernatant was incubated 1:1 with pre-swol-len protein G-Sepharose (Amersham Biosciences) for 2h at 4°C, a step taken to eliminate proteins in the lysate that may bind non-specifically to the protein G. The mix was subsequently spun, and the supernatant was incubated at 4°C overnight with an anti-MMP-1 that recognizes pro-and active forms (Chemicon, AB8105), an isotype matched control antibody (Chemicon, AB5320), anti-␣ 2 ␤ 1 (Chemicon, MAB1998Z), or anti-␤ 1 (Chemicon MAB2252). The mix was next incubated for 2 h with protein G-Sepharose, and after 3 washes the precipitate was analyzed by Western blot using a primary antibody to pro-and active MMP-1 (R&D Systems MAB901), ␣ 2 ␤ 1 (Chemicon 1998Z), ␣ 2 (Chemicon AB1936), or ␤ 1 (MAB2252) as indicated. All incubations (antibody and protein G) were performed on a rotary table at 4°C, and all spins were performed using a desktop Eppendorf centrifuge at 4°C for 5 min at maximum speed (14,000 rpm). ELISA-ELISA for phospho-and total Akt was performed using kits (KHO0111 and KHO0101) from BIOSOURCE International (Camarillo, CA). The phospho-Akt specific ELISA detects Akt, which is phosphorylated on serine 473. Both ELISAs were performed according to the manufacturer's instructions. Although equal sample volumes were added to wells, correction for differing protein concentrations was made before calculation of relative Akt. Determination of Neuronal Cell Death-At the time of experimental treatments, the culture medium was replaced with Locke's buffer containing 154 mM NaCl, 5.6 mM KCl, 2.3 mM CaCl 2 , 1 mM MgCl 2 , 3.6 mM NaHCO 3 , 5 mM glucose, and 5 mM Hepes (pH 7.2). Cell death in each case was monitored by trypan blue exclusion 15 h after treatment of cells as described previously (12,36). Neuronal cell counts were determined from five fields at predetermined coordinate locations. Each field was photographed and coded, and both live and dead cells were subsequently counted. A minimum of 100 cells was typically counted in each field. Each experiment was conducted in triplicate wells so that 15 fields were counted. Results were normalized so that control means were equal to 1. In these experiments MMP-1 was used at 50 ng/ml, ZVAD at 100 M, pertussis toxin at 100 ng/ml, and batimastat at 20 ng/ml and 500 ng/ml (ϳ1 M). ZVAD and pertussis toxin were added in advance of MMP-1 (2 h for pertussis toxin and 30 min for ZVAD). For batimastat studies, MMP-1 was added to medium containing the inhibitor, and the mixture was applied to the cultured cells. Measurement of Mitochondrial Membrane Potential Activity-Six hours after the treatment of neuronal cultures with the indicated stimulus in Locke's buffer containing 154 mM NaCl, 5.6 mM KCl, 2.3 mM CaCl 2 , 1 mM MgCl 2 , 3.6 mM NaHCO 3 , 5 mM glucose, and 5 mM Hepes, pH 7.2, cells were washed in buffer and then incubated for 30 min at 37°C in a 5% CO 2 incubator in the presence of 10 M 5,5Ј,6,6Ј-tetrachloro-1,1Ј,3,3Ј-tetraethylbenzimidazolylcarbocyanine iodide (JC-1) from Molecular Probes, Eugene, OR. Cells were then extensively washed again in Locke's buffer. In the absence of mitochondrial damage membrane potential is high, and JC-1 forms aggregates that give a red fluorescence. When mitochondria are damaged, potential is lost, and JC-1 remains in monomers that give a green fluorescence. Optical measurements were, therefore, acquired with excitation at 485 nm and emission at 527 nm and 590 nm. The levels of fluorescence at both emission wavelengths were then quantified, and a ratio of measurements was assessed. Statistics-Data are given as means Ϯ S.E. After normality testing data were evaluated by ANOVA or Kruskall Wallis, and pairwise comparisons were made by Tukey or Mann-Whitney testing as appropriate. RESULTS Human Neurons Express ␣ 2 ␤ 1 -Previous studies show that pro-MMP-1 binds to the I domain of the ␣ 2 integrin on keratinocytes (22). To determine whether this integrin is expressed on human neurons, we performed immunocytochemistry and Western blot analysis of lysates from these cells. As shown in Fig. 1, a-d, neurons showed immunoreactivity for ␣ 2 ␤ 1 and ␣ 2 . Staining was prominent along the cell body and was also present along cell processes. No staining was observed when normal mouse IgG or secondary antibody alone were used (not shown). In addition, expression of both ␣ 2 and ␤ 1 was confirmed by Western blot analysis of cell lysates (Fig. 1, d and e). MMP-1 Co-immunoprecipitates with Neuronal ␣ 2 ␤ 1 -To determine whether MMP-1 associates with neuronal integrins, as it does with those expressed on keratinocytes (22, 38), we examined immunoprecipitates made using an antibody to ␣ 2 ␤ 1 and to ␤ 1 for the presence of MMP-1. Results are shown in Fig. 2a. The antibody used to make the precipitate is shown at the top, whereas that used to probe the blot is shown at the side. The upper panel is an image obtained after a relatively short exposure of film to membrane, whereas the lower panel is an image showing the immunoprecipitate lanes as they appeared after a longer exposure. In this experiment purified MMP-1 that had undergone some auto-activation was used as a control (shown in the upper panel only). Although the control was overloaded, two bands representing pro and active length MMP-1 can be appreciated. It can also be appreciated that, although cultures from which the precipitates had been made were treated with MMP-1 that contained near equal amounts FIG. 2. MMP-1 co-immunoprecipitates with ␣ 2 ␤ 1 . For the data shown in a immunoprecipitates were made using anti-␣ 2 ␤ 1 or anti-␤ 1 as indicated at the top and analyzed by Western blot using a primary antibody to MMP-1. The lower panel shows immunoprecipitate lanes from an image obtained following a longer exposure of the membrane. 500 ng of purified MMP-1 was run in the control lane for comparison (shown in the upper panel only). The upper band represents the pro form and the lower band represents active length enzyme. For the data shown in b, immunoprecipitates were made using anti-MMP-1 and analyzed by Western blot using a primary antibody to ␣ 2 or ␤ 1 as shown at the right. Neuronal lysates were run as controls. Immunoreactive proteins migrated according to their expected molecular masses (130 kDa for ␤ 1 and 165 kDa for ␣ 2 ). Western blot analysis of two separately prepared neuronal extracts also shows a band at 165 kDa that is immunoreactive for ␣ 2 (e) as well as a band at ϳ130 kDa that is immunoreactive for ␤ 1 (f). of both pro and active length forms and the blot had been probed with anti-MMP-1 that recognizes both forms, only pro length MMP-1 was detected. It is of course possible that active length MMP-1 can also associate with ␣ 2 ␤ 1 , but in our experiments its ability to do so was below the limits of detection. Of interest, the region of MMP-1, which is important to integrin binding, is not within the pro-domain but within the hinge and hemopexin domains, and yet it is also pro-MMP-1 that is detected in keratinocyte immunoprecipitates (22,38). As shown in Fig. 2b, immunoprecipitates that were made using anti-MMP-1 also contained both ␤ 1 and ␣ 2 . Neuronal extracts were used as controls in this figure. MMP-1 Stimulation of Neurons Is Associated with a Decrease in the Phosphorylation of Akt, and This Is Inhibited by a Blocking Antibody to ␣ 2 -We had previously shown that MMP-1 is toxic to human neurons, but the potential mechanisms were not investigated (12). Signaling through ␣ 2 ␤ 1 in other cell types has been linked to protein phosphatase activation and subsequent Akt dephosphorylation (30). Reduced activity of Akt has in turn been linked to cell death (31). We, therefore, measured levels of phospho-Akt in untreated and MMP-1-treated neuronal cultures. MMP-1 stimulation of human neurons was associated with reduced levels of phospho-Akt as determined by ELISA (Fig. 3a). Although MMP-1 from Oncogene was used in these experiments, similar results were obtained using a purified preparation from Chemicon. Also of note, the magnitude of the MMP-1-associated reduction in phospho-Akt was similar to that observed in fibroblasts after their stimulation with an established ␣ 2 ␤ 1 agonist (30). To ensure that the reduction we observed in phospho-Akt was not due to degradation, total Akt levels were also measured by ELISA and were unchanged (data not shown). To examine the potential involvement of an ␣ 2 integrin-containing complex, we also tested MMP-1 on cultures that were pretreated with a blocking antibody to ␣ 2 . Results demonstrate that anti-␣ 2 attenuated the effect of MMP-1 (Fig. 3b). Because ␣ 2 ␤ 1 has been linked to the activation of protein phosphatase 2A, we also tested the protein phosphatase inhibitor okadaic acid at a concentration relatively selective for protein phosphatase 2A for its effect on MMP-1-associated dephosphorylation of Akt. We observed that okadaic acid did reduce the ability of MMP-1 to reduce phospho-Akt while stimulating no significant increase on its own (50 ng/ml MMP-1, 74 Ϯ 10% that of control; 50 ng/ml MMP-1 plus 1 nM okadaic acid, 104 Ϯ 8% that of control, n ϭ 3). The MMP-1-stimulated Reduction in Phospho-Akt Is Independent of Proteolytic Activity-Although immunoprecipitation studies suggest that MMP-1 may directly bind to ␣ 2 ␤ 1 , it is also possible that MMP-1 can generate a ligand that signals through ␣ 2 ␤ 1 and/or other cell surface receptors to stimulate a decrease in phospho-Akt. Pro-MMP-1 undergoes auto activation, and it is difficult to rule out activation of the pro-enzyme after its addition to cultured cells. We, therefore, tested batimastat, an inhibitor of the catalytic activity of MMP-1, for its ability to affect MMP-1-related changes in phospho-Akt. The results are shown in Fig. 4a and suggest that the ability of MMP to reduce phospho-Akt may be independent of proteolysis. As an additional experiment to assess the question of whether MMP-1 was acting in a manner independent of proteolysis, we tested the inactive pro-MMP-1 mutant in which Glu-200 was changed to Ala (pro-MMP-1 (E200A)). A glutamic acid residue at position 200 allows for interaction of a water molecule with the active site zinc, which in turn is critical for substrate catalysis (1,39,40). As shown in Fig. 4b pro-MMP-1 E200A also affected a statistically significant decrease in phosphorylated Akt. MMP-9 Does Not Decrease Phospho-Akt-Like MMP-1, MMP-9 has been associated with neurotoxicity (10, 11). We, therefore, examined whether MMP-9 might also be associated with a reduction in phospho-Akt. In these experiments the active form of MMP-9 was used because it is this form of the enzyme that has been linked to cytotoxicity (11). Although MMP-1 was again associated with a significant reduction in phosphorylated Akt, levels in MMP-9-treated cells were similar to controls (Fig. 5). These data suggest that, at least for the cultures used and the time point examined in our studies the effect of MMP-1 is relatively specific. MMP-1 Is Associated with Decreased Phosphorylation of GSK-3␤-␣ 2 ␤ 1 signaling has also been linked to decreased phosphorylation of GSK-3␤ (30), a potential effect of reduced serine-9 phosphorylation by Akt (41). We, therefore, examined lysates from unstimulated, MMP-1 stimulated, and pro-MMP-1 E200A-stimulated cells for phosphoserine 9 GSK-3␤. As shown in Fig. 6, MMP-1 and pro-MMP-1 E200A were associated with a reduction in serine 9-phosphorylated GSK-3␤, as determined by Western blot. Although additional studies will be necessary to better define the effect of MMP-1 on GSK-3, these data are consistent with effects previously reported for engagement of ␣ 2 ␤ 1 by an alternate ligand (30) and, thus, lends FIG. 3. MMP-1 stimulation of neurons leads to a decrease in intracellular phospho-Akt, and this is inhibited by a blocking antibody to ␣ 2 . Human neurons were stimulated for 45 min with medium alone or medium containing 50 or 100 ng/ml MMP-1 or pretreated for 30 min with anti-␣ 2 (40 g/ml) and then stimulated with 50 ng/ml MMP-1. Extracts were then prepared and compared for phospho-Akt. Decreased phospho-Akt in extracts from MMP-1-treated cells was observed with ELISA, which allowed for the quantification of Akt in numerous samples. Data shown represent the mean plus S.E. for phospho-Akt, calculated in units/mg of protein and then expressed as percent control in 3-5 replicates. In a the difference between unstimulated and MMP-1-stimulated extracts is significant at p Ͻ 0.03 (Mann-Whitney test). Total Akt was also measured by ELISA and was not decreased in association with MMP-1 (data not shown), implying that treatment was associated with dephosphorylation of Akt rather than degradation. further support to the idea that MMP-1 signals through this integrin. MMP-1-associated Neuronal Death Is Attenuated by ZVAD but Not by Batimastat-Akt is thought to act as a pro-survival signal by inhibiting the activity of pro-apoptotic intracellular proteins including Bad. Overexpression of Akt will often promote cell survival and, perhaps more important with respect to the present study, stimuli that decrease Akt activity often induce apoptosis through a caspase-dependent mechanism (31). As shown in Fig. 7 we tested the possibility that MMP-1 stimulation of neurons might lead to cell death in a manner that would be affected by ZVAD, an inhibitor of caspases. To determine whether MMP-1-induced neurotoxicity was dependent on its enzymatic activity, MMP-1 was also tested in the presence of batimastat. ZVAD was associated with a statistically significant inhibition of MMP-1-associated cell death (Fig. 7a). Batimastat, however, did not block cytotoxicity either at a relatively low dose (Fig. 7a) or at a high dose (1 M, Fig. 7b). Because of the differential responsiveness of primary cell cultures, there is some difference in MMP-1-associated toxicity between the two experiments (compare Fig. 7a to 7b). As a further measure of toxicity, we also tested the ability of MMP-1 to affect mitochondrial membrane potential, which is reduced during early apoptosis (42). In these experiments the blocking antibody to ␣ 2 was also tested for its ability to affect MMP-1-related changes. As shown in Fig. 7c, MMP-1 was associated with a reduction in mitochondrial membrane potential, as assessed by a fluorescent probe, and this change was not observed in cultures pretreated with the blocking antibody to ␣ 2 . Along with the data shown in Fig. 7, a and b, these results suggest that MMP-1 is neurotoxic through a mechanism that is independent of proteolysis and involves changes in integrin signaling. DISCUSSION Although named for their ability to degrade matrix proteins and well studied for the same, MMPs are becoming increasingly recognized as effectors of cell signaling (43). For example, MMP-1 can degrade insulin-like growth factor-binding proteins and thereby stimulate IGF signaling (15), and MMP-9 may activate interleukin-1␤ (44). Although most studies have focused on the ability of MMPs to generate potential cell surface receptor binding ligands, MMPs can also bind to specific receptors, raising the possibility that they may more directly affect signaling. For example, MMP-9 can bind to CD44, MMP-1 can bind to integrins ␣ 1 ␤ 1 and ␣ 2 ␤ 1 , and MMP-2 can bind to integrin ␣ v ␤ 3 (20 -22). The present study shows that MMP-1 stimulates dephospho- FIG. 4. MMP-1-associated changes in phospho-Akt are independent of proteolytic activity. a shows the results of an experiment in which batimastat (20 ng/ml) was tested for its ability to inhibit MMP-1-associated changes in phospho-Akt. In this experiment 50 ng/ml MMP-1 was mixed with batimastat (Bat.) in serum-free medium before its addition to cultured cells. Extracts were again prepared 45 min after the addition of medium with or without MMP-1 and then compared for relative phospho-Akt by ELISA. Results represent the mean plus S.E. for an experiment performed with 4 -6 replicates. Although the difference between unstimulated and MMP-1 plus batimastat was significant at p Ͻ 0.04, there was no significant difference between MMP-1 and MMP-1 plus batimastat (Tukey test). In separate studies batimastat was not found to have an effect on its own. b shows the results of a similar experiment in which 50 ng/ml wild type MMP-1 was compared with 50 ng/ml catalytic mutant (pro-MMP-1 (E200A)) for its ability to stimulate a reduction in phospho-Akt. Data represent the mean plus S.E. for six replicates. The difference between unstimulated and wild type MMP-1 as well as that between unstimulated and pro-MMP-1 (E200A) was significant at p Ͻ 0.02 (Tukey test). There was no significant difference between wild type and mutant. rylation of Akt through a mechanism that is attenuated by a blocking antibody to ␣ 2 but not by an inhibitor of enzymatic activity. These data suggest that MMP-1 may affect integrin signaling in a manner that is independent of proteolytic activity. In terms of the receptor complexes and intracellular pathways linking MMP-1 to the dephosphorylation of Akt, it should be noted that integrin signaling has previously been linked to the activation of pathways that typically increase the phosphorylation of Akt. The engagement of ␣ 2 ␤ 1 has, however, recently been reported to activate protein phosphatase 2A and to stimulate the dephosphorylation of Akt (30). Interestingly, a new study shows that Akt may be localized to protein phosphatase 2A-containing complexes, which may in turn be functionally linked to ␤ 1 (45). A specific tryptophan residue in the ␤ 1 cytoplasmic domain can specifically affect Akt signaling without having an effect on other integrin-activated pathways such as phosphoinositide 3-kinase (45). Of additional interest, integrin activation has also been linked to G i protein signaling mediated by integrin-CD47 complexes (29). G protein signaling may also increase protein phosphatase activity (46,47) and may, thus, lead to an overall decrease in the phosphorylation of Akt (30). One possible consequence of MMP-1 signaling through ␣ 2 ␤ 1 would be altered cell migration/process outgrowth. Integrins have been well studied for their effects on cell shape and migration (48). Of interest with respect to the possibility of metalloproteinase-integrin interactions and enhanced cell migration is a previous study showing that ADAM-9 interacts with ␣ 6 ␤ 1 to stimulate fibroblast Rho kinase signaling and migration in a manner that is independent of catalytic activity (27). MMP-1-associated reductions in Akt activity could lead to changes in the activity of GSK-3␤ that would influence the stability of microtubule-associated proteins (41). Such changes might in turn have positive or negative effects on cell migration depending on factors including the type of cell and its status. It is, therefore, tempting to speculate that although MMPs disrupt matrix integrity and by doing so create extracellular conditions which allow for changes in cell shape and/or migration, they might also stimulate requisite intracellular changes. Another potential consequence of altered signaling through ␣ 2 ␤ 1 by MMP-1 is reduced cell survival. Apoptosis is required for tissue remodeling such as that which may occur in disease/ inflammation. MMP-1 may engage unoccupied ␣ 2 ␤ 1 and thereby stimulate intracellular changes linked to reduced cell survival. MMP-1 might also displace an endogenous ligand or otherwise alter the shape and/or adhesive properties of such a FIG. 7. MMP-1 stimulates neuronal toxicity that is attenuated by ZVAD and anti-␣ 2 but not by batimastat. For a and b data represent the mean plus S.E. for neurotoxicity experiments in which the % dead cells in 15 fields was determined (see "Experimental Procedures"). For the data shown in a batimastat (Bat) was used at 20 ng/ml, and ZVAD was used at 100 M, whereas for that shown in b batimastat was used at 500 ng/ml (1 M). For the data shown in a the difference between untreated and MMP-1-treated cells (50 ng/ml) was significant at p Ͻ 10 Ϫ4 , as was the difference between untreated and MMP-1 plus batimastat-treated cells. There was no significant difference between MMP-1 and MMP-1 plus batimastat. Alternatively, ZVAD was associated with a statistically significant inhibition of MMP-1-associated toxicity (p Ͻ 0.001, Tukey test). For the data shown in b, the difference between untreated and MMP-1-treated cells as well as the difference between untreated and MMP-1 plus batimastat-treated cells was significant (p Ͻ 0.03), whereas that between MMP-1 and MMP-1 plus batimastat was not. In c mitochondrial potential for the mean plus S.E. of eight replicates is shown. The difference between untreated and MMP-1-treated cultures was significant at p ϭ 0.008 (Tukey test). There was no significant difference between unstimulated and MMP-1 plus anti-␣ 2 or between unstimulated and anti-␣ 2 groups. ligand, thereby altering ␣ 2 ␤ 1 signaling and/or promoting weak adhesion. Although intermediate adhesion has been associated with cell motility, weak adhesion has been linked to cell death by anoikis (49). The cells used in our experiments were not plated on collagen or other known ␣ 2 ␤ 1 -binding proteins, but we cannot rule out their production of potential ligands or cell-cell interactions leading to ␣ 2 ␤ 1 engagement. In the present study neurotoxicity was inhibited by a blocking antibody to ␣ 2 , suggesting that changes in integrin signaling played a role. Integrin signaling has more generally been linked to cell survival. The ability of changes in integrin signaling to stimulate survival or apoptosis is likely, however, to depend on the nature of both the integrin and the ligand. With respect to ␣ 2 ␤ 1 , smooth muscle cell death has recently been reported to follow stimulation of cells with an ␣ 2 ␤ 1 -activating monoclonal antibody (28), and T cell apoptosis has been linked to the activation of ␤ 1 (50). The recently described functional link between ␤ 1 and protein phosphatase 2A (45) may be relevant with respect to these observations as well as to our own. In the study describing this functional link it was proposed that some ␤ 1 interactions are likely to maintain ␤ 1 -associated Aktprotein phosphatase 2A complexes in a state so that protein phosphatase 2A activity is repressed (45). Conceivably, effects on ␤ 1 by specific ligands or by loss of matrix might derepress protein phosphatase 2A and, thus, lead to decreased activity of Akt and apoptosis. Our findings also suggest that MMP-1-associated toxicity may be caspase-dependent. Although dephosphorylation of Akt would be expected to promote caspase-dependent apoptotic death, additional studies will be necessary to determine the extent to which MMP-1-related death may be dependent on changes in the activity of Akt as well as on changes in the activity of specific caspases. In summary, we have shown that MMP-1 stimulation of neurons leads to Akt dephosphorylation and cell death through a mechanism that is independent of proteolysis. These findings suggest that MMP-1 might contribute to the neuronal damage which occurs in association with degenerative and inflammatory conditions characterized by elevated levels of this proteinase.	2018-04-03T05:54:28.462Z	2004-02-27T00:00:00.000	{ "year": 2004, "sha1": "1aceb5ad14125ac47ec84055d06f50af16b1cd4f", "oa_license": "CCBY", "oa_url": "http://www.jbc.org/content/279/9/8056.full.pdf", "oa_status": "HYBRID", "pdf_src": "Highwire", "pdf_hash": "cc6b80f9b11e4d7251db6878b0d0727c279ca9b3", "s2fieldsofstudy": [ "Biology", "Chemistry" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
246653228	pes2o/s2orc	v3-fos-license	Factors associated with prenatal stress and anxiety in pregnant women during COVID-19 in Spain Aim of the study To describe prenatal stress and state anxiety levels in pregnant women living in Spain during the lockdown of the first wave of COVID-19 and its relation with obstetric factors, perception of health care, and concerns about the socio-sanitary situation. Methods The present study is an observational, correlational, and cross-sectional quantitative study. The participants in the study were pregnant women recruited through non-probabilistic convenience and snowball sampling during the lockdown. A web link was provided to an online questionnaire designed for this research, which collected socio-demographic and obstetric variables, perceptions of health care received during the pandemic and preoccupations associated with COVID-19. It also included the Prenatal Stress Questionnaire (PDQ) and the State Anxiety Inventory (STAI-S). Results Based on the responses of 695 pregnant women, the results showed a mean of 16.98 (SD = 25.20) of prenatal stress and elevated levels of anxiety (M = 25.20/SD = 11.07) in the first wave of the pandemic. Risk factors for prenatal stress and anxiety were the level of preoccupation associated with COVID-19 and previous mental health issues. A specific risk factor for anxiety was having more than one child and a protective factor were perceiving accessibility and availability of health care, with clear and consistent pregnancy care and follow-up protocols. Conclusions The lockdown period for COVID-19 was a stressful experience for pregnant women, highlighting the need to address their psychological well-being through clear and coherent protocols in terms of maternal-foetal health control and follow-up. Introduction The healthcare and social crisis caused by the COVID-19 coronavirus disease reported for the first time in December 2019 in the city of Wuhan, China, has affected the population worldwide. Due to the rapid propagation of the virus, on 14 th March 2020 the Spanish Government declared a ''state of emergency'' throughout Spain, limiting peoples' freedom of movement, save in exceptional circumstances. 1 This involved mandatory lockdown as a measure of contagion prevention and control, 1 which had an impact on the population's mental health. 2 This became obvious from the beginning of the coronavirus COVID-19 coronavirus pandemic, with anxiety increasing in the general population 3 and the exacerbation of other mental problems in the pregnant population. 4 At the start of the pandemic, it was observed that pregnant women who attended prenatal check-ups expressed concerns associated with the lack of information and the limitation of resources for control of the virus in the different health services. They also expressed fear at attending prenatal follow-up check-ups or that they would be cancelled; uncertainty regarding the possible non-compliance of their birth plan; doubts about the possibility of being accompanied by their partner and other family members, and the possible early termination of pregnancy and/or the need for elective caesareans and overly rigorous hygiene measures. 5 The pandemic represented the appearance of additional reasons for uncertainty, contributing to emotional distress during this period. 6,7 These concerns are common in first-time mothers with high risk, unplanned pregnancies. 8,9 Pregnancy requires adaptation in all areas of life, and this leads to doubts and uncertainty, which may be highly stressful for the woman. 10 Specific stress is spoken about referring to the prenatal stage, which is associated with the woman's concerns relating to physical symptoms, health of the foetus, the birth, interpersonal relationships, and development during pregnancy. 11 It has been estimated that over 25% of women suffer from prenatal stress. 12 When levels are high and persistent, they may trigger anxiety disorder. 13 Spielberger et al., 14 define the state of anxiety as an ''immediate, transitory emotional state over time, which involves physiological and cognitive changes characterized by tension, apprehension and nervousness''. During pregnancy, anxiety is a common mental health problem, being observed in 15.2% of the pregnant population. 15 Several studies concluded that women facing stressful events, among which may be included previous pregnancy losses, reproductive problems, fertility treatments, high risk pregnancies, unplanned or unwanted pregnancies, perinatal deaths, a history of psychiatric issues or a previous history of psychopathology, are more likely to experience anxiety during pregnancy. 16 It has become obvious that anxiety is a risk factor for the development of pregnancy, and is associated with a higher risk of pregnancy loss, lower intrauterine growth, premature birth, foetal disturbances in the hypothalamic-pituitary-adrenal axis, low birth weight, and also affective disorders in women, which can have longlasting effects on offspring. 17,18 COVID-19 disease has been associated with various maternal and perinatal risks, posing a major challenge for obstetric and gynaecological practice. 19 However, there is still little consideration of the risks of the socio-health situation caused by COVID-19 on women's mental health and its consequences on foetal development. It is essential to focus attention on the pregnant population, especially during the periods of lockdown associated with the pandemic crisis, as pregnancy could be considered a state of greater psychological vulnerability and therefore requires special care in the field of gynaecological-obstetric health. As a result, the aim of this study was to describe the levels of prenatal stress and state anxiety in pregnant women living in Spain during the lockdown resulting from the first wave of COVID-19, with the focus on specific variables, including the concerns associated with COVID-19, the perception on the accessibility and availability of healthcare, and obstetric factors. This study intended to discover what the risk and protective factors were, from the point of view of healthcare, which contributed to regulating the mental health of the pregnant woman during times of major uncertainty and concern. Method Design An observational, correlational, cross-sectional quantitative study on the factors associated with prenatal stress and anxiety in women living in Spain during the COVID-19 pandemic. Study population and scope The study population were pregnant women in lockdown during the first wave of the COVID-19 pandemic. To reach the sample, a non-probabilistic convenience and snow-ball sampling was used. To calculate the same size for a multiple lineal regression with the necessary quantity of predictors for the study, the G*Power programme version 3.1.9.7 20 was used, specifying a small effect size, an alpha of .05 and a power of .9. TheNrequired for the study was 531 subjects. Inclusion and exclusion criteria Inclusion criteria for the analyses in this study were: having completed the questionnaire during lockdown, being over 18 years of age and having been confined to the home for 10 days or more. Exclusion criteria were: not having completed all the questionnaires and being or having been hospitalised for obstetric complications at the time of answering the questionnaires. Variables Sociodemographic variables (age, nationality, educational level, civil status), obstetric variables (trimester of gestation, planned pregnancy, fertility treatment pregnancy, previous pregnancy loss, risk pregnancy, number of children), health history (chronic disease) and mental health issue history (previous diagnoses of depression, anxiety or other affective disorders) were collected. Information was also collected on the perceived accessibility and availability of health care during the pandemic, concerns associated with COVID-19. Levels of prenatal stress and state anxiety were measured. Measurement tools Following a bibliographic review on the background of the new coronavirus and mental health in previous pandemics, an ad hoc questionnaire was designed which included the proposed variables and this was applied in this study for the first time. The questions associated with sociodemographic, obstetric and health history variables were assessed with single choice questions, dichotomousÿesändnoquestions and self-completion questions. For the variable of perception on the accessibility and availability of health care, a Likert-type scale was constructed from 0 (Strongly disagree) to 4 (Strongly agree), with 3 questions: 1)Do you think that access to basic needs (food, water, medical care) is guaranteed?2)Do you think that the protocols of action are clear and coherent in terms of avoiding risks of contagion?; 3)Ẅas the information you received from your midwife/gynaecologist/health professional regarding the monitoring and care of the birth clear and satisfactory?. The overall scores obtained by the participants were used, adding the scores of each item, where the higher the score, the better the perception of the accessibility and availability of health care. For the variable of concerns associated with COVID-19, a Likert-type scale was constructed from 0 (Not at all) to 4 (Ëxtremely), with 9 questions with the same introduction: "How concerned are you about the impact of the COVID-19 pandemic": 1) ". . .on a social level?"; 2) ". . .on your job?"; 3) ". . .on your economic situation? "; 4) ". . .on the health of your friends and acquaintances?"; 5) ". . .on the health of your relatives (parents, grandparents, siblings, aunts and uncles)?"; 6) ". . .on the health of your partner?"; 7) ". . .on your own health?"; 8) ". . .on the health of your baby?"; 9) "Lately, I am more afraid than usual to go for prenatal check-ups. The overall scores obtained by the participants were used, adding up the scores for each item where the higher the score, the greater the concern. The Prenatal Distress Questionnaire (P DQ), 21 adapted and validated in a Spanish population by Caparros-González et al. al. 22 was used to assess prenatal stress. It is a scale used to provide information on specific concerns and stress during pregnancy, related to physical symptoms, relationships, parenting, medical problems, preparation for labour, delivery and infant health. It consists of 12 items that are structured into 3 factors assessing concerns related to birth, relationships and physical condition. The scale does not establish a specific cut-off point, so in this case the overall scores obtained by the participants were used, adding the scores of each item, where the higher the score, the greater the prenatal stress. In its original format it had a Cronbach's alpha of .80 and .81. The overall scale reliability of the questionnaire for this study had a Cronbach's alpha of .74. The State-Trait Anxiety Inventory (STAI), 14 adapted and validated in Spanish by Buela-Casal et al., 23 was used to assess anxiety. It is an inventory that assesses the state of anxiety characterised by consciously perceived subjective feelings that are linked to a precise and identifiable event. The inventory has 20 items for state anxiety (transient) and 20 items for trait anxiety (anxious propensity, relatively stable). Each item is scored on a 4-point scale ranging from 0 to 3, with higher scores indicating greater severity. Each subscale has a range of scores from 20 to 80, with higher scores indicating greater anxiety. In this study only state anxiety was assessed. The cut-off point used was 33/34, equivalent to the 85th percentile, in accordance with the scales established for the Spanish population. 23 The reliability of the state anxiety scale showed a Cronbach's alpha of 0.94 for this study. Data collection A flyer-type invitation was sent to institutions and professionals in the perinatal area and to pregnant women through social networks (Instagram ® , Facebook ® , Linkedin ® and Whatsapp ® ), which included a web link to the online survey. The online survey system LimeSurvey ® was used for its implementation. This is an open survey, in which any woman of legal age, fluent in Spanish and pregnant, regardless of gestational age or place of residence, was invited to participate voluntarily. Participants who responded between 27 April and 3 May 2020, the mandatory quarantine period in Spain, were included. No initial contact was made 2 with potential participants. The first section of the questionnaire included the objectives of the study, information about the study promoters, the supporting institution and the duration of the survey. At the end of this section, informed consent and information on transparency and data protection were provided. The participant's consent was requested by ticking a checkbox, which was taken as the signature required to give consent to participate and allow the use of data. A cookie was set to avoid repeated participation or changes in responses. At the end of the questionnaire participants were invited to share the questionnaire with other pregnant women. Data analysis Descriptive statistics such as frequencies and percentages for categorical variables and mean, standard deviation, maximum, minimum, coefficient of skewness and coefficient of variation for numerical variables were used. To test the hypotheses of the study, multiple linear regressions were used, reporting the adjusted R2 coefficient of determination, the regression coefficients with their respective hypothesis tests and the Durbin-Watson statistic to evaluate autocorrelation. Two regression models were performed. In the first one the dependent variable wasprenatal stressänd in the second one the dependent variable wasstate anxiety. In both models the possible associated factors incorporated as independent variables were: planned pregnancy, assisted reproduction gestation, number of children, previous pregnancy loss, risky gestation, presence of chronic illness, history of mental health, perception of accessibility and availability of health care and concern associated with COVID-19. The method used for selection of the final predictors was stepwise. The criterion for the rejection of the null hypothesis for the regression coefficients was used as an alpha of .05. For both models the possible effects of confusion in the sociodemographic variables were assessed. The bivariate relationship was observed between the sociodemographic variables and the study-dependent variables, being incorporated into the regression models as adjustment variables only those which presented with significant association. Analyses were performed using the IBM ® SPSS programme version 23. Results In total, 1,043 pregnant women completed the questionnaires. Out of these, 894 (85.7%) resided in Spain and 149 in other countries. Of those residing in Spain, 695 met with all the inclusion criteria (N = 695). The age range fell between 21 and 47 years (M = 33.83; SD = 4.102). As observed in Table 1, 96% stated they were of Spanish nationality, had With regard to the dependent study variables, it may be observed that prenatal stress (determined by the PDQ score) presents a mean of 16.98 (SD = 6.53; Min = 2; Max = 41), with a higher spread (CV = .38 > .1) and slight positive skewness (C. skewness = .503). The anxiety state variable (determined by the STAI-S score) presents a mean of 25.20 (DS = 11.07; Min = 0; Max = 57) with a high spread (CV = .44 > .1) and slight positive skewness (C. skewness = .285). According to the scales used for the Spanish population, in our study 67.3% had a low score in anxiety state and 32.6% presented with indicators of anxiety. With regard to categorical independent variables, Table 2 reveals that most participants in their third trimester of pregnancy (58.4%), state that their pregnancy was planned (84.9%), that the pregnancy was not achieved through assisted reproduction (88.3%), that they had other children (57.1%), that they had not had pregnancy losses prior to their current pregnancy (66.3%), that they had no chronic disease (78.3%), that they did not have a risk pregnancy (80.3%), and they had no history of mental health issues (63.2%). The perception of health care (determined by the score of the Perception of health care scale) presents a mean of 2.62 (SD = .74; Min = .33; Max = 4.00), quite close to the midpoint of the scale, with a high level of dispersion (CV = .28 > .1) and with a slight negative skewness (C. skewness = -.429). Concern associated with COVID-19 (determined by the COVID-19 Associated Concerns scale score) behaves similarly, presenting a mean of 2.62 (SD = .68; Min = .56; Max = 4.00), a high dispersion (CV = .26 > .1), and a slight negative skewness (C. skewness = -.339). Regarding the results for regression models (Table 3), for prenatal stress, the factors included in the model for presenting with a significant association with the variable were: concern associated with COVID-19, showing a positive association (␤ = 3.832; t = 11.549; p < .001); a history of mental health issues, i.e. prior diagnosis of depression, anxiety or other psychological difficulties associated with greater prenatal stress (␤ = 1.912; t = 4.081; p < .001); number of children which shows a negative association with prenatal stress (␤ = −.985; t = −2.854; p = .004) and pregnancy through treatment with assisted reproduction which is associated with a lower level of prenatal stress (␤ = −1.540; t = −2.166; p = .031). The model explains that 18.5% of the variability of prenatal stress (R 2 adj = .185) and there is no evidence of autocorrelation in the model (Durbin-Watson = 1.813). To sum up, concern for COVID-19 and a history of mental health are shown to be factors of risk for prenatal stress, whilst a protective factor is the number of children and pregnancy through assisted reproduction treatment. In the second regression model, referring to state anxiety, the factors included in the model as presenting a significant association with the variable were: concern associated with COVID-19, where there is a positive association (␤ = 4.798; t = 8.786; p < .001); a history of mental health isssues, associated with a higher state anxiety (␤ = 4.504; t = 5.885; p < .001); number of children which shows a positive association with anxiety (␤ = 3.246; t = −5.856; p < .001); perception of healthcare, which shows a negative association (␤ = −2.679; t = −5.312; p < .001); and that the pregnancy was planned, which is associated with a lower state anxiety (␤ = −2.703; t = −2.653; p = .008). The model explains the 24.6% variability of state anxiety (R 2 adj = .246) and there is no evidence of autocorrelation in the model (Durbin-Watson = 1.990). To sum up, during the first wave of the pandemic, risk factors for state anxiety were concern associated with COVID-19, having a history of mental health problems and having more than one child, whilst the protective factors were the perception of available healthcare and planned pregnancy. Discussion This study focused on describing levels of prenatal stress and state anxiety and their association with obstetric factors, perception of health care and concern about the sociosanitary situation generated by the first wave of COVID-19 in a sample of 695 pregnant women living in Spain during lockdown. As observed in the present study, the socio-health restrictions and the consequent lockdown resulted in the emergence of significant concerns from the pregnant population. The participants in this study have a mean score on the prenatal stress scale (PDQ) of 16.98 out of 48 points (SD = 6.53). This result is similar to that reported by Ibrahim and Lobel 17 in their systematic review prior to the pandemic, where they identify a mean prenatal stress score (PDQ) of 16.21 (SD = 6.22). It also coincides with the results reported by Romero-González et al. 24 during the pandemic in Spain, where they identify a mean prenatal stress (PDQ) of 16.87 (6.71). That is, women's specific concerns and stress related to gestational development remain at similar levels to those observed before the pandemic. S9 With regard to state anxiety, referring to subjective feelings of anxiety perceived to be linked to a specific event, 14 which in this case was the pandemic, the study participants had a mean score of 25.20 (SD = 11.07). Anxiety indicators are observed in 32.6% of pregnant women, when the value reported in the literature prior to the pandemic period was 15.2%15. This data contrasts with the results reported in other studies carried out during the pandemic in Spain in the same period of lockdown, where the mean anxiety (STAI-S) varies from 41.7 (SD = 10.6)25 to 43.07 (SD = 11. 73). 26 It is important to note that for the aforementioned studies 25,26 the sample was recruited in the hospital setting, unlike our study where they were recruited online. In a study conducted in Israel, it was observed that one of the most important sources of anxiety during the COVID-19 pandemic in pregnant women was attending pregnancy check-ups and being in public places. 27 It is possible that women recruited in hospital settings felt more exposed to the virus than those who completed online questionnaires. In order to analyse the factors associated with the levels of prenatal stress and anxiety presented by pregnant women between May and April 2020, two predictive analyses were carried out that considered 3 groups of independent variables (obstetric variables, variable of concerns associated with COVID-19 and variable of perception of accessibility and availability of health care). Both regression analyses, both for antenatal stress and state anxiety, indicate that the main predictor was the COVID-19 concern. That is, both specific gestational stress and anxiety related to the subjective perception of feelings of anxiety about the pandemic situa-tion are associated with concerns generated by the fear of personal contagion, of the foetus, of relatives or friends, of attending prenatal check-ups and its consequences at the social and occupational levels. This is consistent with studies carried out in pregnant women in other countries, such as the USA, 6 Israel 27 and Italy, 28 highlighting the concern for the health of the mother and foetus and its association with higher levels of anxiety and stress. Both analyses also identified that women who reported a history of mental health problems were at increased risk of presenting symptoms of prenatal stress and anxiety during pregnancy in times of pandemic. This finding is consistent with the study by Ravaldi et al., 28 which found that the presence of prior psychopathology predicted high levels of anxiety and post-traumatic stress in pregnant women during the pandemic in Italy. With regard to the variables specifically associated with prenatal stress, two variables were identified as being associated with lower levels of prenatal stress. A first variable is achieving pregnancy through assisted reproduction treatments. This is an initially counter-intuitive result that is not related to the results of other studies. 10 Studies conducted before the pandemic show that having undergone assisted reproduction treatments is accompanied by high levels of psychological stress 29 and that anxiety symptoms resolve after successful treatments. 30 Possibly, women who are pregnant after successful assisted reproduction treatment have found in lockdown the possibility to get away from the stressors they have previously faced during treatment and spend more time taking care of food, sleep and healthy practices that have a positive impact on the reduction of specific stress related to pregnancy. 17,31 However, there are no studies indicating an explanation for this either, and no studies have been found that speak to these outcomes in times of pandemic. A second variable associated with prenatal stress in this study is the number of children, where the more children the woman has, the lower the prenatal stress levels. A study in iran, 32 found that women who were pregnant for the first time were at greater risk of experiencing concerns during the pandemic. This could be explained by the fact that women who have more sons and daughters have already previously experienced concerns associated with motherhood. However, in terms of variables specific to state anxiety, the present study shows that having more children is a risk factor. This is consistent with the results from the study by Balluerka et al., 33 where emotional distress was found to be associated with the role of motherhood and the increased burden of household chores during the pandemic, since school closures meant that the children were in the home 24 hours a day. Notwithstanding, a protective factor of state anxiety is associated with having a planned or wanted pregnancy, which is consistent with the majority of research studies in the pregnant population. 16,34 Also of note as a protective factor specific to state anxiety is the perceived accessibility and availability of health care. That is, pregnant women who perceive that they have access to health care, who perceive clear and coherent protocols for action in terms of avoiding the risk of infection and who are satisfied with the care and follow-up provided by gynaecological-obstetric professionals, have lower lev-els of anxiety. Since the beginning of the pandemic, lack of information, as well as the difficulty in accessing prenatal check-ups, has been a major concern for pregnant women, 5 which has translated into additional reasons for uncertainty. The results of the present study are consistent with those reported in pre-pandemic studies, which indicate that lack of clear and timely information regarding the health situation may exacerbate the risk of psychological and psychosocial distress. 2 In addition, a study conducted during the first wave in a pregnant population in Turkey indicates, in this regard, that lack of information during the pandemic is associated with higher levels of anxiety. 35 This study has some limitations. Firstly, it presents selection biases, as it is a non-probabilistic sampling, when for this type of study with large samples a random probability sampling strategy would be more appropriate. Secondly, it is possible to identify certain information biases, as the online questionnaire does not ensure that the questionnaire is administered in a standardised manner. In addition to the above, the obstetric and health variables were not collected through a medical report to ensure that the participant really had such a history (either risky pregnancy, history of mental health or chronic illness) and that the questionnaire used to collect information regarding concerns and perceptions in relation to COVID-19 was developed on purpose for the study, without being validated beforehand, since no similar situation had been experienced before. Thirdly, there are limitations in terms of internal validity, as the study is cross-sectional, so it is not appropriate to refer to predictors per se. However, there are variables that are prior to the appearance of the dependent variables in terms of time, such as number of children, planned pregnancy, assisted reproduction treatments and perception of accessibility and availability of health care. Fourthly, in the present analysis, other variables such as social support or employment status, among others, which could have an impact on prenatal anxiety and stress, were not incorporated into the regression model. For future research, it would be interesting to investigate the relationship between childbearing after assisted reproduction treatment and prenatal stress. On the other hand, considering the protective nature of the perception of access and availability of health care in our results, it would be interesting to collect information on the type (public or private care) or quality of care received. The evaluation of coping strategies of the pregnant woman could also be pertinent, in order to identify specific factors for the promotion of resilience during gestational follow-up in times of crisis in prenatal care settings. Cross-sectional studies of pregnant women during the pandemic are a starting point for further studies to assess the possible consequences of stress and anxiety levels on maternal mental health, the impact on maternal-infant bonding and neonatal development. Given the risks associated with high levels of stress and anxiety during the prenatal period, it is essential to pay attention to the stress experienced by pregnant women in two ways: in relation to the uncertainty and concerns associated with pregnancy itself and to the concerns associated with the pandemic. This study demonstrates the importance of paying attention during these times of crisis to the most vulnerable pregnant women, such as those who report having suffered mental health disorders, and the need to attend to the psychological well-being of this population through clear and coherent protocols for the control and monitoring of maternal and foetal health. Financing This study did not receive any specific funding from public sector agencies, commercial sector agencies or not-forprofit entities.	2022-02-09T14:14:38.042Z	2022-02-08T00:00:00.000	{ "year": 2022, "sha1": "0f3b23a4940d1f1d7ea5ba7b88e7fea3c510ef0f", "oa_license": null, "oa_url": "https://doi.org/10.1016/j.enfcle.2021.10.003", "oa_status": "BRONZE", "pdf_src": "PubMedCentral", "pdf_hash": "a89e9d1cfc27a2247abe8f7ad7a80b1d5f0216f9", "s2fieldsofstudy": [ "Medicine", "Psychology" ], "extfieldsofstudy": [ "Medicine" ] }
207221643	pes2o/s2orc	v3-fos-license	To determine the effect of wearing shoe covers by medical staff and visitors on infection rates, mortality and length of stay in Intensive Care Unit. OBJECTIVE Intensive Care Units (ICUs) experience higher infection rates due to the severity of illness and frequent use of invasive devices. Use of personal protective equipment reduces the risk of acquiring an infection. This study has been conducted to determine the role of using shoe covers by medical staff and visitors on infection rates, mortality and length of stay in ICU. METHODS It is a descriptive study, performed in Shifa International Hospital, Islamabad from January 2012 to July 2012. The rates of infection (by checking patients for common ICU pathogens), mortality and length of stay of patients admitted in MICU and SICU from January 2012 to March 2012 were measured. Use of shoe covers was abandoned during this period. The same parameters were measured for the patients admitted from May, 2012 to July, 2012; the period during which shoe covers were strictly used by all the staff members and visitors. The data was then analyzed and compared using chi-square test with significance value at p< 0.05. RESULTS A total of 1151 patients were studied in 06 months period. Among the two groups of patients, managed with and without using shoe covers in ICU, statistically significant decrease was seen in terms of length of ICU stay(as P value is less than 0.05) in patients managed in duration of shoe covers. However, the time period in which shoe covers were used the infections with three common ICU pathogens MRSA, VRE and acinetobacter were statistically significant more than the periods in which shoe covers were not used. There was no significant difference in mortality for both groups (P value = 0.146). CONCLUSION Use of shoe covers in critical care area is not helpful in preventing infections of common ICU pathogens and length of stay in ICU patients; nor has it decreased the mortality. INTRODUCTION Large hospital-wide infection prevention schemes, focusing largely on increased awareness and improved hand-hygiene, has emphasized upon limiting the development of nosocomial infections. 1 Three organisms are common in hospital settings; the ICU patients are especially vulnerable to these infections. Acinetobacter baumannii is a Gram negative bacteria. It can be an opportunistic pathogen in humans, affecting people with compromised immune system. It has been isolated from soil and water samples in the environment. 2 MRSA and VRE can cause invasive and life-threatening infections, such as osteomyelitis, bacteremia, endocarditis, pneumonia, urinary tract infections, intra-abdominal or pelvic infections, vascular line sepsis, and wound and surgical infections. The primary route of transmission of MRSA from patient to patient is via contaminated hands of healthcare workers but it is not known that whether shoe dusts spread the infection or not. 3 However, Gram negative infections are mainly spread from the endogenous source but cross transmission is being recognized. Current guidelines for the prevention of spread of multi-resistant bacteria in the hospital setting do not distinguish between Gram-positive and Gramnegative isolates. 4 Colonization refers to the presence of microorganisms in or on a host with growth and multiplication, but without tissue invasion or damage. Infection is the entry and multiplication of microorganisms in the tissues of the host leading to local or systemic signs and symptoms of infection. 5 Using personal protective equipment provides a physical barrier between micro-organisms and the wearer. It offers protection by helping to prevent microorganisms from contaminating hands, eyes, clothing, hair and shoes; hence preventing transmission to other patients and staff. 2 Personal protective equipment includes: Gloves, protective eye wear, Masks, Apron Gown, Shoe covers and a Cap / Haircover. 6 Personal protective equipment reduces but does not completely eliminate the risk of acquiring an infection. It is important to use it effectively, correctly, and at all times when contact with blood and body fluids of the patients may occur. Staff must also be aware that use of personal protective equipment does not replace the need to follow basic infection control measures such as hand hygiene. 7 Personal protective equipment should be chosen in accordance with the risk of exposure. Health care workers should assess whether they are at risk of exposure to blood, body fluids, excretions or secretions and choose their items of personal protective equipment according to the assessed risk. 8 Disposable caps and protective foot wear should be used where there is a likelihood that patient's blood, body fluids, secretions or excretions might splash, spill or leak onto the hair or shoes. [6][7][8] Shoe covers should not however be used for the prevention of surgical site infections (nevertheless, shoe covers are required by OSHA regulations when "gross contamination can reasonably be anticipated") (CDC category IB) A-IV. 9 The use of shoe covers has never been proven to decrease the risk or incidence of surgical site infections, or to decrease the bacterial counts of the operating room floors. 10,11 It also has no mortality benefit when compared to unrestricted access to surgical ICU. 12 Our study measures the advantage of using shoe covers in intensive care setup in terms of rate of infection, mortality and length of ICU stay. METHODS It was a prospectively conducted study in Medical and Surgical ICU of Shifa International Hospital, Islamabad. Total duration of study was 06 months. Two groups were made. The first group included patients who were managed in the above mentioned ICU's when all the staff members and the visitors were not using shoe covers while being in ICU i.e. from January to March 2012. The other group included the patients, who were given critical care when shoe covers were strictly being used by the staff and the visitors in the ICU i.e. from May to July, 2012. The parameters studied in all the patients were the rate of infections, mortality and length of ICU stay to assess whether wearing shoe covers would reduce infections and improve these parameters. All the patients presenting in Medical and Surgical ICU in the above mentioned 06 months were included in the study. From January 2012 to March 2012, when shoe covers were not in use, other infection control practices such as hand washing and antisepsis, use of personal protective equipment when handling blood, body substances, excretions and secretions, environmental cleaning and spills-management and appropriate handling of waste, remained the same as for the other group. During May, 2012 to July, 2012; the period during which shoe covers were being used, none of the staff members, consultants, residents, fellows, nurses, paramedical staff and visitors were allowed in the ICU's without shoe covers. A boundary line was drawn and a warning note was pasted at the ICU entrance regarding strict use of shoe covers. Infection control nurses and ICU staff scrutinized every person. The patients were then serially followed for their length of stay, mortality and incidence of infections with three common ICU pathogens Acinetobacter, MRSA and Vancomycin resistant Enterococcus by obtaining cultures from blood, urine, sputum and other body secretions. Data was entered and analyzed using SPSS version 16.0. Descriptive statistics were calculated. Chi-square test was applied to determine the difference in infection, mortality and length of stay in two groups. P value of less than 0.05 is considered significant. RESULTS A total of 1151 patients were admitted in medical and surgical ICU of shifa International hospital, Islamabad during the 06 months period of this study. All were included in the study. Out of these 63.7% (733) were below 60 years of age while the remaining 36.3% (418) were 60 years or more of age. 64% (737) patients were males and 36% (414) patients were females. Among the total 1151, 51.9% (597) were admitted in MICU whereas 48.1% (554) were admitted in SICU. 55.4% (638) patients were admitted in the duration when shoe covers were not being used while 44.6% (513) patients were admitted in the three months when shoe covers were being used by MICU and SICU staff and visitors. The cultures were done for common ICU pathogens (acinetobacter, vancomycin resistant Enterococcus, MRSA) and only 6.6% (76) patients were found to be culture positive for these organisms. Acinetobacter was obtained in tracheal aspirates of 36 patients, while MRSA was found in 12 patients as shown in Bar chart 1. There were a total of 20.7% (238) deaths during the study period. The duration of ICU stay was within three days for 61.8% (711) patients whereas 17.3% (199) patients stayed for more than six days (Table-I). Comparing the culture positive patients, out of total 6.6% (76) culture positive patients 2.8% (32) were from MICU and 3.8% (44) were from SICU, while 30 patients were culture positive before the use of shoe covers and 46 patients came out to be culture positive when shoe covers were in use, p value is 0.004 as shown in Table-I. Among the total 20.7% (238) deaths, 10.6% (122) occurred in the period when shoe covers were not being used while 10.1% (116) were seen in the time period when shoe covers were being used (p value >0.05) ( Table-I). More number of MICU patients expired i.e. 14.1% (162) when compared to SICU patients, where only 6.6% (76) deaths were observed. The p value came out to be 0.001. Considering the length of stay, 65% (100) patients stayed in ICU before using the shoe covers while 57.7% (99) patients stayed in ICU for less than 03 days during the time when shoe covers used (p value <0.05) as seen in Table-I. DISCUSSION Intensive care unit (ICU) is one of the most eventful units of the hospital and uses sophisticated equipment and advanced medical practices. At the same time ICU may experience higher infection rates due to the severity of illness and frequent use of invasive devices such as intravenous catheters, feeding tubes, airways, etc. Not only are there chances for infection from patients' own endogenous microorganisms, there is also risk of infections from other patients' or environmental microbes if hand hygiene measures and other precautions are not ensured. In Pakistan majority of the hospitals do not have any shoe change policy in the intensive care units. Only a small number of hospitals have implemented the use of personal protective measures for infection prevention and restricted access to the intensive care areas. Earlier studies have shown that the use of barrier nursing and protective measures by the staff in ICU will reduce the incidence of infections due to reduced contamination. 13 In a related study carried out to find the efficacy of protective footwear on bacterial infection, no significant difference was Zeeshan Ali et al. found in the infection rates with and without the use of protective footwear. 11,12 The study done in our setup reached the same inference i.e. not only the use of foot wear has no benefit regarding the control of infections in the critical care but also the rates of infection were more, this was probably due to the fact that individuals while putting on the shoe covers the shoes contaminated their hands and thus further transmitted the infection. There are several factors other than protective foot wear, in the medical facilities of the developed countries to limit infection rates in ICU. These include measures like strict hand washing and limited numbers of visitors entering the ICU which already reduces the bacterial floor contamination so uncertain measures like use of shoe covers are not considered important in infection control. 14 Whereas, the hospitals in the developing countries like Pakistan (where the general environment is not clean enough) hand washing is not strictly followed and the number of visitors in ICU cannot be strictly controlled, so use of shoe covers has been used in some hospitals with the idea to control the ICU infection rates and decrease the air and floor colony count but as per the results of this study, no reduction is seen in infection rates, mortality and length of stay by using the shoe covers in intensive care settings. In a study conducted by Gupta A et al, to find the efficacy of protective footwear on bacterial floor colonization, the floor and air colony counts showed no significant difference in the two phases, with and without protective footwear. 12 This further rationalizes the adoption of a more restricted access to ICUs in terms of the number of personnel allowed to enter the ICU and at the same time questions the practice of shoe change policy still used in our hospitals. However, our study had limitations, First it was single centered study and the other hospitals of the territory were not included, Second, sample size of the two groups were varied 55.4% vs 44.6% patients in the shoe cover group, third the patients admitted in ICU have multiple comorbidities in addition to infection which contributed to the mortality, these factors were not considered. CONCLUSION Use of shoe covers in critical care area is not helpful in preventing infections of common ICU pathogens; nor has it decreased the mortality and length of stay in ICU patients. It requires more studies to be carried out involving aspects such as shoe change practices, restricted access, etc.; so that definite policies can be laid down for infection control in critical care patients. While developing such policies, an integrated approach should be undertaken by involving both the administrators and clinicians, in order to achieve optimum results on implementation of the same.	2016-05-12T22:15:10.714Z	2014-03-01T00:00:00.000	{ "year": 2014, "sha1": "f3be80de22bbc04bd1ff78d00ab137b68c7ae640", "oa_license": "CCBY", "oa_url": "https://doi.org/10.12669/pjms.302.4384", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "f3be80de22bbc04bd1ff78d00ab137b68c7ae640", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
119098577	pes2o/s2orc	v3-fos-license	Borel resummation of transverse momentum distributions We present a new prescription for the resummation of contributions due to soft gluon emission to the trasverse momentum distribution of processes such as Drell-Yan production in hadronic collisions. We show that familiar difficulties in obtaining resummed results as a function of transverse momentum starting from impact-parameter space resummation are related to the divergence of the perturbative expansion of the momentum-space result. We construct a resummed expression by Borel resummation of this divergent series, removing the divergence in the Borel inversion through the inclusion of a suitable higher twist term. The ensuing resummation prescription is free of numerical instabilities, is stable upon the inclusion of subleading terms, and the original divergent perturbative series is asymptotic to it. We compare our results to those obtained using alternative prescriptions, and discuss the ambiguities related to the resummation procedure. Transverse momentum resummation The computation of transverse momentum distributions of heavy systems (such as dileptons, vectors bosons, Higgs) plays an important role in collider phenomenology, from the Tevatron to the LHC [1,2]. As is well known, the perturbative QCD expansion of the inclusive distribution contains to all orders powers of α s ln 2 (q T /Q), due to the emission of soft and collinear gluons. When the transverse momentum q T is much smaller than the mass of the final state Q these logs become large and must be resummed in order for perturbative predictions to remain reliable. The resummation, to given logarithmic accuracy, can be performed [3] for the Fourier transform of the differential cross-section dσ dq 2 T with respect to q T . Upon Fourier transformation, q T turns into its Fourier conjugate, the impact parameter b, and large logs of q T /Q become large logs of bQ. Fourier transformation is necessary in order for the contributions included by resummation to respect transverse momentum conservation, thereby avoiding the spurious factorial growth of resummed coefficients [4]. However, the Fourier transform must be inverted in order to obtain resummed predictions for physical observables. This is problematic because the Fourier inversion integral necessarily involves an integration over the region of impact parameters where the strong coupling is not well defined because of the Landau pole. This problem has been treated with various prescriptions. One possibility is to modify the behaviour of the strong coupling in the infrared in the Fourier inversion integral [3] (b ⋆ prescription, henceforth): this procedure is widely used, but it is known to lead to numerical instabilities when the resummed results are matched to fixed-order ones [5]. A second option is based on the observation that the Fourier inversion integral can be computed order by order in an expansion of the resummed results in powers of α s : if only leading log terms are retained in the Fourier inversion, the result is then well defined for all values of q T [5]. This procedure however is unstable to the inclusion of subleading corrections: the Fourier inversion can be performed to next-to-leading log accuracy [6] (as it is necessary if the resummation is performed to this order), but in such case the result differs significantly from the leading log one, and in fact for Q around 100 GeV it blows up for values of q T of order of several GeV, well within the perturbative region. A "minimal" prescription which is free of these difficulties can be constructed [7], along the lines of the similar prescription for threshold resummation [4]. Namely, the integration path in the Fourier inversion is deformed in such a way as to leave unchanged the result to any finite perturbative order, but avoiding the Landau pole and associate cut in the resummed result. This leads to a prescription which is free of numerical and perturbative instabilities: its only shortcoming is that it is difficult to assess the ambiguities related to the resummation procedure, as it can be done in the b ⋆ prescription by varying the way in which the infrared behaviour of the strong coupling is modified. Here we shall show that, analogously to what happens in the case of threshold resummation [8], the ambiguity in the resummation procedure is due to the fact that the perturbative expansion of the resummed result for the transverse momentum distribution itself in powers of α s diverges. After discussing, in the next section of this paper, how existing prescriptions treat this divergence, we will show in section 3 that the divergent series can be treated by Borel summation, as is the case for threshold resummation [8,9]. The Borel transform of the series converges and can be summed. The inversion integral which gives back the original series di-verges, but the divergence can be removed by including a suitable higher twist term. This leads to a resummed result of which the original divergent series is an asymptotic expansion. The ensuing prescription is given in terms of a contour integral which is easily amenable to numerical implementation. The result is free of numerical instabilities, and stable upon the inclusion of subleading corrections. An estimate of the ambiguity on the resummed results may be obtained from a variation of the higher-twist term which is included in order to render the results convergent. In section 4 we will compare the result of our prescription to other existing prescriptions in the case of the Drell-Yan process, and discuss the ambiguities related to the resummation procedure. Some results on Fourier transforms are collected in the Appendix. The need for a resummation prescription Let us consider a parton-level quantity Σ which depends on a large scale Q and a transverse momentum q T , such as the partonic Drell-Yan differential cross-section dσ dq 2 T . Resummation is necessary because the perturbative coefficient of order n in the expansion of Σ in powers of α s (Q 2 ) has the form 3) P n (lnq 2 T ) is a polynomial of degree 2n − 1 in lnq 2 T , Q n (q 2 T ) is regular as q T → 0, and D n are constants (see the Appendix for a definition of the + distribution). Physical observables are obtained, exploiting collinear factorization, as the convolution of parton level cross-sections with parton distributions [3]. When Q 2 is large enough, it sets the scale of parton distributions, and the q T dependence is entirely given by the partonic cross-section. For lower values of Q 2 the scale of parton distributions is set by the impact parameter b, which is Fourier conjugate to q T , the convolution must be performed in b space, and the Fourier transform must be inverted to obtain physical predictions. In either case, the resummation is performed in b space at the level of partonic observables. Upon Fourier transformation, q T is replaced by its Fourier-conjugate variable, the impact parameter b, and the small-q T region is mapped onto the large-b region. Large logs of b can then be resummed, leading to an expression of the form is the large logarithm which is resummed, and O(L 0 ) denotes terms which are not logarithmically enhanced as b → ∞. For future convenience, we have introduced in the definition of L an arbitrary constant b 0 (to be discussed below), and we have further defined β 0 is the first coefficient of the QCD beta function, The inverse Fourier transform of Σ with respect to b is given by using two-dimensional polar coordinates forb ≡ bQ, and the integral representation of the 0-th order Bessel function, Now consider specifically the resummation of is the partonic transverse momentum distribution of a massive final state, andσ 0 the Born-level total cross-section. In this case, the b-space resummed result has the form [3] Σ(α s ,ᾱ L) = exp S(α s ,ᾱ L), (2.12) where and the constants A i , B i can be determined order by order by matching to the fixed-order calculation. The integral in eq. (2.13) can be performed explicitly, and the result can then be expanded as where inclusion of the first k orders in the sum corresponds to the next k -to-leading log (N k LL) approximation. The LL and NLL functions f 0 , f 1 are explicitly given by Note that with y =ᾱL, using the leading log form of α s (Q 2 ), . (2.18) It is apparent from eqs. (2.16,2.17) that Σ(α s ,ᾱL) has a branch cut along the negative real axis in the complex plane of the variable y =ᾱL: This is due to the fact that the strong coupling blows up when its argument reaches the Landau pole, so that S(α s ,ᾱ L) eq. (2.13) is singular when b becomes large enough, i.e. when It follows that the series for Σ(α s ,ᾱL) eq. (2.4) has a finite radius of convergence, and the integrand in eq. (2.9) is not analytic in the whole integration range 0 ≤b < +∞, so the Fourier inversion integral is not well-defined without a prescription to treat the singularity. As mentioned in the introduction, various prescriptions of this kind have been proposed. Before discussing them, let us show that the reason why a prescription is needed is the divergence of the expansion in powers of α s (Q 2 ) of the resummed result obtained computing the inverse Fourier transform eq. (2.9) with Σ(α s ,ᾱ L) eq. (2.12). To any finite perturbative order, the q T -space resummed result is found by expanding eq. (2.12) and inverting the Fourier transform order by order: where we have replaced the argumentq 2 T of Σ bȳ When K → ∞ the series eq. (2.21) diverges. To see this, we compute the integrals in eq. (2.21) using eq. (A.1) of the Appendix: where the function M(η) is defined in eq. (A.2), we have assumedq 2 T = 0, so that distributions can be ignored, and the term with j = k, which leads to a vanishing contribution to Σ K (α s ,L), has been included in the sum over j eq. (2.24) for later convenience. We now change the order of summation, and use the identity where the integration path H is any closed contour which encloses the origin ξ = 0. We obtain Because of the singularity eq. (2.19), the power series eq. (2.4) has a finite radius of convergence equal to one which immediately implies the vanishing of the radius of convergence of the sum over k in eq. (2.27). The situation is thus similar to that which is encountered in threshold resummation [4,8,9]: the resummation is performed on quantities which are related by Mellin transformation to the physical ones, but the resummed results cannot be expressed as a Mellin transform of some function. Namely, their inverse Mellin transform does not exist, as a consequence of the fact that the inverse Mellin transform of their expansion in powers of α s (Q 2 ) diverges. In the present case, the divergence of the perturbative expansion implies that the Fourier inversion integral is ill-defined; of course the problem disappears if one retains only a finite number of terms in the resummed expansion [10,11]. Various commonly used prescriptions replace the ill-defined integral with a well defined one, as we now review. In the next section, we construct a prescription which is instead based on the idea of replacing the divergent series with a convergent one through the Borel summation method. In the last section we will compare the various prescriptions and in particular the way they treat the divergence of the perturbative series. In the prescription of ref. [3], the variable b is replaced by a function b ⋆ (b) which approaches a finite limit b lim ≤ b L as b → ∞, such as for example In this way, the cut eq. (2.19) is never reached. This procedure has some degree of arbitrariness in the choice of the function b ⋆ (b), which is interpreted as a parametrization of non-perturbative effects, whose size can be estimated by varying b ⋆ , for instance by changing the value of b lim . The matching of this prescription to the fixed-order result is however numerically unstable, as pointed out in ref. [5]. A different possibility [5] is based on the observation that if only the leading log contribution (i.e. the terms with j = 0) are included in eq. (2.24), then the series converges, and its sum can in fact be computed in closed form, with the result (see eq. (A.16) of the Appendix) . Therefore, using eq. (2.18), the LL expression eq. (2.30) is seen to become a function of α s (q 2 T ). Therefore, the leading log truncation of the perturbative expansion in powers of α s (Q 2 ) eq. (2.27) has a finite radius of convergence, set by the Landau pole where the last equality holds at leading order. The main defect of this result is that it is subject to large next-to-leading log corrections. In fact, the NLL Fourier inversion integral can also be computed in closed form [6]. The result (given in eq. (A.17)) differs sizably from the LL result even for relatively large values of q T (several GeV for Q = 100 GeV), as we shall see explicitly in Sect. 4 below. In fact, it turns out that the NLL correction diverges at a value of q T which is an increasing function of the scale Q. This instability can be understood as a consequence of the fact that the truncation of the resummed result to finite logarithmic accuracy leads to an expansion in powers of α s (q 2 T ) with coefficients depending on ln(q T /Q), where higher powers of α s (q 2 T ) correspond to higher logarithmic orders. Such an expansion is necessarily poorly behaved at low q T , all the more so when the scale ratio q T /Q is large. Performing the Fourier inversion to leading or next-to-leading logarithmic accuracy thus removes the divergence of the series eq. (2.21): this is analogous to what is found in the case of threshold resummation, where it can be shown [8] that the divergence of resummed results is removed if the Mellin inversion is performed to any finite logarithmic accuracy. However, the ensuing results are then perturbatively unstable. A yet different way of treating the divergence has been proposed more recently in ref. [7], along the lines of the so-called Minimal Prescription of threshold resummation [4]. The basic idea here is that to any finite perturbative order, when the divergent series is replaced by a finite sum, one may choose the integration path in such a way that it avoids the singularities which appear at the resummed level. The result of the Fourier (or respectively Mellin) inversion is then unchanged to any finite perturbative order, but it becomes finite at the resummed level. It can be further shown [4] that the divergent perturbative expansion of the resummed expression is asymptotic to the result obtained in this way. This prescription is widely used [2]: whereas in the case of threshold resummation it leads to dependence of resummed physical results on a kinematically unaccessible region (albeit by power-suppressed terms), in the case of transverse momentum resummation its only shortcoming is speed limitation in its numerical implementation. The Borel prescription We now turn to the construction of a prescription which extends to transverse momentum resummation the Borel prescription proposed in refs. [8,9] for the resummation of threshold logarithms. The basic idea is to tackle directly the divergence of the series (2.24,2.27) by summing it through the Borel method. To do this, we take the Borel transform of eq. (2.27) with respect toᾱ. This amounts to the replacementᾱ k → w k−1 /(k − 1)!, where w is the Borel variable conjugate toᾱ. We obtain where in comparison to eq. (2.27) we have rescaled the integration variable ξ →Lξ, and we have included all terms with 1 ≤ k ≤ j − 1, which vanish upon contour integration. Both sums in eq. (3.1) are convergent as K → ∞. Indeed, the last condition being due to the simple pole of M(ξ) at ξ = 1. Thus, provided the contour H is chosen so that Since M(ξ) has no singularities on the negative real axis, and Σ(α s , w/ξ) has a branch cut on the real ξ axis between −w and 1, the integration contour can now be deformed so thatR(w,L) is well defined for all positive values of w (see fig. 1). The original function eq. The inversion integral is divergent at w → ∞. This is easily seen by inspection of fig. 1: as w becomes large, the branch cut extends to the left, and the integration contour is pushed towards large negative values of ξ, where M(ξ) oscillates with a factorially growing amplitude. We regulate the integral by cutting it off at w = C. We thus get which is the Borel prescription for transverse momentum resummation. The result can be equivalently rewritten by doing a partial integration as which may be more convenient for numerical implementations in that it depends directly on the physical observable Σ, rather than its derivative. Equation (3.8), and its equivalent form eq. (3.7), are the main result of this paper. It is interesting to observe that if we integrate by parts before cutting off the integral, then the surface term vanish. We then end up with the alternative resummation As we shall see shortly, this is an equally valid prescription. In order to see that this is a valid resummation prescription, consider the truncation to order K of eq. (3.6), namely is the truncated gamma function. The difference between the original R K (α s ,L) eq. (2.23) and its Borel resummation R C K (α s ,L) is (3.12) Because 13) R ht K (α s ,L; C) is seen to be power-suppressed at large Q 2 (higher twist): cutting off the w integration at w = C is equivalent to the inclusion of a higher twist term, which cancels the divergence of the resummed expression. Specifically, R ht K (α s ,L; C) is a twist-t contribution with t = 2 (1 + C), (3.14) the choice C = 1 corresponds to the inclusion of a twist-four term. Moreover, it is apparent from eq. (3.12) that which vanishes faster than any power of α s as α s → 0. It follows that the original divergent R K (α s ,L) is an asymptotic expansion of the Borel-resummed result R C (α s ,L) eqs. (3.8,3.7). Furthermore, the alternative prescription R C ′ (α s ,L) eq. (3.9) differs from R C (α s ,L) eq. (3.8) by the first term in square brackets in (3.8), which is a finite higher-twist contribution. Hence, the two prescriptions correspond to two inequivalent but equally acceptable regularizations of the divergent sum which differ by finite terms, and are both asymptotic sums of the divergent series. The main features of the Borel prescription can be appreciated by considering as an explicit example of a resummed quantity Σ(α s ,ᾱL) = γ LL (α s ,ᾱL), with Substituting this form of Σ(α s ,ᾱL) in eq. (3.7), the associate q T -space physical observable computed with the Borel prescription is found to bē The ξ integral is easy to calculate, because the integrand has only a simple pole at ξ = −w: where we have used the leading-log expression of the running coupling. It is thus clear that the divergent integration is cut off by the inclusion of a power-suppressed contribution Note that the suppression is by powers of Λ 2 q 2 T : at finite order K the higher twist contribution is suppressed by a power of Λ 2 Q 2 , as shown in eq. (3.12), but when resummed to all orders, the scale Q 2 is replaced by an effective scale q 2 T . Comparison of resummation prescriptions Let us now compare the results found using the Borel prescription to those of other prescriptions, with the dual goal of understanding the advantages and disadvantages of various methods, and of assessing the ambiguity which is intrinsic to the resummation of a divergent expansion. First, we look at a typical resummed observable. Namely, we consider the transverse momentum distribution of Drell-Yan pairs, eq. (2.11), which we evaluate at the partonic resummed next-to-leading log level, i.e. using eq. (2.12) with S(α s ,ᾱ L) computed including the first two terms in eq. (2.15), given in eqs. (2.16,2.17) with [12,13] A 1 = C F π (4.1) The results are displayed in fig. 2, for Q = 100 GeV. The two lower curves at large q T in this figure correspond to those found using respectively eqs. (A.16) and eqs. (A.17) of the Appendix, namely, to inverting the Fourier transform to leading and next-to-leading log accuracy (with b 0 = 2e −γ E ). The sizable difference between these two results even for q T as large as 10 GeV shows the instability of the truncation of the Fourier transform to finite log accuracy discussed in the introduction and first stressed in ref. [6]. The other prescriptions displayed in fig. 2 are the b ⋆ prescription, where the Fourier inversion is performed after replacing b with b ⋆ eq. (2.29), with b lim = b L , where b L = 7.2 GeV −1 is the NLO Landau pole eq. (2.20); the minimal prescription (MP) where the Fourier inversion is performed along the deformed path of ref. [7], and the Borel prescription eq. (3.7) with C = 1. In fig. 3 we further show the dependence of the Borel prescription on the parameter C which characterizes the higher twist term included in the resummation eqs. (3.13,3.14), as it is varied between twist four and twist eight. Because all these choices provide valid resummation prescriptions, this variation provides an estimate of the ambiguity which is intrinsic of the resummation procedure: indeed, the b ⋆ and minimal prescription, also shown in this figure, are well within the band of variation as q T → 0. These plots show that the ambiguity in the resummation procedure is negligible for q T 5 GeV, it remains small for q T 2 GeV, and it only blows up as q T approaches the Landau pole. We can further elucidate the origin of these results by studying the effect of the various prescriptions when the divergent sum eq. (2.21) is truncated, so the Fourier inversion can be performed term by term. Consider specifically the first term in the series, namely, the inverse Fourier transform of L. The exact result is given by eq. (A.1) for k = 1, The MP reproduces this exact result, because ln(b 2 0 /b 2 ) is analytic on the positive realb axis, and a deformation of the integration contour has no effect; a branch cut on the positive realb axis only arises after summation of the whole series. The Borel prescription yields instead as one can see by setting h 1 = 1 α and h k = 0 for all k = 1 in eq. (3.10). The exact result is modified by the introduction of a correction of twist 2(1 + C). Note that the higher twist correction is tiny at large Q 2 , of order 10 −6 for C = 1 and Q 2 = 10 4 GeV 2 . If we use the alternative Borel prescription R C ′ (α s ,L) eq. (3.9) we get instead so the two prescriptions are indeed seen to differ by a higher twist term. Finally, the result of the replacement of b by b ⋆ eq. (2.29) can be computed analytically in terms of the Bessel function K 1 : Using the asymptotic behaviour K 1 (z) ∼ z→∞ e −z / √ z, we see that the correction factor in eq. (4.7) vanishes faster than any power of 1/(b lim q T ) for q T ≫ 1/b lim . For higher order powers of L the same qualitative behaviour is found using the various prescriptions discussed here. Namely, the MP gives the exact Fourier transform eq. (A.1); the BP gives a result which differs from it by a higher twist term, and the b ⋆ prescription gives a result which differs from it by a term which is exponentially suppressed in 1/(b lim q T ). We thus see that the way different prescriptions tackle the divergence of the perturbative expansion is the following. In the LL and NLL case, the divergent series eq. (2.24) is made convergent by truncating the Fourier inversion to finite order, i.e. by only retaining a finite number of terms in the inner sum over j. This, as discussed in Section 2, leads effectively to an expansion in powers of α s (q 2 T ) which has very poor convergence properties at small q T even when Q is large. The MP and BP both provide an asymptotic sum of the divergent series: the BP removes the divergence by inclusion of a higher twist term, and the MP by a suitable analytic continuation, which corresponds [4] to the inclusion of terms which are more suppressed than any power of Q 2 . At large Q 2 , the higher twist term of the BP is negligible so these two prescriptions are essentially indistinguishable when applied to convergent series. When applied to the divergent resummed expansion displayed in figs. 2-3 they only differ in the region where q T approaches the Landau pole, so the high-order behaviour of the series become relevant. Finally, the b ⋆ prescription modifies the divergent series by inclusion of a term which is more suppressed than any power of 1/(b lim q T ). When applied to a convergent series, this prescription produces a result that differs sizably from that of the BP when q 2 T ≪ Q 2 and it approaches the Landau pole: this is because the scale of the correction term is set by Q 2 for the BP, and by q 2 T for the b ⋆ prescription. At the resummed level, however, the effective scale of power suppressed terms becomes q 2 T also for the BP (compare eq. (3.20)), so all resummation prescriptions lead essentially to the same result. Summary We have constructed a resummation prescription for transverse momentum distributions which extends to this case the Borel prescription previously proposed for threshold resummation [8,9]. The construction is based on the observation that the reason why a resummation prescription is needed in the first place is that the perturbative expansion of resummed results in q T space in powers of α s (Q 2 ) diverges. The Borel prescription tackles this divergence by summing the convergent Borel transform of the divergent series, and then making the Borel inversion finite by inclusion of a higher twist term. The original divergent series is an asymptotic expansion of the result obtained thus. The Borel prescription is easily amenable to numerical implementation; being based on a b-space resummation it is easy to match to fixed-order results, and it is perturbatively stable. There is some freedom in this prescription, parametrized by a real parameter C, related to the twist t of the term included in order to obtain convergence by t = 2(C + 1). Whereas C may be chosen to take any value, it is convenient to choose a value which corresponds to twists which already appear in the expansion of the observable being considered. Indeed, physical observables must be independent of the choice of C, and thus if an unphysical twist term is introduced, it must be compensated by an equal and opposite power suppressed term which is thereby artificially introduced by this choice. Comparison of the Borel prescription to other available resummations, such as the minimal prescription or the b ⋆ method, shows that at large Q 2 they lead to results which are extremely stable and which only differ when q T approaches the Landau pole. In fact, variation of the parameter C of the Borel prescription provides a reliable estimate of the ambiguity in the resummation procedure. For q T 2 GeV this ambiguity appears to be negligibly small, even in the region of a few GeV where the impact of the resummation is sizable. This is in contrast to the case of threshold resummation, where it was found [9] that the ambiguity is almost as large as the effect of the resummation itself in most of the kinematic region where the resummation is relevant. Our results contradict the widespread prejudice that transverse momentum resummation is affected by sizable ambiguities, and it shows that, at least as long as Q is as large as the W mass and q T as large as the nucleon mass perturbative resummation of transverse momentum distributions provides reliable and stable results. The Borel prescription provides a new method for performing this resummation which has more stable matching properties than the b ⋆ prescription and might be numerically advantageous over the widely used minimal prescription. A Appendix In this appendix, we collect some results on two-dimensional Fourier transforms of powers of logarithms. First, we compute the exact Fourier transform with respect tob of the k-th power of ln k b 2 0 b 2 (with b 0 a constant). We get and the + distributions are defined by In order to prove eq. (A.1), we define a generating function where we have used polar coordinates forb, and the integral representation of the 0-th order Bessel function The integral can be computed by means of the identity We find 1 2π We may now replace consistent with the definition eq. (A.3). We get Evaluating the k-th derivative of both sides with respect to η at η = 0 leads immediately to eq. (A.1). Note that the term j = k is excluded from the sum because it vanishes upon differentiation with respect toq 2 T . Forq 2 T strictly larger than zero, both the term proportional to δ(q 2 T ) and the + prescription have no effect. Let us now turn to the evaluation of the Fourier transform to fixed logarithmic accuracy. Equation (A.1) shows that the Fourier transform of the k-th power of ln b is proportional to 1/q 2 T times the (k − 1)-th power of the log of the Fourier conjugate variable lnq 2 T (leading log approximation), but also includes terms proportional to all lower powers of this log. The N n LL approximation corresponds to including terms up to j = n in the sum in eq. (A.1), i.e. such that the power of lnq 2 T is by n + 1 units lower than the power of ln(b 2 0 /b 2 ). The NLL and N 2 LL approximations are particularly simple due to the fact that where γ E ≈ 0.5772 is the Euler constant. It follows in particular that if b 0 = 2 e −γ E , the NLL and NNLL terms in eq. (A.1) vanish [12]. A useful form of the N n LL approximation can be obtained noting that Its Fourier transform to LL accuracy is given by This result was given in ref. [5]. One may think that because of eqs. (A.11-A.12) eq. (A.16) with b 0 = 2e −γ E automatically provides a result which is correct to N 2 LL accuracy. This, however, is not true if F (L) is a physical observable, such as a cross-section. Indeed, in this case the N n LL approximation to it is defined by expansion of its logarithm: for example if F (L) is identified with Σ(α s ,ᾱ L) eq. (2.12), the expansion of it to subsequent logarithmic order is given by the expansion eq. (2.15) of S(α s ,ᾱ L) = ln Σ(α s ,ᾱ L), and not of Σ(α s ,ᾱ L) itself. The NLL approximation to the Fourier inverse of F (L) may however be calculated exactly in terms of G(L) ≡ ln F (L). One finds This is the result found in ref. [6].	2008-11-10T14:47:30.000Z	2008-07-24T00:00:00.000	{ "year": 2008, "sha1": "eef418a595f4aede604cc48d3ab8383898bbf082", "oa_license": "CCBY", "oa_url": "https://doi.org/10.1016/j.nuclphysb.2008.09.041", "oa_status": "HYBRID", "pdf_src": "Arxiv", "pdf_hash": "eef418a595f4aede604cc48d3ab8383898bbf082", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
59579914	pes2o/s2orc	v3-fos-license	The life and death of cosmic voids We investigate the formation, growth, merger history, movement, and destruction of cosmic voids detected via the watershed transform code VIDE in a cosmological N-body dark matter {\Lambda}CDM simulation. By adapting a method used to construct halo merger trees, we are able to trace individual voids back to their initial appearance and record the merging and evolution of their progenitors at high redshift. For the scales of void sizes captured in our simulation, we find that the void formation rate peaks at scale factor 0.3, which coincides with a growth in the void hierarchy and the emergence of dark energy. Voids of all sizes appear at all scale factors, though the median initial void size decreases with time. When voids become detectable they have nearly their present-day volumes. Almost all voids have relatively stable growth rates and suffer only infrequent minor mergers. Dissolution of a void via merging is very rare. Instead, most voids maintain their distinct identity as annexed subvoids of a larger parent. The smallest voids are collapsing at the present epoch, but void destruction ceases after scale factor 0.3. In addition, voids centers tend to move very little, less than 0.01 of their effective radii per ln a, over their lifetimes. Overall, most voids exhibit little radical dynamical evolution; their quiet lives make them pristine probes of cosmological initial conditions and the imprint of dark energy. INTRODUCTION Since cosmic voids are, by definition, relatively empty of matter, they offer a unique and pristine laboratory for studying dark energy (Lavaux & Wandelt 2011;Sutter et al. 2012a), exotic fifth forces (Li et al. 2012;Spolyar et al. 2013), and the early universe (Goldberg & Vogeley 2004). They also offer a complementary probe of the growth of structure via their size and shape distributions (Biswas et al. 2010;Bos et al. 2012;Clampitt et al. 2013). Recently large catalogs of voids identified in galaxy redshift surveys (Pan et al. 2012;Sutter et al. 2012b;Nadathur & Hotchkiss 2014;Sutter et al. 2014b) have opened the way for statistical and systematic measurements of void properties (Ceccarelli et al. 2013 However, given the promising utility of voids, we still lack a detailed understanding of their life cycles. For example, for a given void observed at low redshift, we do not know when it formed, where it formed, whether it grew to its present size via simple expansion or through mergers, nor whether it will continue expanding or eventually collapse. We also do not understand basic statistics about voids over cosmic time: their formation and merger rates, growth rates, and movement. Such understanding of the life cycles of voids will solidify current void-based cosmological analysis and enable future probes. Also, if we are to use voids as cosmological probes we must understand the impact of their dynamics on any primordial cosmological signal. For example, void abundances are difficult to predict with excursion set formalisms because small voids tend to c 0000 RAS arXiv:1403.7525v1 [astro-ph.CO] 28 Mar 2014 collapse. While there have been several attempts to improve the initial theoretical result of , such as by adjusting the void growth and destruction parameters (Furlanetto & Piran 2006;D'Aloisio & Furlanetto 2007;Paranjape et al. 2012), rescaling void sizes (Jennings et al. 2013), and introducing a peaks formalism (Russell 2013), there still remains very little correspondence to voids identified with watershed techniques in galaxy surveys (Sutter et al. 2014b). We may improve excursion set predictions by directly measuring the growth and destruction rate in cosmological simulations. The shapes of voids offer a particularly interesting cosmological probe, whether by their distribution (e.g. Bos et al. 2012) or via an application of the Alcock-Paczynski test (Alcock & Paczynski 1979;Ryden 1995;Lavaux & Wandelt 2011;Sutter et al. 2012a). However, these tests rely on the assumption that the void identified in a galaxy survey corresponds to a physical underdensity in the dark matter. While this is largely an issue of sparsity and galaxy bias (Sutter et al. 2014a), the watershed technique may spuriously merge voids even in the dark matter. These voids will erroneously appear as larger voids that are not completely empty and thus have suspect shapes. We can use a detailed merger history to identify such suspect voids. Recently pointed out that for a given tracer population there exists a compensation scale, where the void-matter bias is identically zero. Below this scale, voids generally collapse due to their surrounding overdense walls, while above this scale voids tend to continue expanding (Ceccarelli et al. 2013). However, these results are based on studies of the velocity profiles and clustering statistics at fixed time. Only by tracing the evolution -and thereby studying the dynamics -of voids could one accurately examine the properties of voids in relation to such a compensation scale. Finally, the growth and merger rates of voids are potential cosmological probes, analogous to the growth rate of cosmic structure. The nature of modified gravity and fifth forces can leave fingerprints on the evolution of the void population at high redshift, potentially constraining the properties of dark energy. Unfortunately, to date this remains a largely unexplored topic. Most early studies of voids in simulations focused on visual identification and characterization (e.g., White et al. 1987). For example, the pioneering works of Dubinski et al. (1993), which discussed the process of void merging, and van de Weygaert & van Kampen (1993), which first noted the hierarchical nature of void buildup, were entirely based on visually examining thin slices of N -body simulations. More recent and more sophisticated analyses have focused on void interiors (e.g., Gottlober et al. 2003;Goldberg & Vogeley 2004;Aragon-Calvo & Szalay 2012;Neyrinck et al. 2013) or on statistics at a fixed time such as those discussed above. In this work we present a comprehensive study of the formation, subsequent evolution, and destruction of voids. We use techniques adapted from building halo merger trees (Srisawat et al. 2013) to follow individual voids across cosmic time. This approach allows us to measure their formation time, identify when mergers occur, track their movement and growth, and, when it does happen, record their time of collapse. We translate this information into rates and correlate these rates with void size, which can then inform theoretical and observational results. In the following section we review our simulation setup, void finding approach, merger identification technique, and some definitions to be used throughout the work. In Section 3 we focus on the formation time of voids, followed by a discussion in Section 4 of their growth and merger histories. In Sections 5 and 6 we present an analysis of void movement and destruction rate over cosmological time, respectively. Finally, we conclude in Section 7 with a brief discussion of implications for theoretical modeling of voids and directions for future work. Simulation We study voids forming in a single cosmological dark matter simulation run using the Gadget-3 N -body code (Springel 2005) with initial conditions drawn from the WMAP-7 cosmology (Komatsu et al. 2011). Voids are identified in 62 snapshots from redshift 0 to ∼30. The snapshots are evenly spaced in ln a, where a is the scale factor. The simulation contains 270 3 particles in a box of comoving length 62.5 h −1 Mpc, giving a dark matter particle mass of 9.31×10 8 h −1 M . For more simulation details see Srisawat et al. (2013). Void Finding We identify voids with a heavily modified version of ZOBOV (Neyrinck 2008;Lavaux & Wandelt 2011;Sutter et al. 2012b). ZOBOV creates a Voronoi tessellation of the tracer particle population and uses the watershed transform to group Voronoi cells into zones and subsequently voids (Platen et al. 2007). By implicitly performing a Delauney triangulation (the dual of the Voronoi tessellation), ZOBOV assumes constant density across the volume of each Voronoi cell, which sets the smoothing scale for the continuous field necessary to perform the watershed transform. There is no additional smoothing. The algorithm proceeds by first grouping adjacent Voronoi cells into z ones, which are local basins. Next, the watershed transform merges zones into voids by examining the density barriers between them and joining them together to form ever-larger agglomerations. We impose a densitybased threshold within ZOBOV where adjacent zones are only added to a void if the density of the wall between them is less than 0.2 times the mean particle density. This prevents voids from expanding deeply into overdense structures and limits the depth of the void hierarchy (Neyrinck 2008). However, this does not place a restriction on the density of the initial zone, and in principle a void can have any mean density. The watershed transform identifies catchment basins as the cores of voids and ridgelines, which separate the flow of water, as the boundaries of voids. In sum, we identify voids as depressions in the tracer density; voids are non-spherical aggregations of Voronoi cells that share a common basin and are bounded by a common set of higher-density walls. These we identify the initial zones as the deepest voids, and as we progressively merge voids across ridgelines we can identify super-voids. There is no unique definition of a void hierarchy, and we take the semantics of Lavaux & Wandelt (2011): a parent void contains all the zones of a sub-void plus at least one more. All voids have only one parent but potentially many (or no) children, and the children of a parent occupy distinct subvolumes separated by low-lying ridgelines. Our simulation gives a mean particle spacingn −1/3 ≈ 0.25 h −1 Mpc, which sets a lower size limit of the detectability of voids due to shot noise. For this work we will study all voids with effective radius R eff > 1 h −1 Mpc. We define the effective radius as where V is the total volume of the Voronoi cells that contribute to the void. We do not impose any other cuts based Figure 2. Cumulative number functions for voids in different levels of the hierarchy. Tree level 0 (dark blue) are the topmost parent voids, and tree level 8 (red) are subvoids deepest in the hierarchy. Though higher-level voids tend to be larger, they span a broad range of sizes. At the topmost level, there are only a few small "field" (i.e., childless) voids and some voids that span nearly the entire simulation volume. We do not show these largest voids so that we may highlight the relative differences of the remaining hierarchy levels. on density contrast or minimum density -we wish to see if marginal voids have similar histories as deeper underdensities. Additionally, for the analysis below we need to define a center for each void. For this work we take the barycenter, or volume-weighted center of all the Voronoi cells in the void: where xi and Vi are the positions and Voronoi volumes of each tracer i, respectively. Figure 2 shows the cumulative void number function for all voids in the final a = 1.0 snapshot, organized by level in the hierarchy. Note that these numbers functions do not turn over at small radii, as predicted by , since we are plotting the cumulative, rather than differential, function, and we are only counting voids well above the resolution limit. At the top-most parent root level (Tree Level 0) there are only a few small field voids and the largest voids in the simulation. As we go deeper into the hierarchy, we see increasingly smaller voids. We construct the void tree such that each void has only a single parent (or no parents at all) and can potentially have many children. One void is a parent of another if it shares all zones of the child plus at least one more. Parents can then become children of even larger super-voids. Without any density thresholds, there will be a single void that encompasses the entire simulation volume. However, since we do apply a density threshold, we have multiple root voids. The relatively small simulation box prevents us from examining the very largest voids; however, the large voids that are discovered in this box are representative of the voids found in larger simulations and galaxy surveys. Merger Identification To match voids from one snapshot to another, we use the tree building routine that is part of the publicly available VELOCIraptor (aka STF) package 1 . The VELOCIraptor tree builder code is a particle correlator: it takes two particle ID lists (named A and B) and for each object in list B identifies those objects in list A (i.e., in the previous snapshot) that have particles in common. This first step produces a graph mapping the connections between objects rather than an actual progenitor tree. To produce a tree the algorithm calculates the merit of each connection: where NA i ∩B j is the number of shared particles between the object i in catalog A and object j in catalog B, and NA i and NB j are the number of particles in object i in catalog A and j in B, respectively. Between two snapshots, the unique main progenitor is the object that maximises the merit. This strategy has proven successful in building halo merger trees (see for instance Srisawat et al. 2013). Note that while technically we are constructing bidirectional graphs, we will see that we are justified in calling these trees since the void evolution is surprisingly simple. However, unlike halos, which are defined by the particles they are composed of, voids are defined by the empty spaces between particles. Ideally we would correlate volumes rather than particles. However, this is computationally expensive and fraught with difficulties: it would require modeling the Voronoi volume around each particle and making arbitrary decisions on when one particle's volume is correlated with another. Instead of trying to determine the overlap in volume, we simply use the volume associated with each particle as determined by ZOBOV, v l , to weight the merit function. Thus the modified merit function becomes where the total volume of a void is V = N l v l , and the shared volume isṼ 2 This sum is over all shared particles, with each particle weighted by the volume associated with it in catalogs A and B. In order to qualify as a progenitor the compared voids must share at least 10 particles, though changes to this value do not produce significantly different results. We note that even a simple particle based approach is justified as even our smallest voids contain hundreds of particles. While the cores of voids are empty, there are sufficient numbers of particle distributed throughout the remaining volume that particles can be used as a proxy for the void volume. The volume weighting scheme used here simply reduces instances where a void in catalog A shares particles with several voids in catalog B. The volume weighted merit function chooses the progenitor which minimises the volume fluctuations on a particle by particle basis and for the void as a whole. For the remainder of this work when we examine the movement and growth of a void, we are discussing the evolutionary chain of the main progenitor. VOID FORMATION We begin with visual examination of voids as a function of scale factor, shown in Figure 3. This figure shows slices of the dark matter particles that are identified as belonging to voids. Initially there are only small, isolated voids just above the minimum size threshold. Since we do not apply a density criterion, these are shallow basins that will eventually empty out (as seen in van de . These depressions in the initial density field then begin to expand, with small basins quickly merging with larger basins. As the walls and filaments begin to coalesce around a = 0.2 − 0.4, a complex void hierarchy begins to form as subvoids group into larger parent points. At this epoch the density contrast in the large scale structure becomes high enough for our void finder to identify the multi-level basins as distinct voids. At late times, the larger voids simply grow and expand into their local environment, and since the watershed method includes all particles within the ridgeline as void members, we see very few gaps (e.g., the dense halos) in the void particle distribution. The shallow basins at early times could be considered as "proto-voids". To separate these from mature voids one could define a density threshold, as is done for halos . However, since voids exhibit only linear and quasi-linear evolution, there is no clear distinction between early-and late-time voids: there is a one-to-one mapping from the initial to the final void state (Lavaux & Wandelt 2010;. To clarify this mapping, Figure 4 shows the formation scale factor a f for each void identified at the present day. We define a f as the scale factor at which the void reaches a fraction fV of its current volume. We examine values of fV equal to 10 −4 , 8 × 10 −3 , and 0.2. These fractional volumes are chosen arbitrarily but pertain to specific increases in void size by corresponding to fractional radii of 0.01, 0.05, and 0.1, respectively. All three values give nearly identical results: some voids have persisted since the beginning of the simulation, and others have only appeared recently. As expected, with larger values of fV the distribution skews to later times (note the much smaller value of the green line at a = 0.0), but this is surprisingly insignificant: once a void is detected, it has essentially its present-day volume. There is also an increase in the formation time just before a = 1.0. While this may be due to numerical effects, it is not significantly different than the a f > 0.4 fluctuations. There is a noticeable spike at a ≈ 0.3. This coincides with the initial growth of the void hierarchy; prior to this time there are only isolated voids. Thus this time indicates the appearance of significant structural hierarchy in the cosmic web. It also coincides with the appearance -though not domination -of dark energy: in these epochs, ΩΛ ∼ 0.1ΩM . Thus a large void population forms at these scale factors through a combination of the crystallization of the cosmic web, allowing basins to be identified, and the emergence of dark energy, which shuts off significant continued void production. Since all chosen values of fV give nearly the same results, voids reach their present-day volume quickly -essentially when they are first able to be identified as voidsand do not grow much. We will return to this subject later. In Figure 5 we examine the distribution of void sizes as they appear in the simulation. Throughout most of cos- mic history, the median void formation size is centered on 2 h −1 Mpc, and a small population of medium-scale 10 − 15 h −1 Mpc voids continually appears. The median void formation drops to nearly 1 h −1 Mpcby the present day, since newer voids can only occupy smaller niches in the cosmic web adjacent to larger, expanding neighbor voids. The very largest voids appear at any times, as the walls between midscale voids empty out and the voids are joined into larger super-voids. The fact that there is no noticeable peak formation time for voids of a specific size but there is a peak in the formation time of voids indicates that the level of these voids in the hierarchy changes with time. Although we are showing the present-day a = 1 void sizes, the fact that voids do not grow much over time indicates that this is also essentially their size at formation. VOID MERGERS & GROWTH We use Figure 6 to show an example of a void merger history. In this figure we show slices of the dark matter density centered on a representative void at a = 1.0. We then show the same slice at the same position at two other scale factors, 0.8 and 0.4. In these additional slices we show the progenitors of the present-day void. While the voids in general have complex shapes due to the nature of the watershed (see, for example, Sutter et al. 2012b) we represent them here as simple circles with radii equal to the effective radius R eff . We choose this void to highlight two distinct processes of void evolution. The first, pure merging, occurs when highdensity barriers between two voids completely dissolve due to outflows. When the barrier becomes too low (< 0.2ρ), the separate voids become indistinguishable from each other, forming a single larger void. However, as we will discuss below, this is a very rare process. Instead, what more frequently occurs is annexation of subvoids as a larger parent void assembles, generating a hierarchy. The subvoids remain as distinct and detectable basins, but comprise only a portion of the volume of a larger parent (Figure 1). Figure 6. Evolution of progenitor voids. We show thin slices through the dark matter density with voids superimposed on top. Voids are represented as circles with radii equal to R eff . Slices are arranged from early (left) to late (right) times and trace the evolution of a singe void to highlight different void merger histories. We only show progenitors of the final a = 1.0 void. To increase the density contrast, each panel is scaled such that black and white are the minimum and maximum density of the cells shown, respectively. The density is constructed using cloud-in-cell weighting and the shading is scaled according to log (1 + δ). Projection effects lead to void centers occasionally appearing to lay on top of filaments. Several processes are highlighted by this evolution, including void merging as barriers dissolve, and the formation of a void hierarchy as the larger parent void annexes smaller, but distinct, subvolumes. For each progenitor tree leading up to each presentday void, we can track the total number of progenitors; in other words, the width of the merger tree. In Figure 7 we plot the average number of progenitors Nprog across all present-day voids as a function of scale factor. The fact that the mean number of progenitors is barely greater than one indicates that almost all voids follow only a single line of descent and experience very few mergers. The number of progenitors begins to decrease at a = 0.4, which coincides with the end of the peak formation time (Figure 4) and the relative lack of new voids after that time. In Figure 8 we show the growth rate history, dlnR eff d ln a , for every void in the simulation as a function of scale factor a. We plot a line for each individual void. As expected, the voids with the highest growth rates are the largest; these are the supervoids that form from the rapid merger of subvoids as the basin empties out of substructure. Since they form in deeply-underdense environments with little substantial structure surrounding them, they act as miniature universes with Ωtot < 1. For these larger voids there are a few steep changes as they merge with a smaller void or fragment into smaller progenitors. Even though the merger or fragmentation ratio is small in terms of volume, it can impact the instantaneous growth rate from one snapshot to another. The small-and medium-scale voids show remarkably steady growth histories, with very few strong deviations. Even though some of these voids do merge, they tend to just absorb their subvoids, so the overall volume gained is small. Interestingly, there is a population of collapsing voids: these are the voids located in overall overdense regions. This is the "void-in-cloud" phenomenon of . Even though there are a few small voids with discontinuous merger histories, almost all the small voids are either gently expanding or contracting. We break down the growth rates into secular and merging components, as we show in Figure 9. We define the secular growth rate as the growth rate of voids which did not experience a merger in that timestep. If instead that void Figure 7. Average number of progenitors as a function of scale factor for voids at a = 1.0. We calculate this quantity by counting the total number of progenitors for all present-day voids and dividing by the number of present-day voids. Hence, this quantity becomes less than one as the lines of descent for individual voids end. did merge with another, its growth from snapshot to snapshot is calculated in the average merger growth rate. Here we plot the mean growth rate over all voids as a function of scale factor. We also separate voids into their level in the hierarchy so that we may examine the nature of larger parent voids and their subvoids separately. First we notice that the merger growth rate far outweighs the secular growth rate by an order of magnitude; voids gain volume typically not by growth of the underlying volume but by merging (when it does occur) with adjacent voids. The merger growth rate is much larger for subvoids deep in the hierarchy than it is for voids higher in the tree. Thus, even though a small fraction of larger voids experience occasional large jumps in their volume, averaged over the entire cohort of voids in that tree level it is entirely insubstantial. We see for voids deep in the hierarchy (that is, subvoids) a peak in the merger growth rate at scale factor 0.3, which is the epoch with the highest formation rate where voids experience a rapid restructuring as the hierarchy forms. Comparatively, the secular growth rate is very small and contributes little to the overall growth of voids. Here we see the opposite trend as for the merger-based growth rate: the voids highest in the tree hierarchy have the highest rates, since they are not surrounded by overdense shells that would restrict their growth. Again we see a collapsing void population. Indeed, they have been collapsing since a = 0.3, when the top-level voids first formed. Finally, Figure 10 shows the instantaneous growth rate at a = 1.0. This is an interesting statistic since it allows us to separate collapsing from expanding voids based purely on their recent history. Similar analyses have been done using using the void-matter cross-correlation ) and the identification of universal density and velocity profiles Sutter et al. 2013). While we do not have sufficient volume to reliable apply those measures here, we do see a transition to overall-collapsing voids around 1-2 h −1 Mpc, although there are subsets of collapsing voids at all but the largest scales. A cleaner separation is based on the position in the void hierarchy, as discussed above. The clusterings statistics of estimate a compensation scale roughly around 1 h −1 Mpc, but this is very uncertain due to the small box size. Despite this, we see very good correspondence between that result and the point at which the voids, on average, are collapsing. VOID MOVEMENT Despite the occasional violent merger, all voids experience very little movement of their barycenters (Eq. 2). Figure 11 shows the mean void barycenter velocity, which we define as d\|X\|/d ln a , and which we express as fractions of the void effective radius at the current epoch. The average is taken over the entire void lifetime tracing from its current state along the branch of its main progenitors. The distribution of mean barycenter velocities has two distinct peaks. One, at 10 −3 , represents the movement of the majority of voids, and is remarkably low. This mean velocity gives rise to an average displacement of only a tiny fraction of the void radius. This is not surprising: once a deep underdensity forms from the initial conditions, it is unlikely to move as the voids expand into the surrounding cosmic web. The second peak, at 1 × 10 −2 , is the small population of voids that experience violent mergers. Most of these voids are small subvoids residing deep in the void hierarchy. As before, we see that this is only a small fraction, roughly five percent, of all voids. Even in these cases, the mean displacement is very small, indicating that even when voids experience mergers they are relatively gentle, since the barycenter remains relatively stable. VOID DESTRUCTION We show the destruction rate, or the fraction of the existing void population lost in each snapshot in Figure 12. While initially high prior to a = 0.3, afterwards the destruction rate drops to essentially zero. The initial relatively high destruction rate is not surprising, since at these early times the basins are just beginning to become deep enough to be identified as voids, and there is significant noise in the classification of these objects. However, even at these low scale factors no more than ∼ 4% of voids are lost in every snapshot. But at the same time the formation rate spikes, a = 0.3, the destruction rate plummets. After this epoch, voids that have already existed or will eventually form never cease to exist. Thus even voids that are collapsing aren't squeezed entirely; instead they merely become subvoids of larger parent voids. In fact, the squeezing may help bolster their ability to be detected: as the overdense shells around them grow higher, the density contrasts increase, allowing the void finder to continually detect them. CONCLUSIONS We have performed a comprehensive analysis of the life cycle -covering formation, mergers, growth, movement, and destruction -of cosmic voids. We have adapted merger tree codes originally designed to track the evolution of halos to account for the large spatial extents of voids. By applying this technique to a high-resolution N -body simulation, we have gained a clear picture of voids as dynamic objects in the cosmic web. Through the use of a watershed void finder, we are able to classify voids according to their position in a hierarchy and use that to identify key epochs and scales in their evolution. The past life of a cosmic void depends intimately on its place in the void hierarchy. Voids near the top of the hierarchy primarily form at a scale factor of 0.3, when the density contrasts in the cosmic web become high enough to support their identification and the introduction of dark energy shuts off continued structure formation. These higher-level voids suffer only minor mergers and tend to maintain consistent growth rates over cosmic time. In contrast, voids lying deep in the hierarchy continue to form and have a somewhat more violent life due to the lower-density nature of their surroundings, but even most of these voids have only a single line of descent. The location of a void in the hierarchy is more important than its size: two voids of equal volume can have radically different merger histories depending on their amount of substructure. Voids typically grow at slow rates. However, there is a population of small collapsing voids. These voids tend to live in overdense environments near filaments and walls. Their overdense surroundings slowly squeeze them as adjacent larger voids expand. This picture is consistent with the theory developed by , the velocity inflow-outflow analysis of Ceccarelli et al. (2013), the clustering study of , and the density profile studies of and Sutter et al. (2013). Despite being slowly crushed, after a = 0.3, these voids never get completely destroyed. Instead, they continue to survive as identifiable voids to the present day. Thus the void destruction rate does not play a significant role in the late-time evolution of voids, and can be ignored in theoretical treatments. Additionally, as pointed out by Russell (2013), the collapsing process is completely negligible for all but the smallest voids. Finally, voids don't move much throughout their lifetimes. Only small voids in the frothy depths of the hierarchy that undergo several mergers appear to have perturbed barycenters. Even for these most active of voids, they typically only move a few percent of their effective radii. The combination of small box volume and high resolution limits our study to relatively small (∼ 1 − 15 h −1 Mpc) voids, while voids in larger simulations and galaxy surveys are typically much larger. However, recently Sutter et al. (2013) were able to show that many void properties scale as a function of sampling density and galaxy bias. Thus, properties and characteristics of voids studied in one population of tracers can, in principle, be immediately translated to voids in another population. Thus the conclusions that we reach in this work are generally applicable to voids discovered in other simulations and galaxies. We have examined the properties of voids defined using a watershed technique. There are, of course, other plausible definitions of voids (see, for example, the comparison work of Colberg et al. 2008). These different algorithms might give different pictures of void histories, especially formation times, since they usually impose density thresholds. Additionally, there are other approaches to defining merger trees. We have noticed that volume correlations based on particles can give some non-intuitive results: voids that appear to occupy similar positions (based on their barycenters and effective radii) may not necessarily share any particles. The relationships between particle correlation and barycenter definition should be investigated further. However, we have applied other merger tree algorithms, such as MergerTree and JMerge (both described in Srisawat et al. 2013), and found qualitatively similar results. Overall, voids live far quieter lives than their overdense counterparts, the halos. Whereas up to 20% of halos have suffered a recent major merger, voids experience essentially no major mergers throughout their lifetime. Likewise, while subhalos can be stripped of their mass as they pass through a larger parent halo, subvoids continue to be identifiable even when a supervoid forms around them. The implication is that voids are a much more pure cosmological probe; the fundamental cosmological signal imprinted from initial conditions and modified by dark energy is not corrupted by significant dynamics. Thus lower-redshift cosmological probes, such as the Alcock-Paczynski test and void-galaxy crosscorrelations, will not be affected by recent spurious mergers in the void population.	2014-09-20T08:08:49.000Z	2014-03-28T00:00:00.000	{ "year": 2014, "sha1": "c34568975afb53a73448813791153a3abc59cc71", "oa_license": null, "oa_url": "https://academic.oup.com/mnras/article-pdf/445/2/1235/18197933/stu1845.pdf", "oa_status": "BRONZE", "pdf_src": "Arxiv", "pdf_hash": "c34568975afb53a73448813791153a3abc59cc71", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
51942235	pes2o/s2orc	v3-fos-license	Experience of Domestic Violence and Psychological Morbidity in Spouses of Alcohol-Dependent Males Background: Prevalence of both domestic violence (DV) and alcohol use is reported to be high in Kerala. The prevalence of DV and psychological morbidity in spouses of alcohol-dependent males has not been studied objectively. Methods: This cross-sectional study was undertaken to study the occurrence of DV and psychological morbidity–major depressive disorder (MDD), anxiety disorders, and adjustment disorders–in spouses of alcohol-dependent males attending the de-addiction center of a tertiary care hospital in South India. Sixty consecutive cases, aged 18–55 years, were recruited after getting informed consent. They were assessed using Domestic Violence Questionnaire (DVQ), Mini International Neuropsychiatric Interview Schedule, and a questionnaire to assess adjustment disorder. The association of DV with psychological morbidities was also studied. Results: DV was reported by 41 (68.3%, 95% confidence interval [CI] = 55.0–79.7) cases. At least one psychiatric morbidity was observed in 51 (85.0%, 95% CI = 72.9–92.5) cases–MDD in 15 (25.0%, 95% CI = 15.1–38.1), anxiety disorders in 6 (10%, 95% CI = 4.1–21.2), and adjustment disorder in 32 (53.3%, 95% CI = 40.1–66.1) cases each. No statistically significant association was observed between DV and any of the psychiatric disorders. However, DVQ scores showed significant correlation with years of marriage (Pearson's r = 0.268, P < 0.05) and with stressful life events over the past 1 year (Pearson's r = 0.424, P < 0.05). Conclusions: High rates of DV and psychological morbidity were seen in spouses of alcohol-dependent males. INTRODUCTION Violence against women is a major human rights abuse and an important public health concern. The World Health Organization (WHO) Multi-country Study on Women's Health and Domestic Violence found the lifetime prevalence of physical and/or sexual partner violence among ever-partnered women in 15 countries to range from 15% to 71% and it was found to be 4%-54% in the past year. [1] A multisite household survey conducted in India found that about 26% of women reported physical violence in the past 12 months. [2] In this study, among urban nonslum sites, the highest prevalence rates for physical and psychological violence (43.1% and 61.6%, respectively) were reported from Thiruvananthapuram, Kerala. [2] Various studies have found that partner alcohol abuse was associated with an increased risk of intimate partner violence (IPV) in women. [3][4][5] A study conducted in Pune, India, had assessed IPV using a 4-item questionnaire which assessed physical and emotional abuse, but not sexual violence. [5] Spousal domestic violence (DV) against women is reported to have far-reaching mental health implications in the victims. Depression, posttraumatic stress disorder (PTSD), anxiety, self-harm, and sleep disorders are well documented to be mental health sequelae of IPV. [6,7] The Study of Abuse in the Family Environment (India-SAFE) reported that alcohol consumption by husband and exposure to physical violence were significantly associated with an increased risk of poor mental health in women. Psychological morbidity was not evaluated objectively, but assessed using a self-reported questionnaire in this study. [8] A study done in Palakkad, Kerala, found that DV was reported by almost two-thirds of spouses of alcohol-dependent males attending a tertiary care center, but the questionnaire used to assess DV was not validated in the local language. On clinical evaluation, two-thirds of these spouses had clinical depression and 11% had suicidal thoughts. [9] In another study, sixty spouses of men with alcohol dependence (AD) were assessed using a structured clinical interview schedule. They found that 65% of the participants had a psychiatric disorder; mood and anxiety disorders were the most common problems reported. [10] A population-based survey done in northern part of Goa assessed women aged 18-49 years using General Health Questionnaire and reported that excessive partner alcohol use led to a two-to three-fold increase in risk for common mental disorders. On logistic regression, they concluded that partner violence had a mediating effect on the association between partner's excessive alcohol use and women's common mental disorders. Increasing age, poor education, and lack of paid employment were also found to increase the risk for common mental disorders in this population. [11] A community-based study conducted in a slum in Kolkata found that alcohol addiction of husband, below poverty line (BPL) status, lack of social support, and property in women were significantly associated with DV in ever-married women of reproductive age group. [12] Of the women attending the adult outpatient unit of a national institute, 56% reported a history of IPV, out of whom, 99% met the criteria for depression and 12% for PTSD. In this study, alcohol use in spouse, harmful use, and lower education of spouse were found to be associated with IPV. [13] Kerala has one of the highest rates of alcohol consumption in India. In a recent survey, high prevalence of alcohol use disorders in males was reported in Kerala. [14] The India-SAFE study had reported high rates of DV from Thiruvananthapuram. [2] In a case-control study done in Government Medical College, Thiruvananthapuram, to assess DV as a risk factor for attempted suicide in married women, it was found that irrespective of the case/control status, DV was reported to be high (almost 65%) in spouses of alcohol-dependent males. [15] None of the studies done in Kerala had assessed DV using a questionnaire validated in the local language. Psychological morbidity was also not evaluated objectively using structured interview schedules in these studies. In this context, this study was undertaken with the primary objective of assessing the occurrence of DV in spouses of alcohol-dependent males attending the de-addiction center of a tertiary care hospital in South Kerala. The secondary objectives were to assess the psychological morbidity in this populationdepressive disorders, anxiety disorders, and adjustment disorders-and to study the association of DV and other sociodemographic and clinical variables with these psychiatric disorders. MATERIALS AND METHODS A cross-sectional study was undertaken in the de-addiction center of a teaching hospital in South Kerala for 9 months. Approval was obtained from the Human Ethics Committee of the institution. Participants were spouses of alcohol-dependent males attending the de-addiction center and belonged to the age group of 18-55 years. Those with psychotic disorders, as diagnosed with the Diagnostic and Statistical Manual of Mental Disorders-IV Edition-Text Revision (DSM-IV-TR) criteria, [16] using Mini International Neuropsychiatric Interview Schedule (MINI)-Malayalam Version, [17,18] and those who refused or were unable to give informed consent were excluded from the study. Sample size was calculated using the following formula: Where Z 1− /2 = the standard normal variate at a Type 1 error of 5% and is 1.96 P = the expected prevalence of DV in similar studies, taken as 65% [15] d = relative precision, taken as 20% of P. It was calculated to be 52, and rounded off as 60. Consecutive individuals who satisfied the inclusion criteria were recruited for the study after obtaining informed consent. Three cases were excluded from the study as they were psychotic and three cases refused to give consent. Confidentiality and privacy were ensured during the interview. Those who reported DV were sensitized regarding the legal provisions available for help and given necessary guidance to seek such aids. This issue was also addressed in the therapy of alcohol-dependent males. Those who were detected to have psychiatric morbidity were referred for appropriate care. Using a pretested pro forma, sociodemographic variables of the participants were assessed. DV was defined as any act of physical, psychological, or sexual abuse by a husband toward his spouse over the past 12 months; measured by a cutoff score of 5 or more using the Domestic Violence Questionnaire (DVQ). This is a 20-item questionnaire which has been validated in the local language. It has 12 items to assess psychological violence (including social and economic restriction, insulting in public, neglecting, expressing suspicions regarding fidelity, and being unfaithful toward spouse); six items to assess physical violence (threatening to harm physically, slapping, beating, twisting arm or pulling hair, kicking or dragging, and choking or inflicting burns); and two items to assess sexual violence (ignoring by not having sexual intercourse and having forcible sexual intercourse). Scoring is done from 0 to 4 based on the frequency of exposure to the act over the past 12 months (0 -never, 1 -once/twice, 2 -three to five times, 3 -six to ten times, and 4 -11 times or more). At a cutoff score of 5, sensitivity was 89.5% and specificity was 87.2%. Cronbach's alpha was 0.92. [19] Depressive and anxiety disorders were diagnosed as per the DSM-IV-TR criteria, [16] using MINI Malayalam Version. [17,18] Adjustment disorder was diagnosed as per the DSM-IV TR criteria, using a questionnaire validated in the local language for another study. [20] Other variables such as the number of years after marriage, duration of onset of AD in husband, duration of treatment for AD, period of abstinence, family history of psychiatric illness in spouses, stressful life events over the past 1 year, social support, and religious beliefs were also assessed. Stressful life events were assessed using Presumptive Stressful Life Events Scale (PSLES) which has been validated for the Indian population. Each life event is given a score. The total score is taken as a measure of the stressful life events that the person is exposed to over the past 12 months. Those with scores above the median cutoff were taken to have greater stressful life events. This questionnaire had been translated into the local language and used for other studies. [21] Perceived social support and religious beliefs were assessed using questions designed to assess the same, which had been used in a previous study. [15,22] Four questions in the local language were used to assess perceived social support. These items inquire about having neighbors/friends to seek help from, having relatives to support, getting support from organizations/agencies, and being a member of organizations. Each item is scored from 0 (no) to 1 (yes). The median cutoff score was taken to dichotomize the variable as having good/poor social support. [22] Religious beliefs were assessed using three questions prepared in the local language. The questions were the following: "Do you believe in God?" "Do you engage in prayers regularly?" and "Do you go to temple/mosque/church regularly?" The first question was scored as 0 for no and 2 for yes. The other two questions were scored from 0 to 2 based on the frequency. The median score was taken as the cutoff and the variable was dichotomized as good religious beliefs for scores above the cutoff and poor for those below. [22] Statistical analysis The data were entered in MS Excel Version 7.0 (Microsoft Corp., USA), cleaned, and edited. Analysis was done using R version 2.13.1. [23] Mean and standard deviation (SD) are provided for continuous variables and proportion and 95% confidence interval (CI) for categorical variables. Association was assessed using odds ratio (OR) and significance was tested using Chi-square test or Fisher's exact test as appropriate. Pearson's product moment correlation coefficient or point bi-serial correlation coefficient was used to assess the correlation between different variables depending on whether the variables were continuous or dichotomous, respectively. RESULTS The DISCUSSION This study assessed the occurrence of DV and psychological morbidity in spouses of alcohol-dependent males, belonging to reproductive age group, attending the de-addiction center of a tertiary care center in South Kerala. It was found that more than two-thirds of the sample (68.3%) had experienced DV over the past 1 year. The India-SAFE study had reported such high rates of DV from Thiruvananthapuram. [2] A hospital-based study done in Palakkad, Kerala, had reported similar rates of DV in spouses of alcohol-dependent males. [9] A community-based study done in Karnataka, in spouses of alcoholics, had found higher rates of 74%-92%. PDC -Predegree course; VHSC -Vocational Higher Secondary Course The instrument used for assessing DV was only face validated for this study. [24] In our study, the questionnaire used to assess DV had been validated for the local population and had an optimum cutoff score with good sensitivity and specificity. That could be the reason for the difference in the rates reported. At least one of the psychiatric disorders studied-MDD, anxiety disorders, or adjustment disorders-was observed in 85.0% of the participants. MDD-current, past, and recurrent-was seen in 25%, anxiety disorders in 10%, and adjustment disorders in 53.3% of the sample. The study done in Palakkad reported that two-thirds of the spouses of alcohol-dependent males had clinical depression. They had assessed clinical depression using a questionnaire, [9] while in our study, the diagnosis of MDD was made using MINI. Hence, a modest rate of the diagnoses is observed. In 2013, Kishor et al. assessed sixty spouses of alcohol-dependent males and reported that 65% of them had a psychiatric disorder. Mood disorders and anxiety disorders were more common; MDD was reported by 43.3% and panic disorder in 15%. They had also used a structured clinical interview schedule for making the diagnosis. [10] In our study, no significant association was observed between DV and MDD, anxiety disorders, adjustment disorders, or any of these psychiatric morbidities. The OR for MDD was 2.21 and any psychiatric disorder was 1.09, but they were not statistically significant. This suggests that probably our study did not have adequate power to assess the association of DV with these variables. For anxiety disorder, the OR was 0.92 and for adjustment disorder, it was 0.76; both of which were not significant. The India-SAFE study had found AD and DV to be significant risk factors for poor mental health in women. [8] In a population-based survey, Nayak et al. found that violence-related attitudes of the partner increased the odds of mental illness in women by almost three times, on multivariate analysis. [11] In the Kolkata-based study, it was found that poor social support was associated with an increased risk of DV in ever-married women. [12] None of the other variables showed significant association with either DV or psychological morbidity in spouses of AD males in our study. DVQ scores had shown positive correlation with years of marriage and PSLES scores. This suggests that DV increased with increasing age in reproductive age group. With increasing age, women might either perceive DV to be more or have less inhibition in disclosing it. Stressful life events experienced by women over the past 1 year also increased with increasing DV over the same time period. Increased stress in women who experience DV might be the mediating factor that increases the psychological morbidity in this population. High rates of DV and psychological morbidity were observed in spouses of alcohol-dependent males, belonging to reproductive age group. Occurrence of psychological morbidity in those with exposure to DV might be mediated through increase in perceived stress in this population. Strengths and limitations of the study This study was done in a homogeneous group of spouses of alcohol-dependent males, of reproductive age group. The study variable, DV, was assessed using a validated questionnaire. Psychiatric diagnoses were also assessed by an interview schedule translated into the local language. Hence, valid measures of the study variables could be obtained. This being a hospital-based study, the findings cannot be generalized to the community. As there is no control group, comparison of the rates of DV and psychological morbidity with spouses of nonalcohol-dependent males could not be done. CONCLUSIONS More than two-thirds of the spouses of alcohol-dependent males, belonging to the reproductive age group, had experienced DV over the past 1 year. High rates of psychiatric morbidity were reported by these women. Adjustment disorder was reported by more than one half of them; a quarter of them had MDD and 10% had anxiety disorders, while 85.0% had any one of these psychiatric morbidities. Moreover, stressful life events over the past year were found to increase with increasing scores of DV, suggesting the former's role in mediating the effect of DV on the mental health of women. This study highlights the need for assessing the exposure to DV and the mental health of spouses of alcohol-dependent males. Addressing the mental health needs of this population could also improve the treatment outcomes of males with AD. Further community-based and case-control studies would help to delineate the causative role of DV in psychiatric morbidity in spouses of alcohol-dependent males.	2018-08-14T20:37:25.872Z	2018-07-01T00:00:00.000	{ "year": 2018, "sha1": "949034293d3974e2a587ba3068d1e5aed63a2ea1", "oa_license": "CCBYNCSA", "oa_url": "https://journals.sagepub.com/doi/pdf/10.4103/IJPSYM.IJPSYM_38_18", "oa_status": "HYBRID", "pdf_src": "WoltersKluwer", "pdf_hash": "6d58c912da6a6d2c6d572aaff74ff13ef0241d76", "s2fieldsofstudy": [ "Medicine", "Psychology", "Sociology" ], "extfieldsofstudy": [ "Medicine" ] }
253040633	pes2o/s2orc	v3-fos-license	Duane Retraction Syndrome With Mechanical and Innervational Upshoot and Secondary Superior Rectus Contracture: A Surgical Challenge Duane retraction syndrome (DRS) with mechanical and innervational upshoot poses a surgical challenge. We discuss a case of DRS with mechanical and innervational upshoot and its surgical management. An 11-year-old boy presented with left eye upward deviation since birth. This deviation was worst on the right gaze. His best corrected visual acuity was 6/6 OD and 6/60 OS. Refraction showed low hyperopia with low astigmatism in both eyes. Stereoacuity was absent and there was suppression on the Worth 4 dot test in the left eye. The left eye had large hypertropia of 50 prism diopter in primary gaze. Extraocular movements showed severe upshoot and narrowing of palpebral fissures on adduction and limited abduction (-2). The patient underwent Y-splitting of the left lateral rectus (LR) muscle of 10 mm, LR recession of 4 mm, and left eye superior rectus recession of 12 mm. A marked reduction in hypertropia in primary gaze was observed on day one and at two months postoperatively with residual upshoot on adduction. His left eye deviation remained stable after six months postoperatively. Introduction Duane retraction syndrome (DRS) is a rare entity and its incidence in patients with strabismus is less than 5% [1]. Concomitant contraction of the lateral rectus (LR) muscle in adduction results in characteristic globe retraction with narrowing of the palpebral aperture and upshoot and/or downshoot in adduction. Abduction deficiency is frequently observed as agenesis of the abducent nucleus is a well-known theory for its underlying etiology [1]. The paradoxical synergistic innervation of the LR muscle by different nerves is evidenced by electromyographic (EMG) studies, and DRS has been further classified into Duane I, II, and III [2]. In 1988, Ahluwalia et al. further classified DRS according to the position of the affected eye: esotropia DRS, exotropia DRS, or orthotropic DRS [3]. The abnormal innervation may involve vertical muscles in the affected eye as demonstrated by an associated overshoot of the eye either over elevation or over depression, or in other words, upshoot or downshoot. These phenomena are reported in approximately 25-39% of patients with DRS [4]. The underlying mechanism for vertical movement abnormality can either be mechanical, innervational, or both [5]. Tight LR on the globe and co-innervation of the vertical or oblique muscles with the LR muscle are said to be the mechanisms involved. Our patient had a large vertical deviation in the primary gaze, a disfiguring overshoot occurred abruptly in adduction. This clinical finding is attributed to both mechanical and innervational pathology, which is a rare finding. The surgical approach for our patient was directed toward (1) hypertropia in the primary gaze and (2) severe and disfiguring upshoot that highlighted the need for vertical muscle intervention. Thus, we report a rare case of DRS with disfiguring severe upshoot (mechanical and innervational-type) and its challenging surgical correction. We believe this case report will assist in decision-making in terms of planning for surgery for this rare type of DRS. Case Presentation An 11-year-old boy presented with an upward deviation of the left eye since birth. This deviation was worst on the right gaze. On examination, his best corrected visual acuity was 6/6 OD and 6/60 OS. Refraction showed low hyperopia and astigmatism in both eyes. The left eye was hypertropic while the right eye was fixated on distant objects in primary gaze. Extraocular movements examination revealed evidence of globe retraction with narrowing of the palpebral fissure and severe upshoot on attempted adduction. On levoversion, the left eye had limited abduction (-2) and these findings are clinically well described in DRS type III ( Figure 1). The patient underwent left eye superior rectus recession of 12 mm and 4 mm left LR recession with Ysplitting of LR muscle 10 mm from origin insertion. Superior rectus recession was done via the direct scleral technique. Traction suture Vicryl 5/0 was placed at the superior rectus muscle insertion to pull the globe inferiorly to achieve maximum exposure to enable visualization. Immediate postoperative examination on day one revealed the reduction of hypertropia of the left eye in primary gaze and levoversion (Figure 2). At the subsequent follow-up two months postoperatively, the residual hypertropia on primary gaze and on levoversion was stable with residual upshoot of the left eye on adduction despite a large 12 mm of superior rectus recession ( Figure 3). The prism cover test showed 18 PD base down and 16 PD base in. Six months later, his left eye condition remained the same, and the prism cover test remained stable ( Figure 4). Discussion DRS is a complex and rare form of congenital strabismus; its incidence in patients with strabismus is less than 5% [6]. DRS is easily recognized clinically because it has a pronounced abduction deficit with varying degrees of adduction limitation, globe retraction with narrowing of the palpebral opening, and oblique elevation or depression on attempted adduction [6]. Despite the fact that the Huber classification is wellknown, paradoxical contraction due to variable degrees of medial rectus (MR), inferior oblique (IO), and vertical rectus muscle innervation is often overlooked [2]. As a result, the term "atypical DRS" is used to describe clinical conditions that do not fit the Huber classification. Variable degrees of upshoot or downshoot movements on adduction and hypertropia in the primary position constitute the evidence for cocontraction of vertical rectus muscle in DRS, which Huber does not explain in his classification [7]. The superior rectus muscle's co-innervation with the LR muscle, or mechanical factors such as the bridle or leash effect created by tight LR, can cause an upshoot or downshoot [8]. When the globe adducts and moves above or below the horizontal plane, the tight LR slips, causing an upshoot or downshoot. In extreme circumstances, the "knife edge effect" has been characterized as manifesting even with the slightest adduction movement. Furthermore, a long-standing upshoot in the main position with substantial hypertropia may have led to the development of superior rectus contracture and subsequent fibrosis as demonstrated in our patient [8]. DRS is defined as a defect in nerve innervation of the LR muscle caused by the absence of the abducent nucleus or nerve, resulting in oculomotor nerve co-innervation and co-contraction [1]. Evidence from MRI studies and histological discoveries supports this postulation [9]. This result backs previous EMG investigations that showed a paradoxical contraction of the LR muscle during adduction [2]. In addition, DRS is inherited as an autosomal dominant trait in 10% of instances, and hence genetic factors play a part in its etiology [10]. To treat DRS type III, a variety of surgical approaches have been proposed. The recession of both the LR and MR of the affected eye can be used to treat unilateral DRS type III with upshoots or downshoots on adduction. Posterior fixation suture of the horizontal rectus, lowering of the insertion of the LR muscle, and vertical rectus recession are all surgical techniques used in DRS. Other techniques include LR recession with bifurcation (Y-split), in which the LR is split into two for 10 mm from the insertion and vertically transposed [10]. Secondary superior rectus contracture is treated with a large recession of the superior rectus muscle [11]. Arora et al. have documented a similar presentation in a 23-year-old female. Their team performed bilateral LR recession of 6 mm, Y-split of the LR, and superior rectus recession of 6 mm. The outcome at three months was good, with greatly reduced hypertropia [8]. In contrast, in our case, only unilateral LR recession and large superior rectus recession were performed as compared to bilateral LR recession and moderate superior rectus recession in the above-mentioned case. However, a good outcome was similarly observed in our patient. Our patient had DRS with considerable upshoot, significant hypertropia in the primary position, and superior rectus contracture. The condition was successfully treated by using superior rectus recession, LR recession, and Y-splitting of the LR muscle. Our patient had residual hypertropia in the primary position after having achieved maximum superior rectus recession. One of the difficulties in managing DRS is the associated superior rectus contracture [12]. Conclusions DRS can be linked to the adduction upshoot or downshoot of the affected eye. Hypertropia on primary gaze in DRS is rare. Clinicians should be aware of the likelihood of executing a combination of surgical techniques and handle their cases accordingly. Additional Information Disclosures Human subjects: Consent was obtained or waived by all participants in this study. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.	2022-10-21T15:33:21.172Z	2022-10-01T00:00:00.000	{ "year": 2022, "sha1": "fe98f0a7b189f8e4751bbc14b145fd6f378e60b4", "oa_license": "CCBY", "oa_url": "https://www.cureus.com/articles/120213-duane-retraction-syndrome-with-mechanical-and-innervational-upshoot-and-secondary-superior-rectus-contracture-a-surgical-challenge.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "5a700ee23119f53b169d93ed2590882491fd69ab", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
9451201	pes2o/s2orc	v3-fos-license	Electromagnetic field with constraints and Papapetrou equation It is shown that geometric optical description of electromagnetic wave with account of its polarization in curved space-time can be obtained straightforwardly from the classical variational principle for electromagnetic field. For this end the entire functional space of electromagnetic fields must be reduced to its subspace of locally plane monochromatic waves. We have formulated the constraints under which the entire functional space of electromagnetic fields reduces to its subspace of locally plane monochromatic waves. These constraints introduce variables of another kind which specify a field of local frames associated to the wave and contain some congruence of null-curves. The Lagrangian for constrained electromagnetic field contains variables of two kinds, namely, a congruence of null-curves and the field itself. This yields two kinds of Euler-Lagrange equations. Equations of first kind are trivial due to the constraints imposed. Variation of the curves yields the Papapetrou equations for a classical massless particle with helicity 1. Introduction Quantum mechanics provides exhaustive description of motion of a particle in limited scales, when typical length of run is comparable with the wavelength, normally, in atomic and sub-atomic ones. When considering motion of a particle in scales which are apparently non-comparable with typical wavelength, that exhaustive quantum mechanical picture becomes less convenient, and to use it, one passes to asymptotical behavior of incident and scattered waves. In astrophysical scales, particularly, when studying deflection of light in gravitational field, classical mechanics is evidently more convenient and one prefers to consider photon as a classical massless particle drawing a null-geodesic in the space-time. This approach to light propagation in curved space-time is quite satisfactory while photon is considered as a scalar particle. If, however, its polarization becomes important the question arises, how to include it into the classical mechanical description. If the wavelength is big enough, i.e., photon momentum is relatively small, the commonplace classical mechanical description becomes incomplete. The point is that under some conditions spin of the particle becomes comparable with some components of its orbital momentum, which usually are regarded as zero. This may happen, for example, when describing a light beam incident to a Schwartzschild black hole. Due to the classical mechanical considerations each photon of the beam has zero longitudinal component of orbital momentum. However, its spin is also longitudinal and, hence, contributes the total longitudinal component of angular momentum. Since, on one hand, spin is always collinear to momentum of photon, and, on the other hand, gravitational deflection of the beam changes the momentum, this description contradicts the conservation law of total angular momentum. It is clear that in order to have the correct picture one should take spin of photon into account such a way that the total angular momentum conserves. If it is done the sum of longitudinal components of orbital and inner momenta is constant, thus, the earlier is not zero after deflecting. Therefore, if photon momentum is small enough, this effect can change the shape of scattered beam. Thus, there exist situations when the well-known corpuscular theory does not work. While in case of spinless particle one can make a choice between the usual (corpuscular) and wave mechanics due to scales under consideration, in case of particle with helicity the earlier does not work, so, the only possibility is to use the latter. The latter provides description in terms of special functions, say, Legendre polynomials, which are convenient while their powers are not very high. However, in some real situations powers of the polynomials are of the order of astronomical distances measured in wavelength units. In these situations corpuscular description in which spin of the particle is taken into account properly, would be much more convenient. The goal of the present work is construct a model of photon with given momentum and helicity ±1 in a curved space-time. Statement of the problem The desired model is assumed to provide certain world line of a massless particle which has helicity 1 and, at the same time, description of electromagnetic field which everywhere draws a locally plane wave. The main difficulty is that spin of the particle is quantized, thus, the desired construction must contain both classical and quantum degrees of freedom. The notion of classical particle with quantum spin seems to be one of the simplest systems of mixed classical-quantum nature, and attempts to built correct theory of this object are lasting for decades [1]. Our task is somewhat wider, because we try not only to obtain equation of the particle world line, but also a locally plane wave, in other words, to describe propagation of circularly polarized photon in terms of both geometric and wave optics. An attempt to build such a model was made in the works [2,3] due to the problem of light propagation in Schwartzschild space-time. Since electromagnetic field presents in the construction a new question arises, how to combine the field in the space-time and helicity which must be attached to the unknown world line. In the work [2] electromagnetic field was removed by a vector field which was defined on the world lines. This substitution made it possible to combine wave and world line under assumption that if the lines are found properly then the fields attached to them constitute the entire electromagnetic field in the space-time. In the present work we revise this approach. Instead of specifying a functional space of curves and attaching a vector field to each curve, we assume that the same result could be obtained from the pure electromagnetic Lagrangian. The main idea of this work is that, after all, both geometric and wave optics should follow from pure electromagnetic theory, therefore, the model should be built in the framework of the theory of electromagnetic field. To do it, we start with the well-known form of Lagrangian of electromagnetic field and restrict the functional space of the field variables with its subspace of the fields behaving as locally plane waves. Since restrictions of this sort are known as constraints we assume that putting relevant constraints on the Lagrangian leads to special Euler-Lagrange equations for the waves in question, and the desired description containing both geometric and wave optics follows from it. Locally plane wave and associated orthonormal frame The notion of plane wave in space-time differs from that in space where the wave vector is orthogonal to the hyperplanes, cannot be tangent to them. In space-time the wave vector is orthogonal to the hyperplanes and, at the same time tangent to them. To see this consider flat space-time and Cartesian coordinates {t, x, y, z} in it which are chosen such a way that the wave has phase φ = ω(t − z). The phase takes constant values on luminal hyperplanes t = z and the wave vector e − = 2 −1/2 (∂ t + ∂ z ) tangent to the hyperplanes: At the same time the wave vector is orthogonal to the hyperplanes because, due to pseudo-Euclidean metric any null-vector is orthogonal to itself. The wave propagates along this vector, therefore e − must be identified with the vector of velocity and plays the role of velocity of photon in corpuscular model. Now we use these two objects to construct an orthonormal frame associated with the wave. By construction, there exists an isotropic vector e − tangent everywhere to the surfaces of constant phase, and, therefore, orthogonal to them. We introduce one more isotropic vector e + whose direction is arbitrary, requiring only that its scalar product with the vector e − is equal to one. As it is done we can introduce two unit space-like vectors e α which are orthogonal to each other and to e ± . The four vectors defined this way constitute a local orthonormal frame. In Minkowski space-time and Cartesian coordinates considered above the space-like vectors e α , α = 1, 2 are defined as follows: They are also tangent to the wave fronts and orthogonal to them. These vectors are used for specifying polarization of the wave. The four vectors e α , e ± where coordinates are chosen such that e + = 2 −1/2 (∂ t − ∂ z ) form an orthonormal frame with metric The frame of 1-forms dual to it, is Now, let us pass to a curved space-time and account main features of an electromagnetic wave which can be called locally plane and monochromatic. We start with a wave which possesses locally a scalar function φ called phase. Gradient of this function is an isotropic 1-form, and its (hyper-)surfaces of level are wave fronts, i.e., it is possible to introduce local Cartesian coordinates in which the wave can be represented locally the way just discussed. This is possible under some special condition, due to which the wavelength is much less than typical scales specified by the space-time curvature. Hereafter we assume that this condition is satisfied. The field of orthonormal frames associated with the wave can be constructed similarly. As this is done it remains to fix one detail. The point is that the metric of this frame given by the equations (2) remains unchanged if one of isotropic vectors is multiplied by an arbitrary factor and another one is divided by it. Employing this operation we can fix action of the vector e − on the phase: while, by construction, the equation (1) remains in force. Now, as the orthonormal vector frame associated with the wave {e ± , e α } is defined, one can find out the orthonormal covector frame {θ ± , θ α } dual to it and the connection 1-form ω b a· ≡ γ b ca· θ c for the frame. Hereafter we assume that this is done, and pass to considering constraints whose imposed on electromagnetic field leaves only locally plane waves. Electromagnetic field with constraints If the field in question draws everywhere a locally plane wave there exists a congruence of null-curves {x i (s)} which are integral lines of the vector e − . Since this vector is identified with the velocity the wave vector is collinear to it, and potential of the field α has no ± components, so, we have a constraint given by The vectors e 1 , e 2 are chosen to specify polarization of the wave, hence, their span does locally the instant (two-dimensional space-like) wave front. Therefore we can neglect changes of the potential in their directions. This constraint can be written as where the operator e β • stands for differentiation along the vector e β . By analogy with plane waves in Cartesian coordinates the amplitudes can be represented in the form where the contants a β are complex numbers chosen such a way that the wave has left or right circular polarization and the function φ, specifies the phase of the wave. In these denotions the constraint (5) coincides with the equation (1). Finally, all the constraints imposed above (1, 3-7) reduce to the following: the 1-form of field potential has only one non-zero derivative where φ and ω do not depend on the parameter s. Action principle for constrained fields The action functional for electromagnetic field is given by the well-known integral where α is 1-form of potential of the field, <, > stands for scalar product and ε denotes the unit 4-form: ε ≡ ε ijkl /4! θ i ∧θ j ∧θ k ∧θ l that corresponds to four-dimensional integration in the space-time. Straightforward computation of variation of the action (8) yields Maxwell equations, valid for all possible shapes of electromagnetic field. Our goal is to restrict the functional space of the field with subspace of fields which draw everywhere locally plane waves by imposing constraints ((1, 3-7)). For this end we expand the quadric form < dα, dα > in the frame {θ} as follows: Though most of terms of the expansion are zero, some of them have non-zero variations. Substituting the constrained Lagrangian into the action integral (8) yields: where we take into account the fact that for convenience we use complex valued field components. As usual, the components A and their complex conjugatesĀ are regarded as independent variables. Due to the constraints ((1, 3-7)) the Lagrangian under integral (9) can be transformed as follows: The third term can be ignored because, as will be shown below, action of the vector e − annulates the field, consequently, the expression e − •Ā , e − • A is product of two zero factors, hence, both the third term itself and its variation are identically zero. Though, due to the constraint (1) the termȦ (andȦ) is equal to zero, we do not ignore it because its variation plays important role in the action principle. Thus, finally, the action integral has the form Helicity of constrained fields The action integral is evidently invariant under rotations of the frame in the plane formed by the vectors e 1 and e 2 . This invariance yields some conservation law due to the Noether theorem. To find the law we consider the change of the form of the action integral under rotations of the local frames specified by an infinitesimal matrix δη a b (s) defined as a function of the parameter s on each curve. Rotation of the frame changes components of the vectors A,Ā but does not change the vectors. Thus, the first term of the Lagrangian (10) containing scalar product < A,Ā > does not contribute variation. At the same time the rotation changes derivative ofȦ: where the covariant derivative D b (δη a c· ) is exactly variation of the connection 1-form. The first term in the variation ofȦ does not contribute variation of the action due to field equations. By the result, variation of the action is ε. So, due to the Noether theorem we obtain conserved spin current with single non-zero component where is the spin tensor of the wave. Due to the constraint (6) it is zero for linearly polarized waves, and for circularly polarized waves has single non-zero element S 12 = ±ω\|a 1 a 2 \|, whose sign depends only on helicity. The only consequence of this result we need is that the spin has single non-zero component, and as for its magnitude, we accept its quantum value 1 in dimensionless units. Note that the spin is always pointed along the vector of velocity, so no special equation is needed for it. This fact provides implementation of the Tulczyjew constraint which requires that conversion of the particle momentum with its spin is zero [5]. Field equations Now we return to the action functional (10) and compute its variations only under small variations of the field which has two components A 1 and A 2 . Variation of the first term in the constrained Lagrangian is identacally zero because it contains the factor θ + , θ + which is not varied, therefore we ignore it. Variation of the rest part of the action is where action of vectors e ± is considered as differentiation along the vectors. The next step is to extract total derivatives: Note that due to the metric of the null-frame (2) combinations like e + • V − are parts of divegence of a vector V, and if the vector has only one component '− ′ this expression coincides with its divergence. In particular, the combination e {+ • e −} • f is exactly Dalembert operator applied to a function f which is constant on the wave fronts. Consequently, the first term in the right-hand side of the equation above is exactly a divergence, hence the integral can be taken by parts and variation of the action integral becomes: where the first term reduces to a surface integral and fugure brackets at subscripts mean symmetrization. As usual, the surface integral vanishes at infinity and all the rest reduces to the following Euler-Lagrange equation: Evidently, this covariant equation reduces to Dalembert equation in local Cartesian coordinates provided that the curves x(s) are locally null straight lines. In fact, any vector whose covariant derivative along the curve is zero: and so for the complex conjugate, satisfies this equation. The constrained fields satisfy this equation due to the equations (1-6), consequently, this part of the entire variation of the action integral is zero due to the constraints. Though the field equation leaves ω an arbitrary function of the phase we restrict our analysis with monochromatic waves for which this value is constant. Papapetrou equations It remains to consider the second part of variation of the action integral, produced by variation of local frames under fixed field variables. Since the local frames are defined as co-moving frames on the congruence of null-curves, it is possible to introduce variation of the congruence and derive variation of the frames from it. Small change of shape of a curve x(s) causes small change of the tangent vector without change of its length. Consequently, variation of the vector e − is orthogonal to it and to the complementary null-vector e + , hence, belongs to the span of the two polarization vectors e β . Since the vector e + is also orthogonal to the span, it suffers no change. Therefore, variations of both e + and the corresponding covector θ − are zero. Thus, only variations of the vector e − and the corresponding 1-form θ + contribute variation of the action integral (10). The first term in the Lagrangian (10) contains the factor θ + , θ + = e − , e − ≡ 0, therefore its contribution is predetermined only by variation of the vector e − =ẋ. Consider variation of four-dimensional integral with respect to variation of the curves. Variation of the factor ẋ ,ẋ is well-known from the variation principle for geodesics [6]. Here we can use the fact that variation of the integrand reduces to the scalar product of the vector of variation of the curve δx(s) and the covariant acceleration Dẋ ds : The second term in the Lagrangian (10) has single zero factorȦ in the scalar product, consequently, non-zero contribution to the variation of the action integral appears only when varying this factor. The zero factor to be varied isȦ: Variation of the covariant derivative contains derivative on s and the term containing the connection γ a bc· . Variation of the first of them does not contrtibute variation of the action integral because neither polarization vectors e α nor components of the field change under varying the congruence of curves. The only term suffering some change is connection γ a bc· . Therefore we can write down contribution of this term as follows: where the sign minus appears due to the lower index at the field component. Thus, the next task is to find variation of the connection δ γ a bc· . It is easier to find variation of the 1-form ω b a· because the variation is to be taken in a fixed point under changing the field of frames which is dragged by the vector δx. This variation is, by definition, Lie derivative of the connection 1-form with respect to this vector. So, to find variation of the connection 1-form it suffices to take its Lie derivative with respect to the vector δx. Lie derivative of a 1-form λ with respect to a vector v is [4] £ v λ = dλ( v) + d(λ( v)). Unlike ordinary 1-form the form of connection has components referred to local frames. Since the frames are built on the vector of velocity on the congruence its variation causes some infinitesimal rotation of the frames. Denote the corresponding matrix of rotation η a c . This rotation transforms components of the connection 1-form and must be taken into account. To do it it is necessary to obtain explicit form of this matrix from the dragging vector δx. Consider a point in the space-time a curve passing through it and the local frame and small variation of congruence of curves given by small vector δx which drags the congruence. This dragging replaces the curve passed through this point with the curve dragged by the vector and the local frame built in this point is also to be replaced by the frame dragged by this vector from a neighboring point. Since, on one hand both the frames are orthonormal variation of the frames is small rotation. On the other hand, since this rotation is specified by dragging orthonormal frame from a neighboring point, this transformation is given by the connection itself, in other words the matrix of rotation η a c is exactly the value of the form of connection on the dragging vector: η a b = ω b a· (δx). Thus, Lie derivative of the connection 1-form with respect to the vector δx is Subsituting the matrix of rotation we obtain the desirted Lie derivative: where we have obtaind the curvature 2-form Ω a b ≡ R cda b θ c ∧ θ d . Substituting now this into variation ofȦ gives: Variation ofȦ is similar, so after composing the total variation of the second term in the action integral we obtain the two terms. One is total derivative of ω(A a γ a bc· δx bĀc ) on s, which vanishes on the endpoints of the curves. Thus, the whole of variation of this part of Lagrangian is given by another term which is where we introduce spin by its only component S 1 2 . The Euler-Lagrange equation for the curves x(s) coincides with Papapetrou equation: Conclusion Geometric-optical description of electromagnetic wave in curved space-time, with account of its polarization is obtained straightforwardly from the classical variational principle for electromagnetic field. For this end the entire functional space of electromagnetic fields is reduced to its subspace of locally plane monochromatic waves. Therefore, first of all, the notion of locally plane monochromatic wave in curved space-time should be defined. It turns out that waves of this sort exist provided that their wavelengths are small compared with scales under consideration. Assuming this, we have formulated the constraints under which the entire functional space of electromagnetic fields reduces to its subspace of locally plane monochromatic waves and imposed these constraints. These constraints not only reduce field variables but also introduce variables of another kind which specify a field of local frames associated to the wave and contain some congruence of null-curves x i (s). These curves become the main object in the construction because it specifies the field of local frames and the field variablesȦ andȦ are referred to this frame. Returning to the action principle for the constrained electromagnetic field we have Lagrangian (10) which contains variables of two kinds, namely, a congruence of curves x i (s) and the field itself and have two kinds of Euler-Lagrange equations. Equations of first kind reduce to local Dalembert equation for the field componentsȦ andȦ which are trivial due to the constraints imposed. Variation of the curves yields all the rest equations which contain the main result of this investigation. It turns out that the Euler-Lagrange equations they yield are exactly the Papapetrou equations for a classical massless particle with helicity 1. This equation determines the shape of the 0-curves which, by construction, can be considered as world lines of photons with the same wavelengths and helicities. They apparently differ from null-geodesics and, thereby manifest influence of spin-gravitational interaction on propagation of electromagnetic waves in gravitational fields. Effect of this interaction is proportianal to the wavelength [3], therefore this fact can be observed in radioastronomy.	2014-10-01T00:00:00.000Z	2006-01-12T00:00:00.000	{ "year": 2006, "sha1": "81b657885519d406df377afc7b46abf77b8fad17", "oa_license": "CCBYNCND", "oa_url": "https://www.degruyter.com/document/doi/10.1515/zna-2006-3-407/pdf", "oa_status": "HYBRID", "pdf_src": "Arxiv", "pdf_hash": "ff75a908ccc777992fd42b15a99bfdcf33014dba", "s2fieldsofstudy": [ "Physics", "Mathematics" ], "extfieldsofstudy": [ "Physics" ] }
27442361	pes2o/s2orc	v3-fos-license	Phosphatidylinositol-4-phosphate 5-Kinase Localized on the Plasma Membrane Is Essential for Yeast Cell Morphogenesis* Phosphatidylinositol 4,5-biphosphate (PtdIns(4,5)P 2 ), an important element in eukaryotic signal transduction, is synthesized either by phosphatidylinositol-4-phosphate 5-kinase (PtdIns(4)P 5K) from phosphatidylinositol 4-phosphate (PtdIns(4)P) or by phosphatidylinositol-5-phosphate 4-kinase (PtdIns(5)P 4K) from phosphatidylinositol 5-phosphate (PtdIns(5)P). Two Saccharomyces cerevisiae genes, MSS4 and FAB1 , are homologous to mammalian PtdIns(4)P 5Ks and PtdIns(5)P 4Ks. We show here that MSS4 is a functional homolog of mammalian PtdIns(4)P 5K but not of PtdIns(5)P 4K in vivo. We constructed a hemagglutinin epitope-tagged form of Mss4p and found that Mss4p has PtdIns(4)P 5K activity. transformants that grew in SD-Trp, 15 strains grew on FOA plates at 23 °C but not at 37 °C. Finally, one temperature-sensitive mutation ( mss4-1 ) on plasmid (pYO1970) was further analyzed. The mss4-1 mutation and the wild-type MSS4 gene were integrated into the genome by one-step plasmid integration strategy. We utilized plasmids pYO1974 and pYO1975 digested with Sac I and Avr II to transform YOC802 and selected the integrants for the LEU2 marker. After being incubated at 23 °C for 2 days, the cells were streaked on FOA plates and were incubated at 23 °C for 3 days so that the MSS4 - URA3 plasmid was eliminated. Single colonies were picked up and were tested for temperature sensitivity. A temperature-sensitive mss4-1 strain was thus obtained and designated YOC808, whereas the MSS4 gene integrated in the same locus was labeled YOC807. Immunofluorescent staining of was to early phase at 30 °C HA-tagged Mss4p was visualized by indirect immunofluorescence 16B12 anti-HA mouse monoclonal antibody as first antibody an fluo-rescein isothiocyanate-conjugated goat anti-mouse IgG (Wako ,6 -diamidino-2-phenylindole di- hydrochloride, rhodamine-phalloidin white M2R ent protein kinases and calcineurin, a type II B phosphoprotein phosphatase (4). Diacylglycerol, on the other hand, activates the conventional isoforms of protein kinase C, which in turn play a critical role in the regulation of a number of cellular functions in mammalian cells (5). In the budding yeast Saccharomyces cerevisiae, a protein kinase C-homologous gene (PKC1) was isolated (6), whose product was shown to function in cell wall integrity and cell cycle progression (7,8). In vitro studies of Pkc1p, however, indicated that Pkc1p is strongly activated by phosphatidylserine in the presence of Rho1p, but not by diacylglycerol (9). The stimulation by phosphatidylserine alone is characteristic of the atypical isoform of protein kinase C, which is stimulated by phosphatidylserine alone. Since the biochemical property of Pkc1p is different from that of the conventional isoforms of mammalian protein kinase C, it remains unclear whether and how diacylglycerol acts as an important second messenger in S. cerevisiae. PtdIns(4,5)P 2 is also known to function as a regulator of actin-binding proteins (10) such as profilin (11), gelsolin (12), and ␣-actinin of vertebrates (13). Recently, profilin was reported to be localized both in the plasma membrane and cytosolic fractions in S. cerevisiae, with the membrane association presumably facilitated by its interaction with phosphatidylinositol metabolites (14). Therefore, it is likely that through its regulation of actin-binding proteins, phosphatidylinositol metabolites affect the cytoskeleton in yeast. Moreover, PtdIns(4,5)P 2 stimulates GDP to GTP exchange of ADP-ribosylation factor 1 (ARF1) (15). As the GTP-bound form of ARF1 triggers the attachment of the coat proteins (16 -18), PtdIns(4,5)P 2 may play a critical role in coat assembly. Interestingly, PtdIns(4,5)P 2 was found to work as a cofactor for brain membrane phospholipase D (PLD) (19). These findings led to the proposal that PLD and phosphatidylinositol 4-phosphate 5-kinase (PtdIns(4)P 5-kinase) with their respective products, PtdIns(4,5)P 2 and phosphatidic acid, form a positive feedback loop that causes a vesicle fusion with the acceptor membrane (19). Since PtdIns(4,5)P 2 as well as phosphatidic acid activates an ARF GTPase-activating protein (20), they further postulated that the positive feedback loop is halted by the conversion of active ARF-GTP to ARF-GDP. Thus, PtdIns(4,5)P 2 may work as a crucial factor in membrane trafficking. We report here that Mss4p has PtdIns(4)P 5K activity in vitro and that expression of murine type I␤ PtdIns(4)P 5K functionally replaces MSS4 in vivo. Unlike Fab1p, Mss4p is located primarily on the plasma membrane. Analyses of a temperature-sensitive mss4 mutant revealed that Mss4p is involved in the establishment of cell morphology. MATERIALS AND METHODS Yeast Strains and Genetic Manipulations-The yeast strains used are listed in Table I. The complete and minimal yeast media as well as the sporulation medium and procedures of tetrad analysis were as described (30). YPGS medium contains 2% galactose, 0.1% sucrose, 1% Bacto-yeast extract, and 2% polypepton, whereas YPA medium for pre-sporulation consists of 1% Bacto-yeast extract, 2% polypepton, and 1% potassium acetate (Wako Pure Chemical Industries, Osaka, Japan). Yeast transformation was carried out with lithium acetate (31). Plates containing 0.2% 5-fluoroorotic acid (FOA, Sigma) were used to select yeast cells capable of losing a URA3 marked plasmid. E. coli strains, DH5␣ (Life Technologies, Inc.) and SCS1 (Stratagene), were used for gene manipulation. DNA sequencing was carried out with an automated DNA sequencer (model 373A, Applied Biosystems, Foster City, CA). Construction of Plasmids-The plasmids used in this study are described in Table II. Plasmid pYO1953 was cloned from the YEp13 genomic library (32). The insertion of the 3.9-kb BamHI-XhoI fragment of MSS4 into the vector pBluescript SK ϩ resulted in pYO1956, which was used for the construction of other MSS4-containing plasmids and as a template for error-prone polymerase chain reaction. pYO1958, which was designed to aid MSS4 gene disruption, mss4::HIS3, was constructed by ligation of the 5.0-kb EcoRI-EcoRI fragment of pYO1956 and the 1.3-kb BamHI-XhoI fragment of pJJ215 containing the HIS3 gene. pYO1959, pYO1960, and pYO1962 were made by inserting the 3.9-kb BamHI-XhoI fragment of pYO1956 containing MSS4 into the BamHI-XhoI gap of pRS315, pRS314, and pRS316, respectively. pYO1964 was formed by replacing the 1.2-kb NdeI-KpnI fragment of pYO1960 with the NdeI-KpnI linker, which was made by annealing the oligomers TATGTGAGATCTGGTAC and CAGATCTCACA. The murine PtdIns(4)P 5K type I␤ and human PtdIns(5)P 4K genes were obtained by polymerase chain reaction using the published sequences (25,23) with the BamHI and BclI restriction sites, respectively, attached at both ends. The PtdIns(4)P 5K and PtdIns(5)P 4K genes were then inserted to the BglII site of a single-copy plasmid with the GAL1 promoter (pYO761) to yield pYO2116 and pYO2117, respectively, and to a multicopy counterpart (pYO767) to produce pYO2118 and pYO2119, respectively. We also inserted the murine and human PIPK genes to the BglII site of a single copy plasmid with the GAP promoter (pYO2141) to get pYO2142 and pYO2143, respectively. Similar insertions of the PIPK genes to a multicopy plasmid with the GAP promoter (pYO2144) resulted in pYO2145 and pYO2146. We selected and used only those plasmids with the genes whose sequences agreed with the published ones and inserted in the right direction. A 3HA tag was introduced to the N terminus of Mss4p as follows. An AvrII adaptor was made by annealing the two oligonucleotides, CCG-GATCCTAGG and CCGGCCTAGGAT, and was inserted to the AccIII site of pYO1959. After checking the direction of insertion by DNA sequencing, we digested the plasmid with AvrII and ligated it with the NheI-digested 3HA tag of pYO1365. The SphI-XhoI fragment of the resultant plasmid carrying 3HA-tagged MSS4 was inserted to the SphI-XhoI gap of pRS314 and pQR324 to produce plasmids pYO1965 and pYO1966, respectively. Immunoprecipitation and PtdIns(4)P 5K Assay-Cell lysates were made in RIPA buffer (50 mM Tris-HCl, pH 8.0, 1% Nonidet P-40, 0.15 M NaCl, 5 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin, 2 g/ml aprotinin, and 1 mM sodium orthovanadate) by vortexing six times for 30 s each with acid-washed glass beads (425-600 m in diameter, Sigma). After preadsorption with protein A cellulofine (Seikagaku-kogyo, Tokyo), the samples (300 g of protein, assayed by Bio-Rad protein assay kit) were subjected to immunoprecipitation with saturating amounts of 16B12 anti-HA monoclonal antibody (Berkeley Antibody, Richmond, CA) and then adsorbed to protein A cellulofine. The adsorbed immunoprecipitates were then washed four times with RIPA buffer and four times further with buffer T (50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 50 M ATP, 0.25 M sucrose, and 0.15 M NaCl). To determine the PtdIns(4)P 5K activity in the immunoprecipitates, we incubated 10 l of sample in 50 mM Tris-HCl, pH 7.5, 1 mM EGTA, 10 mM MgCl 2 , 50 M ATP, 80 M PtdIns(4)P (Sigma), and 5.0 or 0.5 Ci of [␥-32 P]ATP (Amersham Pharmacia Biotech) in the presence or absence of 50 M phosphatidic acid in a total volume of 50 l. After 60 min, the reaction was terminated by the addition of 0.4 ml of chloroform/methanol/12 N HCl (100:200:1 by volume). The lipids were extracted by the method of Bligh and Dyer (33), dried, and, together with PtdIns(4,5)P 2 , which was used as standard, were spotted on Merck Silica gel 60 TLC plates impregnated with 1.2% potassium oxalate, with the exception of the experiment whose result is shown in lanes 4 and 5 of Fig. 1C, in which a similarly treated Whatman 60A plate was utilized. The samples were separated with the solvent system of chloroform/methanol/ acetone/acetic acid/water (42:30:12:12:12 by volume), and [ 32 P]Ins(4,5)P 2 was visualized by autoradiography except for the product on the Whatman plate, which was processed by BAS2000 Fuji bioImaging analyzer. Isolation of Temperature-sensitive mss4 Mutants-We first made an mss4 strain carrying the mutant gene on a centromer plasmid; the 3.1-kb BamHI-XhoI fragment of pYO1958 carrying the mss4::HIS3 gene was used to transform the diploid strain, YPH501. His ϩ transformants were selected, and the disruption of one of the chromosomal MSS4 gene copies was confirmed by Southern hybridization. The MSS4/mss4::HIS3 diploid strain, named YOC801, was transformed with pYO1962 carrying MSS4 and URA3, and the transformants were subjected to tetrad dissection. His ϩ Ura ϩ asci were selected and designated YOC802 (mss4::HIS3 (pYO1962)). Random mutations were introduced by error-prone polymerase chain reaction mutagenesis (34) in the PI(4)P 5-kinase-conserved region of MSS4 using the two synthetic oligonucleotides, CCTTCTCAAAAGT-CAAAGCA and TCGTACTACCGTTCCGGTA, corresponding to bases 841-860 and 2055-2025, respectively. The amplified 1.2-kb fragment was purified, digested with NdeI and KpnI, and then inserted to the NdeI-KpnI gap of pYO1964. Approximately 4,000 independent clones were made, and DNA of the plasmid pool was employed for transformation of YOC802 strain. The transformants that grew on SD-Trp medium at 23°C were streaked on FOA (ϪTrp) plates and grown at 23 or 37°C so that the MSS4-URA3 plasmid is eliminated. Out of 496 Involvement of PtdIns(4)P 5-Kinase in Yeast Morphogenesis transformants that grew in SD-Trp, 15 strains grew on FOA plates at 23°C but not at 37°C. Finally, one temperature-sensitive mutation (mss4-1) on plasmid (pYO1970) was further analyzed. The mss4-1 mutation and the wild-type MSS4 gene were integrated into the genome by one-step plasmid integration strategy. We utilized plasmids pYO1974 and pYO1975 digested with SacI and AvrII to transform YOC802 and selected the integrants for the LEU2 marker. After being incubated at 23°C for 2 days, the cells were streaked on FOA plates and were incubated at 23°C for 3 days so that the MSS4-URA3 plasmid was eliminated. Single colonies were picked up and were tested for temperature sensitivity. A temperature-sensitive mss4-1 strain was thus obtained and designated YOC808, whereas the MSS4 gene integrated in the same locus was labeled YOC807. Immunofluorescence Microscopy-Immunofluorescent staining of yeast cells was carried out according to Pringle et al. (35). Cells were grown to early exponential phase at 30°C in YPD medium. HA-tagged Mss4p was visualized by indirect immunofluorescence using 16B12 anti-HA mouse monoclonal antibody as the first antibody and an fluorescein isothiocyanate-conjugated goat anti-mouse IgG (Wako Pure Chemical Industries, Osaka, Japan) as the second antibody. DNA, actin, and chitin were stained with 4Ј,6Ј-diamidino-2-phenylindole dihydrochloride, rhodamine-phalloidin (Molecular Probes), and calcofluor white M2R new (Sigma), respectively. Cell morphology and fluorescent staining were observed and photographed using a BX60 microscope (Olympus, Tokyo). Cell fractionation experiments were performed using the previously described techniques (36). Mss4p Is a Functional Homolog of Mammalian PtdIns(4)P 5-Kinase-A BLAST search of protein sequence data bases revealed that yeast Mss4p has 36, 33, and 31% identity with murine type I␣, type I␤ PtdIns(4)P 5K, and human PtdIns(5)P 4K, respectively, in agreement with previous reports. To examine whether Mss4p is a functional homolog of any of the mammalian phosphatidylinositol phosphokinases (PIPKs) in yeast, we constructed plasmids carrying the genes encoding murine type I␤ PtdIns(4)P 5K and human PtdIns(5)P 4K hooked up to either the constitutive GAP promoter or the galactose-inducible GAL1 promoter. After these expression plasmids were introduced to YOC802 strain carrying mss4::HIS3 and a URA3-MSS4 plasmid, the growth on FOA plates was examined. We found that all the transformants expressing the type I␤ PtdIns(4)P 5K gene were capable of growing on FOA plates (Fig. 1A, panels b and c). On the other hand, expression of the PtdIns(5)P 4K gene failed to complement the MSS4 gene disruption, irrespective of copy number or the promoters used (Fig. 1A, panels b and c). We next made a temperature-sensitive MSS4 mutant and tested the suppression of the temperature sensitivity by mammalian PIPKs. Mutations were introduced into the conserved region for PIPK within the MSS4 gene, and one of the mutants that grew at 23°C but not at 37.5°C (mss4-1) was studied (see "Materials and Methods"). The temperature-sensitive mss4-1 strain was transformed with plasmids containing the two mammalian PIPK genes under the control of the two different promoters to test if the temperature sensitivity is suppressed. When murine type I␤ PtdIns(4)P 5K was expressed under the GAL1 promoter on either the single copy or multicopy plasmid, the strain grew on a galactose-containing plate at the restrictive temperature (Fig. 1A, panels e and f). On a glucose-containing plate on which the expression of type I␤ PtdIns(4)P 5K under the GAL1 promoter is reduced, however, suppression of mss4-1 was observed only with the multicopy plasmid, indicating the failure of suppression when expression is greatly reduced. When the same gene was placed on plasmids under the GAP promoter, the plasmid-harboring strains grew at the re- (52) strictive temperature irrespective of the copy number (Fig. 1A, panels e and f). Murine type I␤ PtdIns(4)P 5K therefore suppresses the temperature sensitivity of mss4-1. In contrast, human PtdIns(5)P 4K failed to suppress the temperature sensitivity in any of the combinations of the plasmids and the promoters tested. Therefore, the functional complementation analysis with the deletion and the temperature-sensitive mutations of mss4 suggest that MSS4 encodes PtdIns(4)P 5K. Expression of an Epitope-tagged MSS4 in Yeast Cells-To analyze the Mss4p functions, we inserted the 3HA-epitope tag at the N-terminal of Mss4p. Introduction of the 3HA-epitope tag preserves its essential function because the tagged MSS4 with a single copy plasmid can fully complement mss4::HIS3 at all the temperatures examined (23, 30, and 37°C). These results indicate that the 3HA-tagged Mss4p is functional in vivo. Western blotting analysis of the cells expressing the tagged Mss4p has shown that the anti-HA monoclonal antibody recognized a single band with a molecular mass of 86 kDa, which matched the predicted molecular weight of Mss4p (data not shown). The Tagged MSS4 Gene Product Has PtdIns(4)P 5-Kinase Activity-To examine PtdIns(4)P 5K activity of the MSS4 gene product, we immunoprecipitated the 3HA-tagged MSS4 protein expressed in yeast with the anti-HA monoclonal antibody and determined the kinase activity in the immunoprecipitate (Fig. 1B). The immunoprecipitate from the YOC804 cells carrying the tagged MSS4 gene on a multicopy plasmid had the highest PtdIns(4)P 5K activity, followed by that from the YOC803 cells, which harbored the same gene on a single copy plasmid, whereas that from the cells with untagged Mss4p (YOC806) exhibited little activity. Furthermore, the PtdIns(4)P 5K activity in the immunoprecipitates was found to be stimulated by the addition of 50 M phosphatidic acid (Fig. 1B), a characteristic property of PtdIns(4)P 5K but not of PtdIns(5)P 4K (37). These results demonstrate that the tagged Mss4p possesses PtdIns(4)P 5K activity and that the amount of the kinase activity is copy number-dependent. The mss4-1 Protein Has Less PtdIns(4)P 5K Activity When Cultured at the Restrictive Temperature-We examined whether the PtdIns(4)P 5K activity of the mss4-1 mutant changes at the restrictive temperature. We first made a strain with the MSS4 gene disrupted but harboring a 3HA-tagged mss4-1 gene on a multicopy plasmid and designated it YOC823. The strain was cultured at 23°C, was transferred to 38°C at early exponential growth phase, and was further cultivated for 0, 2, 4, 6, or 8 h before being harvested. The lysates were made, immunoprecipitated with the anti-HA antibody, and the PtdIns(4)P 5K activities in the immunoprecipitates were assayed. As can be seen in Fig. 1C, the kinase activity starts decreasing immediately upon the temperature shift, and the reduction is complete by 4 h at the restrictive temperature. Western blotting of cell lysates showed that the strain had less amount of the mutant protein when cultured at the restrictive temperature than at the permissive temperature (data not shown). These results indicate a temperature-sensitive defect in the synthesis and/or heat lability of the mutant protein. The MSS4 Gene Product Is Localized on the Plasma Membrane-To examine intracellular localization of the tagged Mss4p, we first investigated the partitioning of 3HA-tagged [␥-32 P]ATP. Lanes 4 and 5, the PtdIns(4)P 5K activity in YOC804 cells was similarly assayed in the presence or absence of 50 M phosphatidic acid (PA) with 0.5 Ci of [␥ -32 P]ATP per sample. C, Lanes 1-5, the YOC823 (⌬mss4::HIS3 (YEp3HA:mss4-1)) strain was cultured at 23°C and was then placed at 38°C for indicated times, the lysates prepared and immunoprecipitated, and the PtdIns(4)P 5K assay was carried out at 23°C with 5.0 Ci of [␥-32 ]ATP in each sample. Phenotypes of the Temperature-sensitive mss4-1 Mutant-Growth of the temperature-sensitive mss4-1 mutant (YOC808) was compared with those of the wild-type strain (YOC807) and the mss4::HIS3 strain expressing untagged Mss4p on a single copy plasmid (YOC806) cultured on YPD plates (Fig. 3) and in YPD liquid media (Fig. 4). Judging from the colony size, the mss4-1 strain grew as well as the wild-type strain at 23°C, grew slowly at 37°C, and completely failed to grow at 37.5°C (Fig. 3). The growth defect of the mss4 mutant was not suppressed by addition of 100 mM CaCl 2 . The doubling time of mss4-1 was 4.3 h at 23°C, whereas that of the wild-type cells was 4.0 h, indicating that the mss4-1 mutant grows almost as fast as the wild type at the permissive temperature. At 38°C, however, the growth of the mutant cells stopped within 4 h after the temperature shift (Fig. 4). When observed under the light microscope, the mss4-1 mutant cells were found to stop growing at 38°C with an enlarged size and round shape (Fig. 5). There was no indication of cell lysis under the microscope after a 4-h incubation at 38°C. Most of the mutant cells contain single nuclei or divided two nuclei (Fig. 6, panel I). Even at the permissive temperature (23°C), the mutant cells were round-shaped, whereas the wild-type cells were all oval-shaped (data not shown). As similar morphological phenotypes were observed with mutations of actin-binding proteins, such as profilin and capping protein (38,39), we then examined the actin distribution of the mss4-1 cells by staining with rhodamine-conjugated phalloidin (Fig. 6). In the wild-type cells, the actin staining showed cables running longitudinally and cortical patches, whereas in budding cells, it occurred exclusively on the bud as previously reported (40 -42) (Fig. 6, panel F). A 1-h incubation at 38°C of the wild-type cells made actin cables fainter and altered the localization of actin patches to the bud (Fig. 6, panel G), but further incubation restored polarized distribution of actin patches to the bud and, until 8 h at the restrictive temperature, kept the localization unchanged (Fig. 6, panel H). In contrast, phalloidin staining of mss4-1 mutant cells revealed a random distribution of cortical actin patches. In the mutant cells, actin patches became partially polarized after a 4-h incubation (Fig. 6, panel C), but the polarization was completely lost after 6 and 8 h of incubation at 38°C (Fig. 6, panels D and E). Even at 23°C, actin cables were faint, and actin patches were frequently distributed not only at buds but also in mother cells in the temperature-sensitive strain (Fig. 6, panel A). At 38°C, the staining of actin was obviously brighter in the mutant cells than in the wild-type cells after the 8-h incubation at 38°C (Fig. 6, panel E), whereas staining of cell wall chitin revealed deposition of chitin all over the cell surface in the mutant cells (Fig. 6, panel K). These results indicate that cell polarity and actin distribution are significantly impaired in mss4-1 even at 23°C, and the loss becomes greater at 38°C. We further examined a genetic interaction between mss4-1 and cls5-1, the latter of which possesses a missense mutation in the profilin gene. 2 We found that a haploid strain with mss4-1 and cls5-1 mutations harboring a URA3-MSS4 plasmid could not grow on FOA plate at 30°C (data not shown). This indicates that the mss4-1 cls5-1 double mutant fails to grow, although either an mss4-1 or a cls5-1 single mutant grows well at both 23 and 30°C. The synthetic lethal interaction between MSS4 and the profilin gene is consistent with our observation that actin cytoskeleton is impaired in the mss4-1 mutant. DISCUSSION The first piece of evidence that MSS4 encodes PtdIns(4)P 5K comes from the kinase assay; immunoprecipitates from cells expressing the epitope-tagged form of Mss4p had PtdIns(4)P 5K activity, and the tagged Mss4p is functional as demonstrated by the full complementation of the mss4 deletion by single-copy expression of 3HA-Mss4p. The substrate used, PtdIns(4)P, was purified from bovine brain, is approximately 98% pure on TLC according to the manufacturer, and may contain a small amount of PtdIns(5)P. It is formally possible that MSS4 actually encodes PtdIns(5)P 4K, and the PtdIns(5)P 4K activity produces PtdIns(4,5)P 2 from the trace amount of PtdIns(5)P contained in the substrate. However, the following lines of evidence make this alternative explanation unlikely. The kinase activity of the gene product is enhanced in the presence of phosphatidic acid, a characteristic of mammalian PtdIns(4)P 5K isoforms, but not of PtdIns(5)P 4K (37). In addition, the murine PtdIns(4)P 5K gene, but not the human PtdIns(5)P 4K gene, complemented the mss4 gene disruption and, when the expression level was reasonably high, suppressed the temperature sensitivity of the mss4-1 strain. This identification is consistent with the higher homology Mss4p has with mammalian PtdIns(4)P 5K isoforms than with Pt-dIns(5) 4K. We therefore conclude that MSS4 encodes PtdIns(4)P 5K. Our finding that MSS4 encodes PtdIns(4)P 5K explains the previous genetic studies on MSS4 well. Since overexpression of MSS4 suppresses the cell lysis phenotype of ⌬stt4 at 23°C and the temperature-sensitive stt4-1 mutation, MSS4 was suggested to function downstream of Stt4p, a PtdIns 4-kinase (29,53). The present demonstration is in agreement with the idea that Stt4p phosphorylates PtdIns to produce PtdIns(4)P, which in turn is further phosphorylated by the action of Mss4p to yield PtdIns(4,5)P 2 . In the same paper, it was reported that overproduction of MSS4 did not affect PtdIns 4-kinase activity of wild-type yeast cells and that PtdIns 4-kinase activity of the ⌬stt4 cells carrying a multicopy MSS4 plasmid was as low as that in the ⌬stt4 cells. These results are consistent with the notion that Mss4p is a PtdIns(4)P 5K and has little if any PtdIns 4-kinase activity. Another S. cerevisiae gene, FAB1, whose product has a significant homology to mammalian PIPKs, is not essential (27). Fab1p was suggested to have PtdIns(4)P 5K activity on the vacuolar membrane, and the product of the kinase reaction, PtdIns(4,5)P 2 , was proposed to function as a regulator of vacuole homeostasis (27). However, the possibility that FAB1 encodes PtdIns(5)P 4K cannot be excluded, especially because no kinase assay has been reported. On the other hand, MSS4 is an essential gene (29) whose product is mainly localized to the plasma membrane (Fig. 2). Thus, it seems that Mss4p functions as PtdIns(4)P 5K on the plasma membrane and the product of the reaction, PtdIns(4,5)P 2 , plays an essential function at or near the plasma membrane. The idea that the two PIPKs that produce PtdIns(4,5)P 2 play different roles at different compart-FIG. 4. mss4-1 strain stops growing at 38°C. At zero time, YOC807 (MSS4) and YOC808 (mss4-1) cultures grown in YPD at 23°C to early exponential phase were diluted with the same medium, and the diluted cultures were grown at either 23 or 38°C. The cell densities at time zero and subsequent times were determined with a hemocytometer. FIG. 5. mss4-1 strain shows aberrant cell morphology. YOC808 (mss4-1) cells were cultured at 23°C, diluted, and further incubated for 8 h at 23 or 38°C. The cells were observed under the phase-contrast microscope. ments is supported by our recent observation that MSS4 on a multicopy plasmid suppresses the temperature sensitivity of cmd1-228, a calmodulin mutant with defect in calmodulin localization (43), whereas FAB1 does not (data not shown). One explanation of the phenotypes of the temperature-sensitive mss4-1 mutant cells is that the phenotypes are caused by a reduced level of PtdIns(4,5)P 2 , whereas another interpretation is that they are brought about by defects in hitherto unidentified function(s) of Mss4p. We favor the former possibility because it is consistent with our finding that, when the temperature-sensitive mutant cells are shifted to the restrictive temperature, PtdIns(4)P 5K activity decreases before the mutant phenotypes become evident (Figs. 1C and 6). Through what pathways does PtdIns(4,5)P 2 give rise to the mutant phenotypes? The mss4-1 mutant phenotypes are strikingly similar to those of mutants of two actin-binding proteins, i.e. profilin null mutants (38) and the capping protein deletion mutants, ⌬cap1 and ⌬cap2 (39). Profilin is a ubiquitous actinand PtdIns(4,5)P 2 -binding protein in eukaryotic cells (44) and is required for the proper organization of actin cytoskeleton into actin cables, which occur at regions of active growth and for proper maintenance of cell polarity (38). It was also shown that depletion of PtdIns(4,5)P 2 in the plasma membrane leads to profilin translocation to the cytosol (14). The binding of capping protein to the growing end of actin filaments was demonstrated to be prevented by micromolar concentrations of PtdIns(4,5)P 2 (39). We show here a synthetic lethal interaction between PtdIns(4)P 5K and profilin. Thus, it is plausible that a lower level of PtdIns(4,5)P 2 in mss4-1 cells hinders proper functioning of profilin and capping protein, leading to disorganization of actin cables. Ins(4,5)P 2 is known to be hydrolyzed by phospholipase C, encoded by the PLC1 gene in S. cerevisiae, to produce IP 3 and diacylglycerol. Temperature-sensitive plc1 mutant cells were reported to be swollen with large buds and two nuclei at the restrictive temperature, and the growth defect of the mutant was suppressed by addition of 100 mM CaCl 2 (45). An increased incidence of aberrant chromosomal segregation was also observed with another temperature-sensitive mutant, plc1-1 (46). Since mss4-1 cells at the restrictive temperature do not show these phenotypes, we consider it improbable that the primary effect of Mss4p is through PtdIns(4,5)P 2 hydrolysis. Alternatively, PtdIns(4,5)P 2 may be required to stimulate GDP to GTP exchange of yeast ARF, which is important for secretion (47). The localization and the mutant phenotypes, however, do not support the idea that the major pathway related to Mss4p involves ARF. PtdIns(4,5)P 2 is also known to work as a cofactor for PLD. Yeast PLD encoded by SPO14 (48) is essential for meiosis but not for vegetative growth. The necessity of this gene only for meiosis again makes PLD an unlikely candidate for the major target of PtdIns(4,5)P 2 . In summary, we propose that the MSS4 gene product functions in regulation of actin-binding proteins through generation of PtdIns(4,5)P 2 from PtdIns(4)P in or near the plasma membrane. We believe that novel factors involved in the PtdIns(4,5)P 2 signaling pathway can be identified by genetic approach with the conditional mutant of mss4. Further investigations of the MSS4 gene will greatly elucidate phosphatidylinositol signal transduction cascade.	2018-04-03T00:56:55.315Z	1998-06-19T00:00:00.000	{ "year": 1998, "sha1": "b471d78f27d62137c3e4e26d84af2a94f3e8b949", "oa_license": "CCBY", "oa_url": "http://www.jbc.org/content/273/25/15779.full.pdf", "oa_status": "HYBRID", "pdf_src": "Adhoc", "pdf_hash": "2ae975a7e0e9b676cf3f4e934a7a54a8e04ff7c0", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
169302769	pes2o/s2orc	v3-fos-license	Does Diversification and Executive Compensation Affect Corporate Values in Family Firm: Case of Indonesia Free trade in the Asia-Pacific region (AFTA) and Southeast Asia (MEA) becomes a challenge for family firms in developing their business activities. Strategies that can be taken by family firms to cope with existing market pressures can be pursued by implementing a diversification and compensation strategy. This study aims to explain and analyze the influence of diversification in related models, diversification on unrelated models, and executive compensation to firm value. In this study the population taken is a family firm in the manufacturing sector listed on the Indonesia Stock Exchange during the year 2012-2016 amounted to 140. The important finding of the research is that the diversification in the related model has no significant effect on firm value, the diversification on the unrelated model has no significant effect on firm value, the executive compensation in the related diversified company has a significant negative effect on firm value, then the executive compensation on the unrelated diversified company has a significant positive effect on the value of the company Introduction The family company is a form of company whose ownership structure with the compo- The existence of family elements in the company's ownership structure creates an agency conflict between the family ownership structure and the non-family ownership structure. The family ownership structure in the company is a majority ownership structure that can put pressure on management to provide more benefits to the family as 2nd ICIEBP shareholders. The pressure from family elements on the management of the company will give a loss to minority shareholders, where the interests of minority shareholders become more unnoticed. Free trade in the Asia-Pacific (AFTA) and Southeast Asia (MEA) regions is a challenge for family companies in developing their business activities. The existence of free trade provides an opportunity for foreign companies to market their products in Indonesia, so that companies in Indonesia must have readiness in the competition. In addition, free trade in the Southeast Asia region provides opportunities for companies in Indonesia to increase capital for operational activities. This is because, free trade provides convenience for foreign investment into Indonesia, and provides convenience for companies to increase company's capital. A family company is a company that seeks to carry out operational activities based on the financial capabilities they have. As a result of these conditions, family companies in Indonesia must be able to overcome all the limitations they have in order to improve their competitiveness and to increase the value of the company. Strategies that can be taken by family companies to meet these expectations, and face the existing market pressures can be pursued by implementing a diversification strategy. Diversification strategy is a business development strategy through the expansion of business and geographical segments (Harto 2005). Diversification strategies are carried out by opening new business lines, expanding existing product lines, expanding product marketing areas, opening branch offices, conducting mergers, acquisitions and others (Harto 2005). Diversification strategy is chosen and applied by the company when the company is in certain conditions, namely when the company feels profit and company growth begins to decline in the initial industry of its business, besides diversification is also carried out in order to strengthen competitive advantage with competitors and in order to reduce investment risk because if the company only doing business in a single sector, the investment risk is quite large (Damciwar 1999). When diversifying, the company will become a multi-business company that does not only move on a single line of business, the more diverse business lines the company has, the more revenue sources that the company has. This research emphasizes the form of strategies related to products created by the company. The company carries out two forms of product diversification, namely: related and unrelated diversification. Related diversification is a company effort to diversify business into another business that still has a close relationship with the previous business so that business strategies that are mutually compatible between each business can be developed (Hariadi, 2005 (2013) prove that related diversifications will provide benefits from the efficient use of production factors to produce products that have similarities to core products. The creation of related products accompanied by efficiency in the use of production factors will be useful in increasing company profits, so that the value of the company will increase. The research conducted (Park and Jang 2013) contrasts with research conducted by (Hitt et al. 2001). (Hitt et al. 2001). found empirical evidence that related diversification does not affect the value of the company because the efficiency that occurs when the company conducts related diversification does not have a large enough influence to reduce operating expenses. This will have an impact that the company's profit performance will be more likely to remain and not experience a significant increase. Companies that carry out unrelated diversification strategies have the aim of creating financial economies. The company does this by allocating capital to other companies that are not the same business. The company also conducts asset restructuring, namely buying other company's assets and then repairing and selling the assets at a price that exceeds the costs. In relation to company value, companies that do unrelated diversification can make new products that are different from the core business and can increase market power. With the existence of unrelated products, the core company has additional profit that directly contributes to total profit (Chen and Yu 2012). Research conducted (Hyland 2013) found empirical evidence that unrelated diversifications have a positive impact on increasing firm value. (Hyland 2013) proves that unrelated diversifications will benefit from the expansion of market segmentation due to the increase in different products with the core product. Expansion of market segmentation will provide benefits for the company to increase sales volume so as to have an impact on improving the company's profit performance. The existence of a significant 2nd ICIEBP increase in profits will have a large influence on improving investor perceptions, so that the value of the company increases. The research conducted (Hyland 2013) contrasts with research conducted by (Hitt et al. 2001). (Hitt et al. 2001) found empirical evidence that unrelated diversification had no effect on firm value because increased investment and expenses incurred to conduct unrelated diversification resulted in a decrease in the performance of profits generated by the company, so that unrelated diversification strategies did not become an attraction for investors. In addition to implementing a diversification strategy, family companies also implement a policy in the form of compensation for management that aims to increase the value of the company. Compensation is income that can be in the form of money, direct goods, or indirectly received by management in return for services provided by the company (Yunita 2013). Compensation is a means of communication and performance evaluation for company management. With compensation, it is hoped that it can motivate directors to make the right decisions in increasing the value of the company. The concept of agency theory according to ( Jensen and Meckling 1976) describes a relationship that arises because of a contract between the principal and another party called the agent, where the principal delegates a work to the agent. Investors are principals in companies whose capital comes from investors' share ownership, while the management of the company is the agent. There are many mechanisms that can be pursued to reduce agency problems. Agency Problem The theory agency describes a relationship that arises because of a contract between the principal and another party called the agent, where the principal delegates a work to the agent ( Jensen and Meckling 1976). Investors are principals in companies whose capital is derived from investor share ownership, while the management of the company is the agent ( Jensen and Meckling 1976). In addition to the conflict between principal and agent, agency theory also defines the possibility of conflicts that occur between principals (Eisenhardt 1989). The conflict occurs between majority shareholders and minority shareholders. Company management will get more pressure from the majority shareholders to satisfy their interests. These conditions have an impact on the loss received by minority shareholders due to these conditions. Diversification influence on the related firm model on firm value Diversification in the related model is an enterprise's efforts to diversify business into another business that still has a close relationship with the previous business so that business strategies can be developed that are mutually compatible between each of these businesses (Hariadi 2005). Related diversification is diversified due to consumer considerations so that consumers are not saturated with one type of product. With the existence of new types of products, companies can explore economic scope (economies scope). This will make the value of the company better because it can reduce production costs and sell products at an existing competitive level, advantages that can not be done by companies with a single segment. Based on the agency theory hypothesis there is agency conflict in question is the existence of conflict when the company is having a high free cash flow then the agents will direct it to investments that will increase company sales such as diversification even though the principals want dividend distribution from the free cash flow. This is because the evaluation of management performance is associated with the level of sales of the company. Chang and Wang (2007) succeeded in proving that there was an impact of product diversification strategies internationally with company performance. It was found that related diversification has a positive effect on performance. H1a: Diversification in related models has a positive influence on firm value Diversification effect on unrelated models on firm values Diversification in an unrelated model is an enterprise's efforts to diversify business into another business that does not have a close relationship with the previous business so that a business strategy that is mutually compatible between each business can be developed (Hariadi 2005). Companies that do unrelated diversification are companies that are in a stable condition in the core business and its derivatives (related diversification). The company is expected to always increase the value of the company or improve its performance in a sustainable manner. Therefore, unrelated diversification becomes a way to improve its performance because the existence of unrelated diversification will increase market segmentation which will have an impact on increasing sales volume. The increase in sales volume from unrelated diversification activities will have an impact DOI on increasing performance so that investor perception will increase from the unrelated diversification efforts. Based on the agency theory hypothesis there is agency conflict in question is the existence of conflict when the company is having a high free cash flow then the agents will direct it to investments that will increase company sales such as diversification even though the principals want dividend distribution from the free cash flow. This is because the evaluation of management performance is associated with the level of sales of the company. Research conducted by Mackey (2006) stated that companies that do unrelated diversification benefit more from the diversification of the company. Based on the explanation, the proposed hypothesis is: H1b: Diversification in unrelated models has a positive influence on firm value. Effect of executive compensation on firm values Compensation is a reward given by the company to the company manager. Rewards are given for the work performance carried out by managers to increase company wealth. The company manager seeks to achieve personal wealth by taking advantage of the opportunity to obtain compensation promised by the company. The opportunity is carried out by managers by utilizing the ability to practice financial report manipulation to show improvement in company performance. An increase in the performance of the company in accordance with what is expected by the owner of the company will give the impact of compensation received by the manager. This shows that there is motivation for management to get high compensation by improving the company's performance position. An increase in company performance will provide benefits for the company to increase the value of the company. The company considers compensation as a cost that must be suppressed so that the compensation provided is not in accordance with the executive needs / expectations. With a compensation system that is not in line with expectations and added to the basic human nature (moral hazard), the executive will try to find additional income in an incorrect way. This situation will encourage executives to maximize their personal interests.Based on agency theory, the compensation given to managers is a policy choice that allows to increase agency conflict. This is based on information held by managers greater than shareholders, so managers try to use the information for personal gain. Efforts to improve the performance of companies that will have an impact on increasing the value of the company is a way that is run by Company value (Dependent variables) The value of the company is the investor's perception of the whole of every equity owned by the company which is associated with the stock price. The price of the stock used generally refers to the closing price, and is the price that occurs when the stock is traded on the market (Lindenberg and Ross 1981). The value of the company in this study was measured using Tobin's Q modification. Tobins'Q can be calculated using the following formula: Diversification (Independent variables) Diversification strategies in companies are to diversify businesses related to core businesses (related) and businesses that are not associated with core (unrelated) businesses in order to obtain high profits from business segments owned (Hariadi 2005). Measurement of related diversification refers to measurements taken (Choe et al. 2014). The method of measurement is to use the following formula: Executive compensation (Independent variables) One of the independent variables used in this study is executive compensation. Compensation is income that can be in the form of money, direct goods, or indirectly received by management in return for services provided by the company (Yunita 2013 Company size The size of a company is the size of a company that can be seen through the company's ability to generate company revenue through the resources owned. The size of the company is one of the scales to classify companies. Company size is measured using the natural logarithm of the book value of the total assets of the company that refers to the study (Hanlon 2005). Company age The age of the company can be interpreted to indicate the length of time the company runs a business since the company was founded. According to (George and Kabir 2012), the age of the company shows the maturity of each company in running a business. Age (I, t) = the age of the company starts operating (I, t) Leverage Leverage is a ratio that shows the level of debt use to fund a company's assets. Leverage is proxied as the ratio between total debt to total assets of the company. The higher leverage will make managers careful in investing and will improve company performance (Park and Jang 2013). Leverage can be described in the following models: Profitability Profitability is defined as the company's ability to generate profits from the sale of goods or services it produces (Astuti, 2004). This ratio is preferred by shareholders and company management as one of the tools for making investment decisions, whether this business investment will be developed, maintained and so on (Komalasari and Anna 2013). Profitability can be described in the following models: Profitability = (operating income)/total sales Company growth The company's growth is one of the goals expected by internal and external parties of a company because it has an impact both for the company and investors, creditors and shareholders. The company's growth is the impact of the flow of corporate funds from operational changes caused by growth or decrease in business volume (Helfert 1996). In this study, the company's growth is calculated by the following formula (Akben Population and Sample The population used is family companies in the manufacturing sector listed on the Indonesia Stock Exchange (IDX) for the period 2012-2016 which has been adjusted to the characteristics determined by the researcher and those needed in the research to answer the problem formulation and test the research hypothesis. The number of population in this study were 18 companies for each year, so that in this study conducted for five years, namely for the year 2012-3016, the total population was 90 companies. In this study sampling was conducted using purposive sampling technique. According to (Anshori and Iswati 2009: 105) purposive sampling is a technique for determining samples by considering or based on certain criteria. so that the total research sample is equal to the total population of the study that is as many as 90 companies. Types and data sources The type of data used in this study is quantitative data. Quantitative data can be processed and analyzed using mathematical or statistical calculation techniques. Data DOI Model and analysis techniques The analysis technique used in this study is multiple linear regression analysis. Multiple linear regression analysis in a study is used to measure the influence of more than one independent variable on one dependent variable. This research was conducted using two regression models, namely model 1 which was conducted to determine the effect of diversification on related models and executive compensation on firm value and model 2 which was conducted to determine the effect of diversification on unrelated models and executive compensation on firm value. This is because there are two research hypotheses that must be tested where each hypothesis examines the effect of one independent variable on one dependent variable. Data analysis techniques were processed using Eviews 8 software. The analysis technique used in this study was descriptive statistical testing, classical assumption test that is normality test, and hypothesis test which consisted of coefficient of determination Descriptive statistics The total sample in this study for model 1 is 65 and for the model 2 is 25 family companies in the manufacturing sector. Descriptive statistics are statistics used to analyze data by describing or describing data that has been collected as it is without intending to make conclusions that apply to the public or generalization (Anshori and Iswati 2009: 116). Regression result (model 1) Based on the results in Diversification effect on models related to firm value Hypothesis 1a stated that diversification in related models does not have a significant effect on firm value. The results showed that family companies that were able to develop similar products with core products were not able to give a strong influence on the decisions of capital market investors to invest their funds. Capital market investors have a perception that companies that diversify on related models will not provide a significant increase in the sales volume of their new products. This is because new products that are created will not be different from existing products, so fulfillment of consumer preferences for unique new products is not fulfilled. This will have an impact on the absence of more profits for capital market investors from the efforts of family companies to conduct related diversification. In addition, the addition of subsidiaries and the addition of new assets for family companies that diversify in related models do not influence the market perception of investors. capital. Capital market investors have the view that family companies that make additional subsidiaries and new assets do not provide more profits for the company so that they do not have an impact on capital market investors. Based on agency theory, family companies are companies that experience negative perceptions from stakeholders on the operationalization of the company. This perception is formed due to the efforts of family companies to give excessive benefits to family members who have share ownership in the company. To minimize these negative perceptions, family companies will publish efforts to diversify their products as information to stakeholders so that information bias can be minimized. However, information on diversification in similar products carried out by family companies lacks good response from capital market investors due to a perception of the company's inability to develop competitive advantage. The results of this study are in line with research conducted by (Hitt et al. 2001) which found empirical evidence that related diversification does not affect the value of the firm. Effects of diversification on unrelated models on firm values Hypothesis 1b stated that diversification in unrelated models has no significant effect of family companies to create new subsidiaries that are different from the parent and the effort to invest in new assets will require substantial funding. The funding will be obtained by the company through retained earnings held by the family company and the profits generated by the company. These conditions lead to a loss of opportunity for capital market investors to gain more profits from the profits generated by companies owned by family elements. This will encourage capital market investors to withdraw their investment funds to family companies that invest heavily in the formation of new subsidiaries and create new assets. Based on agency theory, family companies are companies that experience negative perceptions from stakeholders on the operationalization of the company. This perception is formed due to the efforts of family companies to give excessive benefits to family members who have share ownership in the company. To minimize these negative perceptions, family companies will publish efforts to diversify their products as information to stakeholders so that information bias can be minimized. The results of research conducted by (Tantra and Ida Ayu 2017) stated that companies that implement unrelated diversification strategies have no influence on the value of the firm. Effect of executive compensation on firm values Hypothesis 2 stated that compensation affects the value of the company. The results showed that executive compensation in diversified companies that have related diversification had a negative effect and proved to be significant to the value of the company, while executive compensation in diversified companies that carried out unrelated diversification had a positive effect and proved to be significant to the value of the company. The results showed that the compensation given to managers in family companies that do related diversification is an effort of the shareholders of family elements as an effort to perpetuate the desire for family ownership in the company to get the maximum profit from the operation of the company. Family elements in company ownership will encourage managers to carry out certain opportunistic behaviors so that shareholder profits can increase. As a result of these efforts, the shareholders of the family element will give the manager benefits in the form of compensation or incentives. The manager's efforts to meet the expectations of shareholders from family elements by carrying out opportunistic actions will lead to negative perceptions of capital market investors. This is due to fraudulent efforts made by companies that will have an impact on the sustainability of the company in the future, so that it can cause losses to capital market investors. The results of the study prove that compensation given to managers of family companies that conduct unrelated diversification is an effort from shareholders to give appreciation for the manager's efforts to improve the company's performance through product development efforts. The existence of these efforts will provide a positive perception of the amount of effort that will be carried out by managers in an effort to improve their performance so that an increase in company performance can occur annually. The increase in performance will encourage capital market investors to make efforts to increase investment in family companies that do unrelated diversification and are able to provide substantial compensation to managers. Based on agency theory, compensation is an attempt to motivate managers to improve their performance properly, and avoid opportunistic behavior. The existence of information related to compensation provided by companies in the financial statements is an attempt to minimize the information gap between the principal and agent, and the majority shareholders with minority shareholders. The compensation information presented by family companies that carry out related diversification will have an impact on the high negative perception of investors as minority shareholders, so that the value of the company will decrease, while the compensation information presented by family companies that run unrelated diversification will impact the high positive perception of investors as minority shareholders. Research conducted by (Duffhues and Kabir 2008) found a negative relationship between executive compensation and company value. But the results of this study contradict the research conducted by (Indreswari 2013) saying that compensation has a positive effect and is proven to be significant to the value of the firm. Conclusion Based on the results of the discussion of the research described previously, it can be concluded as follows: 1 Suggestion Suggestions that need to be considered due to limitations in the study are the scope of research only on family companies in the manufacturing sector, so that the results of this study cannot be generalized to other sectors. Therefore, in subsequent studies can use family companies from the non-manufacturing sector to provide a more comprehensive picture related to the influence of related and unrelated diversification and executive compensation on company value.	2019-05-30T23:47:21.931Z	2019-03-31T00:00:00.000	{ "year": 2019, "sha1": "74b01dbf25a47af537f741f17028dd1b3f7195a8", "oa_license": null, "oa_url": "https://knepublishing.com/index.php/KnE-Social/article/download/4284/8797", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "8258f04fc2c07a668601bee01e0ff9efdcb18a12", "s2fieldsofstudy": [ "Business", "Economics" ], "extfieldsofstudy": [ "Business" ] }
259008142	pes2o/s2orc	v3-fos-license	The high mutation rate at the D614G hotspot-furin cleavage site region increases the priming efficiency of the Spike protein by furin protease: analysis of Indonesian SARS-CoV-2 G614 variants obtained during the early COVID-19 pandemic D614G mutation plays a significant role in the transmissibility of SARS-CoV-2. Identification of other mutations related to D614G mutation within the Spike protein is pivotal as they might contribute to the pathogenicity of SARS-CoV-2. This study aims to analyze the mutation rate of furin cleavage site (FCS) region of Indonesian origin SARS-CoV-2 and to predict the effect of mutation against Spike priming efficiency by furin. A total of 375 sequences of Indonesian isolates obtained during the early pandemic were used for mutation analysis. Mutation analysis includes mutation pattern, variability, frequency of mutation, amino acid conservation, and mutation rate. The effect of mutation against Spike priming efficiency by furin protease from eight sequences with mutation in the FCS region was analyzed by protein–protein docking. We showed that mutations related to the G614 variant were increasing through time, in contrast to the D614 variant. The FCS region at the position 675–692 contained the most variable (66.67%) as well as the highest mutation frequency (85.92%) and has been observed to be the hotspot mutations linked to the D614G mutation. The D614G hotspot-FCS region (residue 600–700) had the highest amino acid change per site (20.8%) as well as the highest mutation rate as 1.34 × 10–2 substitution per site per year (95% CI 1.79 × 10–3–2.74 × 10–2), compared with other Spike protein regions. Mutations in the FCS region were the most common mutation found after the D614G mutation. These mutations were predicted to increase the Spike priming efficiency by furin. Thus, this study elucidates the importance of D614G mutation to other mutations located in the FCS region and their significance to Spike priming efficiency by furin. Supplementary Information The online version contains supplementary material available at 10.1007/s13337-023-00827-w. D614G mutation is located in the Spike protein at position 614, where aspartic acid from the original viral isolates (D614 variant) has been replaced by glycine in the new variant (G614 variant) [10]. This mutation has been reported to increase the infectivity of SARS-CoV-2, but it does not seem to affect severity [26]. Various geographical locations previously dominated by the D614 variant had quickly shifted into the G614 variant [29]. Despite the increasing infectivity that has been reported in both in vitro and in vivo studies, the exact mechanisms related to it remain unclear [19,13]. Earlier reports suggested that the D614G mutation promoted the more open conformation of the receptor-binding domain (RBD) than that had been observed in the D614 variant [35]. The later was supported by the increased susceptibility of the G614 variant to antibody neutralization reported in other studies [23,28]. Another study suggested that the G614 variant had enhanced binding affinity to the host receptor ACE2 [39]. However, recent reports supported different perspectives suggesting that the increased infectivity of D614G mutation might be because of the significant reduction of the S1/S2 shedding and the higher Spike incorporation into virion in the G614 variant [7,37]. S1/S2 shedding is pivotal for priming the Spike protein before cell entries [20][21][22][23][24][25][26][27][28][29][30][31][32]. The importance of furin for proteolytic cleavage at the S1/S2 site is supported by the presence of the furin motif (RXXR) at the uniquely 4-amino acid-insertion region of SARS-CoV-2 (PRRA-R) [22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38]. This inserted furin cleavage site (FCS) has played a role in SARS-CoV-2 pathogenesis both in vitro and in vivo [16]. FCS was also related to stabilizing the open conformation of RBD [30,31]. Accordingly, continuous surveillance of FCS-related mutations was necessary [3]. SARS-CoV-2 mutations are continuously detected as the transmission of SARS-CoV-2 keeps surging [15]. The genomic mutation rate of SARS-CoV-2 has been described in several reports as 8 × 10 -4 or as high as 1.37 × 10 -3 substitution per site per year [11]. However, the mutation in each specific gene or even within a particular region of a gene might show a different rate as described in the influenza virus [24][25][26][27][28][29][30][31]. This different mutation rate will generate a specific site to be the hotspot of the mutation [18,20]. Thus, this study described different mutation rates as well as a different amino acid variation per site within the Spike glycoprotein of Indonesian origin SARS-CoV-2. Furthermore, this study elucidates the importance of D614G mutation to other mutations located in the FCS region and their significance to Spike priming efficiency by furin. To our knowledge, this is the first report that illustrates the mutation rate of regions within SARS-CoV-2 Spike and the relationship of D614G mutation with FCS region mutation. Sample isolates and sequence data RNA samples of SARS-CoV-2 originating from Yogyakarta and Central Java Province, Indonesia, were received from the Laboratory of COVID-19, Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada, Indonesia. Subsequent experiments in this study used 52 RNA samples isolated from nasopharyngeal swabs collected during the early pandemic from May 2020 to October 2020 (GenBank accession number: ON413789-ON413803 and ON413806-ON413842). The research protocol has been approved by the Medical and Health Research Ethics Committee (MHREC), Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada, Indonesia (EC: KE/ FK/0511/EC/2020). In addition to our laboratory sample isolates, we collected 327 Indonesian origins of SARS-CoV-2 complete genome sequences from the GISAID database (https:// www. gisaid. org/) submitted until February 07, 2021, to represent a dynamic evolution of SARS-CoV-2 during the first one year of the pandemic in Indonesia. These sequences were derived from 27 provinces in Indonesia. The datasets were subsequently curated by removing sequences containing ambiguous bases (R, Y, S, W, K, M, B, D, H, V, and N) using Python (v3.8.3). A total of 323 curated sequences has been generated for further analysis. In addition, four sequences containing incomplete date descriptions were excluded for transmission dynamics analysis, resulting in only 319 curated sequences. For mutation analysis, redundant sequences were removed, and selection was based on the earliest identification date. The date-sorting procedure was performed using Python (v3.8.3), whereas redundant sequence removal was performed using CD-Hit (http:// weizh ong-lab. ucsd. edu/). cDNA synthesis, PCR, and sequencing Fifty-two selected RNA samples were used as the first-strand cDNA synthesis template using Reverse Transcription Kit II (ExcelRT™ series, Smobio). The procedure was conducted according to the manufacturer's instructions. RNA samples (9 µl) were incubated with 50 μM oligo (dT) and 100 μM random hexamers (1 µl) for 5 min and 1 min at 70 °C and on ice, respectively. RT buffer (4 µl), RTase (1 µl), and DEPCtreated water (5 µl) were added and incubated for 10 and 60 min at 20 °C and 50 °C, respectively. A total of 20 µl reaction was terminated by incubating at 85 °C for 5 min. Sequencing was conducted with the Sanger method using BigDye® Terminator v3.1 cycle sequencing kit chemistry (Applied Biosystems). PCR product with a total of 60 µl was sent to DNA Sequencing Services (Genetika Science, Indonesia; 1 st Base, Singapore) for sequencing. Mutation analysis The pattern of mutation related to D614 and G614 variants was analyzed cumulatively using 371 sequences, consisting of 319 GISAID sequences originating from Indonesia and 52 laboratory isolate-sequences derived from Central Java and Yogyakarta Provinces, Indonesia (this study). The amino acid conservation of the eight different regions of Spike protein was analyzed by measuring the number of amino acid changes per site. The higher number of amino acid changes per site corresponds to the lower amino acid conservation of the region. The number of amino acid changes per site was calculated by dividing the total amino acid variation by the number of sites in each region. The hotspot mutation at the D614G hotspot-FCS region was determined by observing the variability and the frequency of mutation at the 5 different regions near the active site of the FCS region. The first to the fifth regions were located at the residue 682-685, 679-688, 677-690, 675-692, and 673-694, respectively. Mutation variability of the FCS region was analyzed by dividing the number of different mutations found in the FCS region (675-692) by a number of different mutations located in the D614G hotspot-FCS region (600-700). On the other hand, the mutation frequency of the FCS region was calculated by dividing the frequency of mutation in the FCS region by mutation frequency at D614G hotspot-FCS region. D614G mutation was excluded in both mutation variability and frequency analysis. The value described in the graphs was represented in percentages. Clock rate analysis by Bayesian Markov Chain Monte Carlo (MCMC) Mutation rate analyses were conducted for five different regions of Spike, comprising of D614G hotspot-FCS region (600-700), NTD domain, S1 subunit (14-685), S2 subunit (686-1273), and Spike (Full region, 1-1274). Sequences for each dataset (n = 371 for D614G Hotspot-FCS region; n = 319 for other regions) were extracted based on amino acid position in the designated region, and the redundant sequences were removed. A series of independent curated sequence datasets with a total of 22, 29, 53, 23, and 117 was generated for the D614G hotspot-FCS region, NTD domain, S1 subunit, S2 subunit, and Spike, respectively. Curated sequence datasets for each region were included in taxa for an independent clock rate analysis using BEAST (v1.10.4). Bayesian Markov Chain Monte Carlo (MCMC) analysis was performed with the chain length as 50 million and sampling every 10 thousand by considering several parameters: substitution model of HKY + Γ with 2 partitions [(position 1 + 2), 3]; strict clock; and constant coalescent model. Other parameters including priors were left as default. All parameters were arranged and generated using Beauti (v1. 10.4). The results of log file with 10% burn-in were analyzed using Tracer (v1.7.1). ESS value was ensured to be greater than 200 for each parameter. Homology modeling and molecular docking analysis Partial Spike protein model containing the D614G hotspot-FCS region was modeled using the Swiss-Model web-server (https:// swiss model. expasy. org/). Spike glycoprotein chain B of SARS-CoV-2 with PDB ID 7KDI.B (Cryo-EM Structure = 3.26 Å) was used as the template for homology modeling of the D614G Hotspot-FCS region. We modeled 8 sequences containing different mutations at the FCS region, comprising 5 laboratory sample-isolates and 3 GISAID sequences. To ensure only the high-quality model was built, QMEAN score was greater than -4.0. Furin with PDB ID 6HZA (X-Ray Structure = 1.90 Å) was used as the receptor for Protein-Protein Docking analysis in HADDOCK 2.4 web-server (https:// wenmr. scien ce. uu. nl/ haddo ck2.4/). The native ligand was removed using Chimera (v1.14), and the protein conformation was fixed by removing the alternate locations using Python (v3.8.3) in PyMol (v.2.4.1). The protein-protein docking parameters were set as default. The best cluster of docking results with the lowest Z-score was subjected to the HADDOCK refinement interface for docking interface refinement. The best cluster from docking refinement identified with the lowest Z-score was subsequently used for further analysis. Prediction of the binding energy (ΔG) and binding affinity (Kd) of each refined-docked model was conducted using PRODIGY. The temperature parameter was set to 37 °C. The following equations calculated the fold change of ΔG and Kd: where ΔG 0 and Kd 0 represent binding energy and binding affinity, respectively, of non-mutant isolates (3.09), whereas ΔG1 and Kd1 represent binding energy and affinity, respectively, of mutant isolates/GISAID data. Statistical analysis Data arrangement was performed using Microsoft Excel 2016. All statistical analyses were performed using GraphPad Prism (v8.0.1). The mean difference between the frequency and the variability of mutation in the FCS region of the D614 and G614 variants was independently compared using a t-test. Multivariate analysis with One Way ANOVA was conducted to determine the mean difference between the group in clock rate analysis. A statistically different group was determined by p-value < 0.05. Mutation pattern in the D614G hotspot-FCS region related to D614 and G614 variants Mutations found in the Spike protein spanning from amino acid positions 600 to 700, called as D614G hotspot-FCS region were identified in this study. We define this region as the "hotspot" region since based on our current study, this region which spans from the site of D614G mutation, and the FCS region has the highest mutation frequency compared to the other regions. We analyzed 371 sequences for mutation pattern analysis, comprised of 319 GISAID sequences originating from Indonesia and 52 laboratory isolate-sequences obtained from this study. The pattern of mutation in the D614G hotspot-FCS region was different between the D614 and G614 variants. Whereas the frequency of mutation in the D614 variant did not appear to increase, the frequency of mutations in this region (D614G mutation was excluded) showed a significant increase in the G614 variant through time (Fig. 1). Thus, we found a different mutation frequency between D614 and G614 variants in the hotspot-FCS region. Mutations in the FCS region were linked to the D614G mutation Mutations in the FCS region have been shown to be linked to D614G mutation (Fig. 2). Analysis of mutation at five different FCS regions ( Fig. 2a; Supplementary Table 1) suggests that both mutation frequency and variability in this region were significantly higher in the G614 variant compared with the D614 variant (Fig. 2b). Analysis of the number of mutations that occurred at the region near the FCS active site (RRAR, position = 682-685) suggests that the FCS region at position 675-692 has been observed to be the hotspot mutations linked to the D614G mutation, which contain the most variable (66.67%) as well as the highest frequency (85.92%) of mutation spanning in the region of 600 to 700 of the Spike protein (Fig. 2c). Figure 2 shows that both mutation frequency and variability in the FCS region is associated with the D614G mutation. D614G hotspot-FCS region preserved the lowest amino acid conservation and the highest mutation rate Each Spike protein domain showed different properties related to amino acid change and mutation rate, resulting (Fig. 3a). On the other hand, the S1 subunit of the Spike protein contains the most variable amino acids residue, suggesting that this subunit contains a Spike domain with low conservation sites. Our analysis showed that 2 Spike domains located in the S1 subunit had a contrary amino acid conservation value. Whereas the RBD domain (319-541) showed the lower amino acid change per site as only 2.7%, the NTD domain (14-305) appeared to be one of the least conserved domains, showing moderately high amino acid change per site as 10.3%. However, amongst all domains found in the Spike protein, the D614G hotspot-FCS region had the highest amino acid change per site as 20.8%. Further analysis was conducted to investigate whether the D614G hotspot-FCS region had a different mutation rate compared with the other regions located within the Spike protein, such as the NTD domain (14-305), S1 subunit (14-685), S2 subunit (686-1273), and the full Spike gene (1-1274) (Fig. 3b). By calculating the molecular clock using Bayesian Markov Chain Monte Carlo (MCMC) approach, this study showed that the D614G hotspot-FCS region had the highest mutation rate as 1.34 × 10 -2 substitution per site per year (95% Confidence Interval (CI) 1.79 × 10 -3 -2.74 × 10 -2 ), whereas the NTD domain, S1 Subunit, S2 Subunit, and full Spike gene possessed lower mutation rates as 5.24 × 10 -3 (95% CI 9.04 × 10 -4 -1.06 × 10 -2 ), 2.40 × 10 -3 (95% CI 1.31 × 10 -3 -3.55 × 10 -3 ), 2.75 × 10 -3 (95% CI 5.35 × 10 -4 -5.45 × 10 -3 ), 1.61 × 10 -3 (95% CI 8.09 × 10 -4 -2.64 × 10 -3 ) substitution per site per year, respectively. This finding emphasizes that different regions in the Spike protein have different mutation rates. The four highest amino acid changes related to the D614G mutation were found in the FCS region A total of 21 mutations have been identified in the D614G hotspot-FCS region. Each mutation had different proportions, ranging from 0.27 to 81.94% (Supplementary Table 2). Q675H, N679K, S689R, Q677H, and D614G were the top 5 highest mutation frequency at this region, with relative mutation frequency of 1.08, 1.35, 2.96, 6.20, and 81.94%, respectively (Fig. 4). Aside from the D614G mutation, the Effect of mutations at D614G hotspot-FCS region against furin protease We evaluated the impact of mutations located at the D614G hotspot-FCS region on both the binding energy (ΔG) and the binding affinity (Kd) of the Spike protein against furin protease (Table 1). We built a partial model of Spike comprising the D614G hotspot-FCS region using the Swiss-Model web-server (https:// swiss model. expasy. org/) and conducted protein-protein docking using the HADDOCK web-server (https:// wenmr. scien ce. uu. nl/ haddo ck2.4/). An amino acid change of aspartic acid to glycine at position 614 of the Spike protein (D614G) was found to reduce the binding energy and affinity to furin protease by 0.1-fold and 3.73-fold times, respectively (Fig. 5). However, additional mutations linked to the D614G mutation located in the FCS region resulted in a significantly higher binding energy as well as the binding affinity of the G614 variant compared with the D614 variant, except for S680P. The strongest binding energy and binding affinity were found in the G614 variant with P681H mutations (ΔG = −11.7, Kd = 5.5), followed by the G614 variant with Q677H mutation (ΔG = −11, Kd = 16). The data suggest that the specific mutations may result in more efficient proteolytic cleavages by furin protease. Discussion Our study described that mutations found in the FCS region of Indonesian SARS-CoV-2 sequences related to the D614G mutation. This study elucidates the importance of D614G mutation in association with other mutations located in the FCS region as well as their significance to influence the Spike priming efficiency by furin. To our best knowledge, this is the first report that illustrates the mutation rate of regions within SARS-CoV-2 Spike and the relationship of D614G mutation with FCS region mutation. The higher number of mutations found in the G614 variant than in the original D614 variant suggests that mutations located in the FCS region were linked to the D614G mutation. We validated this hypothesis by comparing the variability and mutation frequency in the FCS region between D614 and G614 variants. We found that both variability and mutation frequency in the FCS region were significantly higher in the G614 variant than those found in the D614 variant. Furthermore, we found that the FCS region at positions 675 to 692 was the hotspot mutation linked to the D614G mutation. This region contained the most variable region and the highest mutation frequency. However, despite the fact that this region was the hotspot mutation, we found no mutation located at the active site of the FCS region (RRAR, position = 682-685). This finding indicates that this active site might play an essential role in either the transmission or the pathogenicity of SARS-CoV-2 infection since this site is recognized and directly interact with furin protease [6,31]. Our study also illustrated that each domain found in the Spike protein possessed different amino acid conservations and mutation rates. We focused on the region of D614G hotspot-FCS region, spanning the amino acid position 600 to 700. This region contains the D614G mutation and other mutations found in the FCS region. We found that this region was the least conserved region containing the highest amino acid variation per site compared with the other domains of the Spike protein. Furthermore, this region had the highest mutation rate 1.34 × 10 -2 substitutions per site per year (95% CI 1.79 × 10 -3 -2.74 × 10 -2 ) compared with other regions. However, one should be considered that the mutation rate found in the NTD domain, S1 subunit, S2 subunit, as well, as Spike (full gene) described in this study were consistently higher than the mutation rate of the SARS-CoV-2 complete genome (8 × 10 -4 / 1.37 × 10 -3 subs per site per year) [11,15]. This finding suggested that SARS-CoV-2 mutations tend to accumulate in the Spike protein, which might affect the viral transmissibility. Moreover, the highest mutation frequency and variability were found in the D614G hotpot-FCS region within the Spike protein. This was supported by the finding of this study demonstrating that the D614G mutation and 4 other mutations (Q675H, Q677H, N679K, and S689R) found in the FCS region had the highest proportion of mutation in this D614G hotspot-FCS region. It has been found that D614G mutation affects the fitness and transmissibility of SARS-CoV-2 by stabilizing the RBD "up" conformation as well as the less proteolytic cleavage of S1/S2 by furin [7,35,37]. Furthermore, FCS has been reported to play a pivotal role in the pathogenicity of SARS-CoV-2 [16]. Thus, we hypothesize that SARS-CoV-2 mutated in a certain manner that might affect the fitness and pathogenicity of the virus. To validate this hypothesis, we evaluated the effect of several mutations, comprised of 4 mutations identified by our laboratory isolates and 3 from GISAID sequences, against the efficiency of furin through homology modeling and protein-protein docking, as shown in Table 1. We observed that the efficiency of furin, by means of the binding energy (ΔG) and binding affinity (Kd), was reduced in the D614G mutation compared with the wild type (D614 variant). This finding might be the underlying factor that the S1/S2 of the G614 variant was more resistant to proteolytic cleavage described in two previous studies that used pseudovirus [7,37]. However, other studies using isogenic SARS-CoV-2 found no significant difference between proteolytic cleavage of S1/S2 between the D614 and G614 variants [13,23]. In addition to the D614G mutation, several mutations could effectively increase both the binding affinity and energy against furin protease, as could be seen in Q675H, Q677H, N679K, P681H, and S689R, but not S680P. This effect might play a role in the transmission as well as the pathogenicity of the virus, as this study found that these 4 mutations (Q675H, Q677H, N679K, S689R) have the highest proportion after the D614G mutation. In addition, we observed that P681H was the earliest mutation located in the FCS region from Indonesian sequences, firstly identified in December 2020 (GISAID data). This P681H mutation was also found in B.1.1.7 variant, which has been spread widely, mainly in Europe and North America, and is considered to be more transmissible [4,8,9,22,34]. This might suggest that the mutations that support more efficient proteolytic cleavages by furin protease could increase the virulence of SARS-CoV-2. Biocomputational analysis of this study is limited to the raw estimation of the differences of binding energy and binding affinity affected by different mutations using the protein-protein docking approach. This rapid approach enables us to predict the effect of a particular mutation to the protein interaction in a short period of time without using high computer resources. However, without performing validation of homology models and determining the mode of interaction, these predictions are only limited to provide a quick glance of any mutations that might affect both of the protein interaction and conformation, where inaccuracy of the prediction has also been expected. Therefore, further validation using Molecular Dynamics Simulation is required to obtain more convincing results. Nevertheless, several recent computational and in vitro studies support our hypothesis showing that Q675H, Q677H, and N679K contribute to higher virulence of SARS-CoV-2 by increasing viral infectivity and syncytium formation, priming the Spike protein through proteolytic digestion, as well as enhancing resistance to neutralization . Another limitation of our study is that it is a laboratorybased study, we did not gather the information of the patient characteristic from the referring hospital. Indeed, inclusion of the patients' clinical characteristics in addition to the sequence data would have been more informative to unravel the association between the effect of various mutations on viral infectivity and clinical severity. Therefore, further comprehensive studies are needed to investigate the associations between different mutations in the FCS region of the Spike protein with virus infectivity and the host immune responses. Conclusion In summary, our study suggests that the FCS region was the hotspot mutation within the Spike protein linked to the D614G mutation. Mutations found in the FCS region generally increase the binding energy (ΔG) and the binding affinity (Kd) of the Spike protein against furin protease, which may affect the transmissibility and pathogenicity of SARS-CoV-2 infection. However, our study was limited to SARS-CoV-2 sequences originating from Indonesia during the early pandemic, which certainly did not cover all possible factors that might affect the different mutation patterns found in other regions or worldwide. In addition, further validation using Molecular Dynamics Simulation is also required to confirm our findings. Besides, this is the first study that described the high mutation rate and the least amino acid conservation found in the FCS region as the hotspot mutation and its relationship with D614G mutation, which might affect the transmissibility and pathogenicity of SARS-CoV-2 infection.	2023-06-02T15:08:07.274Z	2023-05-31T00:00:00.000	{ "year": 2023, "sha1": "136968cce8109d916dced0448626c2ca1d04b62e", "oa_license": null, "oa_url": "https://link.springer.com/content/pdf/10.1007/s13337-023-00827-w.pdf", "oa_status": "BRONZE", "pdf_src": "PubMedCentral", "pdf_hash": "a1175c6924e646cc4157c2678bf25ce86cba89ca", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
221912234	pes2o/s2orc	v3-fos-license	HIF‐1α ameliorates tubular injury in diabetic nephropathy via HO‐1–mediated control of mitochondrial dynamics Abstract Objectives In diabetic nephropathy (DN), hypoxia‐inducible factor‐1α (HIF‐1α) activation in tubular cells plays an important protective role against kidney injury. The effects may occur via the target genes of HIF‐1α, such as haem oxygenase‐1 (HO‐1), but the exact mechanisms are incompletely understood. Materials and methods Mice with proximal tubule‐specific knockout of HIF‐1α (PT‐HIF‐1α−/− mice) were generated, and diabetes was induced in these mice by streptozotocin (STZ) injection. In addition, to mimic a hypoxic state, cobaltous chloride (CoCl2) was applied to HK‐2 cells. Results Our study first verified that conditional knockout of HIF‐1α worsened tubular injury in DN; additionally, aggravated kidney dysfunction, renal histopathological alterations, mitochondrial fragmentation, ROS accumulation and apoptosis were observed in diabetic PT‐HIF‐1α−/− mice. In vitro study showed that compared to control group, HK‐2 cells cultured under hypoxic ambiance displayed increased mitochondrial fragmentation, ROS production, mitochondrial membrane potential loss and apoptosis. These increases were reversed by overexpression of HIF‐1α or treatment with a HO‐1 agonist. Importantly, cotreatment with a HIF‐1α inhibitor and a HO‐1 agonist rescued the HK‐2 cells from the negative impacts of the HIF‐1α inhibitor. Conclusions These data revealed that HIF‐1α exerted a protective effect against tubular injury in DN, which could be mediated via modulation of mitochondrial dynamics through HO‐1 upregulation. It is known that hypoxia-inducible factor-1 (HIF-1) is a critical molecule for mitigating hypoxia-induced damage and exists as a heterodimer comprising two subunits: a variable α-subunit and a constitutively expressed β-subunit. Moreover, the α-subunit is typically rapidly degraded by the proteasome in normoxia and is stabilized in hypoxia. 6,7 HIF-1 can perform its transcriptional function only subunits are complexed. [6][7][8] Previous studies have shown that an oxygen deficit is present in DN and that enhancing HIF-1 signalling ameliorates the progression of DN. 3,[9][10][11] Although HIF-1α likely plays a renoprotective role, its precise mechanisms in DN have not yet been fully elucidated. Unsurprisingly, HIF-1α has a connection with mitochondria because both are related to oxygen metabolism. It has been reported that HIF-1α can stimulate the expression of genes encoding proteins involved in the mitochondrial tricarboxylic acid (TCA) cycle 12 and autophagy 13 regulation to improve mitochondrial morphology and function. 14,15 Previous studies by our laboratory showed the importance and characteristics of mitochondrial dynamics in DN, 16,17 and our results are consistent with those from other studies. 16,18,19 More intriguingly, haem oxygenase-1 (HO-1), a target gene of HIF-1α, [20][21][22] has been reported to restrain hypoxia-induced mitochondrial fission. 23 Collectively, these findings highlight the possibility that HIF-1α impacts mitochondrial function and morphology in kidneys of DN. However, the exact molecular mechanism by which HIF-1α protects against mitochondrial dysfunction in tubular cells in the setting of DN is unknown. In the present study, mice with proximal tubular cell-specific genetic ablation of HIF-1α were generated, and mice were induced to diabetes by streptozotocin (STZ) injection. In addition, a hypoxic cell model was established by treatment with cobaltous chloride (CoCl 2 ), and these cells were then cultured and subjected to various treatments to investigate related mechanisms, aiming to exploring new therapeutic targets for DN. \| Generation of mice with conditional knockout of HIF-1α HIF-1α-floxed mice were purchased from the Jackson Laboratory. \| Design in the animal experiment Nine-week-old male PT-HIF-1α −/− mice and littermate fl/fl mice were randomly divided into four groups: fl/fl mice, PT-HIF-1α −/− mice, diabetic fl/fl mice and diabetic PT-HIF-1α −/− mice. These mice were fasted overnight and then intraperitoneally injected with STZ (50 mg/kg body wt; Sigma-Aldrich) consecutively for 5 days as previously described. 25 Three days after the STZ injection, mice with blood glucose levels of >16.7 mmol/L were selected as diabetic mice for the experiment, and the mice were euthanized after 12 weeks. All mice were monitored by the Animal Care and Use Committee of Central South University (China), and all experiments were performed in precise accordance with the established regulations. \| Assessment of metabolic and physiological parameters The body weight and blood glucose level were measured biweekly, and blood and urine were collected before euthanasia. The blood glucose level and blood samples were assessed as described previously. 17 Urine N-acetyl-β-d glucosaminidase (NAG) was assessed using an automated colorimetric method (Pacific). Urinary creatinine and albumin were measured with a creatinine assay kit and an Albuwell M kit (Exocell). 24 \| Histopathological analysis of kidney Kidney tissue was excised, cut, fixed with paraformaldehyde and embedded in paraffin. Then, these kidney tissue blocks were sliced into four micrometre thick sections. To evaluate the kidney tissue damage, the sections were stained with haematoxylin-eosin (HE) and periodic acid-Schiff (PAS), 26 and these changes were scored with a semiquantitative scoring system (0-3). 27 \| Immunohistochemical (IHC) staining For IHC studies, paraffin-embedded kidney sections were deparaffinized and hydrated using slide warmers and alcohol. After antigen retrieval, the sections were permeabilized with 3% H 2 O 2 and blocked with 5% bovine serum albumin (BSA). Then, the sections were individ- \| Immunofluorescence (IF) staining The above-described paraffin-embedded kidney sections were also used for IF staining according to previously published procedures. 26 Briefly, the sections were successively labelled with an anti-HIF-1α primary antibody (ab179483) and the corresponding secondary antibody. \| Establishment of a hypoxic cell model and treatment of cells A human proximal tubular cell line (HK-2) was acquired from ATCC, and cells were cultured in medium as described previously. 26 To establish the hypoxic cell model, the cells were treated with 300 mmol/L CoCl 2 for 24 hours to 28 \| Examination of mitochondrial morphological changes by electron microscopy (EM) and fluorescence staining Mitochondria in renal tubules were observed by EM as previously described. 17 HK-2 cells were stained with MitoTracker Red (Life Technologies) as previously described. 17 \| Analysis of apoptosis and reactive oxygen species (ROS) production Paraffin-embedded sections were subjected to terminal deoxynucleotidyl transferase dUTP nick end-labelling (TUNEL) following the manufacturer's instructions. 29 A dihydroethidium (DHE) (Invitrogen) probe and MitoSox Red (Invitrogen) were used to evaluate intracellular ROS accumulation in renal tubular tissues and HK-2 cells, respectively. \| Western blot analysis Renal tissue and HK-2 cells were collected, and total protein was extracted. In addition, mitochondrial and cytoplasmic proteins from HK-2 cells were isolated using a Cell Mitochondria Isolation Kit according to the manufacturer's instructions. 26 Primary proximal tubular epithelial cells were isolated from mice (Method S1), and total protein was extracted. A total of 30 µg of protein were loaded onto an 8%-12% gel and separated via sodium dodecyl sulphate (SDS)polyacrylamide gel electrophoresis (PAGE). Then, proteins were transferred to a membrane and incubated with the following primary antibodies at the appropriate concentrations: antibodies against HIF-1α (ab2185) and HO-1 (ab13248); antibodies against Mfn1 \| Mitochondrial membrane potential (ΔΨm) assay The mitochondrial membrane potential was evaluated with a specific dye. In brief, HK-2 cells were stained with tetramethylrhodamine (TMRE, Molecular Probes) at a final concentration of 1 µmol/L for 30 minutes and were then visualized with confocal microscopy. \| Statistical analyses The statistical significance of the difference between two groups was evaluated by an unpaired t test in animal and cell studies, and the significance of the difference was denoted by the P value (P < .05, P < .005, P < .001). All statistical analyses were implemented in Graph Pad Prism 8. Figure 1H). As shown in Figure 1I, the scores of evaluation of tubular and glomerular damage in diabetic PT-HIF-1α −/− mice were the highest. Moreover, the DHE probe and TUNEL assay revealed notable increases in ROS production and apoptosis, respectively, in diabetic PT-HIF-1α −/− mice compared to diabetic fl/fl mice ( Figure 1J,K). In addition, we found greatly increased expression of cleaved caspase-9 and caspase-3 in diabetic PT-HIF-1α −/− mice compared with diabetic fl/fl mice ( Figure 1L,M). (Figure 2A,B). IHC staining showed that HO-1 expression was substantially increased in diabetic mice compared to non-diabetic mice and was clearly reduced in diabetic PT-HIF-1α −/− mice compared to diabetic fl/fl mice ( Figure 2C, D). Figure S1. Moreover, hypoxia induced proapoptotic proteins expression, such as cleaved caspase-9 and caspase-3 ( Figure 3G,H). \| Disruption of HIF-1α exacerbates hypoxiainduced mitochondrial damage in HK-2 cells To explore the effects of HIF-1α, HK-2 cells were cultured in hypoxia and treated with the HIF-1α inhibitor KC7F2 or transfected with a HIF-1α plasmid. As shown in Figure \| Lack of HIF-1α aggravates mitochondrial dysfunction in HK-2 cells exposed to hypoxia through HO-1 Cellular IF staining showed that HO-1 agonist hemin further increased HO-1 expression compared to that in hypoxia and that KC7F2 treatment could not block the effect of hemin ( Figure 5A). \| D ISCUSS I ON This study demonstrated that HIF-1α improved mitochondrial dysfunction and restricted mitochondria-dependent apoptosis in tubular cells of DN via the HO-1 pathway ( Figure 6). This study provided the first demonstration of the protective role of HIF-1α in tubular cell injury in mice with STZ-induced DN. Moreover, a novel mechanism was proposed wherein the HIF-1α/HO-1 pathway is the pivotal pathway mediating tubular cell mitochondrial dynamics in DN. Many studies have demonstrated that hypoxia occurs when there is an imbalance between oxygen supply and consumption, and this imbalance is deemed the major driver of DN. 1,3 In view of this information, we attempted to define the relationship between hypoxia and kidney damage in DN. Semenza et al 7 were the first to F I G U R E 4 Effects of HIF-1α on hypoxia-induced mitochondrial damage in HK-2 cells. A, Representative Western blot bands indicating HIF-1α expression in HK-2 cells treated with KC7F2 (a HIF-1α inhibitor) or transfected with a HIF-1α plasmid in hypoxia. B, Representative mitochondrial morphology, ROS generation and mitochondrial membrane potential as detected by MitoTracker Red (upper), MitoSox Red (middle) and TMRE (below) staining, respectively. C-E, Histograms depicting mitochondrial fragmentation (C), the ROS production (D) and mitochondrial membrane potential (E). F and G, Representative Western blot bands (F) and relative densities (G) of Mfn1, Mfn2, Drp1, p-Drp1(S-616) and Fis1 expression in HK-2 cells treated as described in A. H and I, Expression of mito-Cyto.C, mito-Bax, cyto-Cyto.C, cyto-Bax, c-caspase-9 and c-caspase-3 was revealed by Western blotting(H) and related densitometric analysis (I). J, The HO-1 mRNA levels in HK-2 cells were measured by real-time PCR analysis. K and L, The protein expression of HO-1 was assessed by Western blotting (K) and quantitative analysis (L). P < .05 vs the normoxia group; # P < .05 vs the hypoxia group. n = 3 \| 11 of 14 JIANG et Al. report on HIF-1, and their study expanded on the identify of HIF-1 as a heterodimer that can respond to hypoxia. The investigation of HIF-1α is particularly valuable in kidney disease since HIF-1α is expressed predominantly in tubular cells, and tubules are susceptible to hypoxia. 8,31,32 Furthermore, studies have shown that HIF-1α governs the initial adaptive response to hypoxia. 3,8 Moreover, recent observations have shown that proximal tubular cells are the initiators of and critical therapeutic targets in diabetic kidney disease. [32][33][34][35] To explore the effects of HIF-1α on tubular damage in DN, mice with proximal tubular cell-specific HIF-1α ablation were generated, and these mice were treated with STZ to establish a mouse model of DN. The absence of HIF-1α in proximal tubular cells aggravated the kidney damage as evidenced by alterations in morphology and function. These data revealed that HIF-1α plays a protective role in diabetic kidneys, consistent with the observation that HIF-1α deficiency promotes renal injury in DN. 36,37 Interestingly, studies have suggested that HIF-1α can influence mitochondrial morphology and function. 12,14,15 In addition, our previous studies proved that mitochondria play a vital role in DN, especially regarding mitochondrial dynamics. 16,17,29,38 However, whether mitochondria are involved in the protective effect of HIF-1α against tubular injury in DN is still unknown. To explore this possibility, CoCl 2 was applied to HK-2 cells to mimic hypoxia because it is widely used and recognized as a cell model for chemically induced hypoxia; this model is relatively simple, and the conditions are easy to control. Moreover, treatment with an appropriate concentration of CoCl 2 simulates a hypoxic state, including the induction of HIF-1α expression 39 in cells such as tubular cells. 28 In the present study, we found that the lack of HIF-1α in tubular cells of diabetic kidneys aggravated mitochondrial injury. This study showed that the expression of HIF-1α was increased in DN kidneys of DN, consistent with the results of other studies. 3,9 A remaining question is why the level of HIF-1α is elevated in diabetic kidneys when it has protective impacts in renal tissue. The increasing expression and activity of HIF-1α is one type of impaired compensatory response. 40 After the level of HIF-1α is increased, its downstream target genes are activated to suppress hypoxia-induced damage to cells and organs, which exerts renoprotective effects. 9,21 In addition, evidence from recent studies indicates that the activity of HIF in tubules is a lack of compensatory response in DN. 9,30 The possible reasons are as follows: first, superoxide (O 2 − ) may reduce HIF-1 activity by inducing α subunit degradation 3,40,41 ; second, hyperglycaemia could suppress HIF-1α responsive transactivation in tubular cells. [42][43][44] In vitro experiments confirmed that the high glucose-stimulated increase in HIF-1α expression was not significant in tubular cells. F I G U R E 5 Effects of HO-1 on HIF-1α-regulated mitochondrial fragmentation, ROS generation and apoptosis in HK-2 cells. A, Confocal images showing HO-1 expression (upper) and mitochondria (middle) in HK-2 cells in the normoxia, hypoxia, hypoxia + KC7F2, hypoxia + hemin and hypoxia + hemin+KC7F2 groups. Colocalization of HO-1 with mitochondria was assessed (bellow). B, Confocal images of HK-2 cells stained with MitoSox Red and TMRE. C-E, The proportion of fragmented mitochondria (C) and the fluorescence intensity of MitoSox Red (D) and TMRE (E) were quantified and are shown in the histogram. F and G, Western blot analysis showing the protein expression levels of HO-1, Mfn1, Mfn2, Drp1, p-Drp1(S-616) and Fis1 in HK-2 cells treated as indicated (F), and the bands were analysed by densitometry(G). H and I, Western blot analysis of the levels of Bax, Cyt.C, c-caspase-9 and c-caspase-3 (H) and densitometric analysis of the bands (I). P < .05 vs the normoxia group; # P < .05 vs the hypoxia group; ※ P < .05 vs the KC7F2 treatment group. n = 3 F I G U R E 6 HIF-1α is a transcription factor that regulates HO-1 gene expression, and HO-1 facilitates mitochondrial fusion (Mfn1 and Mfn2) and inhibits mitochondrial fission (Drp1 and Fis1), thus maintaining mitochondrial homeostasis. Under diabetic conditions, kidney tissues are hypoxic, which induces the expression of HIF-1α through an impaired compensatory response. However, tubular-specific deletion of HIF-1α in the kidneys of diabetic mice exacerbated mitochondrial dysfunction and ROS accumulation and then caused tubular damage, thereby leading to diabetic kidney injury Mitochondrial dynamics describes the two features of mitochondria, which manifest as fusion and fission. [45][46][47] In addition, HO-1 may be the intermediary molecule linking HIF-1α and mitochondrial dynamics. HO-1 is a target gene of HIF-1α 22 and exerts protective effects against kidney damage under diabetic conditions. 23 Studies have shown that HO-1 exerts anti-apoptotic, antioxidant, anti-nitrosative and anti-inflammatory effects, 48-50 by reducing the level of the pro-apoptotic protein Bax and proinflammatory/prooxidant protein iNOS, [49][50][51] and increasing the level of the anti-apoptotic protein Bcl-xl and anti-nitrosative protein bilirubin. 49,50,52 Importantly, recent studies have verified that HO-1 activity affects the function and morphology of mitochondria, especially mitochondrial dynamics. [53][54][55] On the basis of these findings, we sought to determine whether HO-1 is the key modulator of HIF-1α-mediated regulation of mitochondrial dynamics in tubular cells. As expected, under hypoxic conditions, HK-2 cells exhibited increased mitochondrial fission, mitochondrial ROS generation and apoptosis, while these effects were further enhanced by HIF-1α inhibitor treatment. While clearly of great importance, in HK-2 cells cotreated with a HIF-1α inhibitor and a HO-1 agonist after exposure to hypoxia the HO-1 agonist rescued the cells from the negative impact of the HIF-1α inhibitor. These results suggest that HIF-1α is the pivotal node upstream of HO-1 expression, identifying a novel pathway through which HIF-1α modulates mitochondrial dynamics via HO-1 in tubular cell injury in DN. A recent question addresses the specific molecular mechanism by which HO-1 modulates mitochondrial dynamics. Studies speculated that HO-1 can regulate mitochondrial dynamics 53,54,56 and showed that HO-1/CO may mediate mitochondrial dynamics in leukaemia. 55 These results indicate that HO-1-mediated regulation of alterations in mitochondrial dynamics in tubular cells of DN may occur through the HO-1/CO pathway, a possibility that requires further investigation. Iron overload can damage mitochondria, and HO-1 participates in iron metabolism by degrading haem into ferrous iron. 57,58 However, the increase in HO-1 is also accompanied by ferritin upregulation and leads to effluxion of cellular iron. 59,60 Most importantly, the increase in HO-1 causes a decrease in intracellular free iron content and thus mitigates mitochondrial dysfunction. This study provides a novel perspective on HIF-1α in DN-by not only developing a new research field in diabetic kidney disease but also laying a foundation for the identification of promising therapeutic targets. In summary, HIF-1α plays a vital role in tubular cell injury in DN and is thus a potential therapeutic target. CO N FLI C T O F I NTE R E S T No competing interests related to this article are reported. AUTH O R CO NTR I B UTI O N S NJ performed the experiments, generated the data and wrote the manuscript. HZ, YCH, SX, LFZ and XFX performed statistical analyses of the data and discussed the results of the manuscript. LL, MY, YX, LW and PG partially edited the manuscript. YL provided technical support for this study. LS is the guarantor of this study, who conceived and designed this study and edited and discussed this manuscript. DATA AVA I L A B I L I T Y S TAT E M E N T The data that support the findings of this study are available from the corresponding author upon reasonable request.	2020-09-26T13:05:57.278Z	2020-09-25T00:00:00.000	{ "year": 2020, "sha1": "41111d45f7aa07e3bc986688384c189c121fb020", "oa_license": "CCBY", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1111/cpr.12909", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "2506de4a6f61c34e04f2ebf7e66dbcf2501e2998", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Chemistry", "Medicine" ] }
211833939	pes2o/s2orc	v3-fos-license	Cytokine Expression in Dengue Fever and Dengue Hemorrhagic Fever Patients with Bleeding and Severe Hepatitis Abstract. Dengue is the most common mosquito-borne flaviviral infection in the world today. Several factors contribute and act synergistically to cause severe infection. One of these is dysregulated host immunological mediators that cause transient pathophysiology during infection. These mediators act on the endothelium to increase vascular permeability, which leads to plasma leakage compromising hemodynamics and coagulopathy. We conducted a prospective study to explore the expression of pro- and anti-inflammatory cytokines and how they relate to clinical dengue manifestations, by assessing their dynamics through acute dengue infection in adults admitted to the Hospital for Tropical Diseases, Bangkok, Thailand. We performed cytokine analysis at three phases of infection for 96 hospitalized adults together with serotyping of confirmed dengue infection during the outbreaks of 2015 and 2016. The serum concentrations of seven cytokines (interleukin [IL]-2, IL-4, IL-6, IL-8, IL-10, tumor necrosis factor alpha, and interferon gamma) were measured in duplicate using a commercial kit (Bio-Plex Human Cytokine Assay). In this study, the cytokine profile was suggestive of a T-helper 2 response. Most patients had secondary infection, and the levels of viremia were higher in patients with plasma leakage than those without plasma leakage. In addition, we observed that bleeding and hepatitis were associated with significantly higher levels of IL-8 during the early phases of infection. Furthermore, IL-6 levels in the early phase of infection were also elevated in bleeding patients with plasma leakage. These results suggest that IL-6 and IL-8 may act in synergy to cause bleeding in patients with plasma leakage. INTRODUCTION The dengue virus is a flavivirus of the family Flaviviridae and has been characterized into four serotypes (Dengue viruses 1 to 4). Transmitted by Aedes species, dengue infection is the most commonly occurring arthropod-borne viral infection globally. Several factors such as globalization, urbanization, and lack of effective vector control have facilitated the spread of this disease beyond the subtropics. 1 It is estimated that 50-400 million infections occur annually. In Thailand, the first epidemic of dengue hemorrhagic fever (DHF), a severe form of dengue virus infection, occurred in 1958. Multiple outbreaks with greater epidemic potential have occurred since then, with a range of 15,000-105,000 cases occurring each year in Thailand. 2 The infection causes a systemic viral infection, and complications are associated with the pathogenic role of endothelial cells. [3][4][5] Cytokines 6,7 increase vascular permeability and hemorrhage during dengue infection. These molecules are proteinaceous and are secreted during innate and adaptive immunological responses, acting as inflammatory mediators or modulatory molecules during dengue infection. Viral infection activates the innate immune response to express pro-inflammatory cytokines that recruit and activate cells involved in inflammation and the induction of adaptive immunity. Cytokines also help distinguish between T-helper 1 and T-helper 2 (Th2) cells, and their expression in dengue has been reviewed extensively. 8 The dengue virus is tropic and replicates in human monocytes, macrophages, and hepatocytes, which are known to express cytokines. 9 The endothelium plays a pivotal role in the pathophysiology of dengue infection. 6 Activation of the endothelium by the expression of interleukin (IL)-6, IL-8, and tumor necrosis factor alpha (TNFα) have been previously described. 3,10 This activation of the endothelial system not only contributes to the bleeding, which is common in dengue fever (DF) and DHF, 11 but also to liver damage, leading to hepatitis complicating dengue. 12 Bleeding manifestations are seen in 20-60% 13 of cases, and hepatitis with a 100-fold increase in transaminase was reported in 7% of cases. 14 Thrombocytopenia with impairment of the coagulation system is associated with bleeding in dengue, 15 and multiple factors contribute to the development of hepatitis. These factors include hypoxic injury caused by decreased perfusion, direct damage by the virus, and dysregulated immune-mediated injury in response to dengue virus. Interleukin-8 is expressed by hepatocytes during dengue infection, 16 with reports of cytokines correlating with elevated transaminase levels 17 and the expression of TNFα following the apoptosis of hepatocytes in dengue infection. 18 Others reported a correlation between severity and IL-6 and TNFα levels. 19 However, only limited data are available for the relationship between the increased expressions of cytokines with hepatitis and bleeding. In the present study, we investigated the expressions of cytokines during dengue infection with bleeding and hepatitis and assessed the degree of viremia with plasma leakage. MATERIALS AND METHODS This was a prospective study of hospitalized adults with symptomatic dengue infection at the Hospital for Tropical Diseases, Bangkok, Thailand. Ethical approval to conduct this study was obtained from the Ethics Committee, Faculty of Tropical Medicine, Mahidol University. During the dengue outbreak in 2015 and 2016, we recruited adults who presented within 5 days after developing symptoms and with a positive non structural protein 1 antigen or a positive anti-dengue IgM test into our study. We excluded any vulnerable groups and participants who had received a transfusion of blood products during the study period. Because this was an observational study, all consenting participants meeting the study criteria were recruited during the 2year study period. The participants were grouped based on the WHO 1997 classification, and a chest X-ray was performed to detect pleural effusion. Complications in dengue infection included bleeding and severe hepatitis (aspartate aminotransferase/alanine aminotransferase > 400 U/L). Cytokine measurement. Blood specimens were collected for cytokine analysis during the three phases of infection. A commercial assay (Bio-Plex Human Cytokine Assay; Bio-Rad Inc., Hercules CA) was performed to detect the levels of IL-2, IL-4, IL-6, IL-8, IL-10, interferon gamma (IFNγ), and TNFα. In brief, sera obtained from participants during the three phases were mixed with beads coated with antibodies to cytokines and a unique fluorescent intensity. Subsequently, the mixtures were incubated with biotinylated anti-cytokine antibodies. Finally, phycoerythrin-conjugated streptavidin was added, and the fluorescent signals were detected using a multiplex array reader (Bio-Plex 200 System, Bio-Rad Inc., Hercules, CA). Raw data were initially measured as the relative fluorescence intensity and then converted to cytokine concentration based on a standard curve generated from the reference concentrations. The acute phase specimen corresponded to blood collected within 5 days following the development of symptoms of dengue infection. The defervescence specimens were obtained 8 hours after the participants had remained afebrile with a recorded body temperature of 37.7°C or lower. The convalescence specimens were obtained during the follow-up within or later than 2 weeks after the onset of symptoms. The standard provided by the manufacturer was used as a control, and the observed readings were determined as the detection limits in pg/mL. The clinical findings are represented as the actual number and percentage. The laboratory parameters are represented as the median ± SD. The chi-square test was used to determine the differences in clinical findings in both groups. The Mann-Whitney U-test was used to determine the P-value between two groups using the median value. Bold indicates statistically significant P-values. Quantification of RNA and sequence of infection. To determine the dengue serotype, a commercially available dengue subtyping multiplex kit by the Genesig company (Chandler's Ford, United Kingdom) was used as per the manufacturer's instruction. 20 Then, RNA quantification was performed using a One-Step SYBR ® Prime Script RT-PCR Kit II, Takara Bio Inc., (Kusatsu, Japan); cDNA synthesis from RNA was performed using reverse transcriptase Prime Script RTase; and polymerase chain reaction amplification was performed by TaKaRa Ex Taq HS. This technique was previously described. 21 For dengue serology, commercially available dengue IgM and IgG immunochromatography kits from Panbio Dengue Duo Cassette Abbott Inc., (Chicago, IL) were used in accordance with the manufacturer's protocol. A serum aliquot representing the day of defervescence was used in these test kits. Primary infections were diagnosed when only anti-dengue IgM was positive together with the appearance of the control band. Secondary infections were determined when both anti-dengue IgM and IgG were positive, including the appearance of the control band. Statistical analysis. The Mann-Whitney U-test was used to assess differences in the cytokine levels between DF and DHF. To establish the correlation between cytokine levels and clinical parameters/findings, a correlation matrix was applied. Results are given as the correlation coefficient; r (range, from −1 to +1). A two-tailed P-value < 0.05 was considered significant for all tests performed. One-way analysis of variance was used to evaluate differences among cytokine levels in the different groups. All statistical analyses were performed using Microsoft Excel, SPSS version 18, and GraphPad Prism 7 for Windows, version 7.30 (San Diego, CA). RESULTS In this study, 96 hospitalized adult patients with a confirmed dengue infection were recruited during the outbreak in 2015 and 2016. The demographic and clinical data as well as laboratory parameters of the study population and clinical details are shown in Table 1. Our results indicate that symptoms of systemic illness including headache (P = 0.02), myalgia (P = 0.03), lymphadenopathy (P = 0.02), and bleeding (P = 0.05) were more common in DHF. A longer duration of hospitalization was observed in DHF cases than DF cases (P = 0.01). The hematological profile was typical for dengue infection, with leukopenia and thrombocytopenia (platelet count < 100 × 10 3 /μL) being obvious findings in our subjects. From the analysis of liver function tests, transaminitis was observed in 11% of subjects. From the analysis of liver function tests, transaminitis was observed in 11% of subjects. The mean ± SD levels of aspartate aminotransferase and alanine aminotransferase for DF were 224 ± 399 U/L and 127 ± 192 U/L, and for DHF were 308 ± 528 U/L and 122 ± 184 U/L, respectively. Cytokine expressions during dengue infection. Figure 1 shows the trend of the expressions of IL-4, IL-6, IL-8, IL-10, TNFα, and IFNγ during the three phases of dengue infection. The expressed cytokines peaked during the acute phase. Interleukin-10 was most expressed during the acute phase followed by TNFα, IL-8, IL-6, IFNγ, and IL-4. Interleukin-2 was not detectable. At defervescence, TNFα was the most expressed, and during convalescence, IL-4 had diminished to low levels and was undetectable in most cases. We investigated the cytokine profile by grouping our subjects based on their clinical classification. These included 59 subjects with DF and 37 with DHF, of whom 17 were of DHF grade I, 19 were of DHF grade II, and a single participant was of DHF grade IV as per the WHO 1997 classification. 22 Most patients had secondary dengue infection and had an uneventful milder clinical course during hospitalization. Complications identified in our study were shock in a single subject, and bleeding in 42% and hepatitis in 11% of study subjects. The bleeding manifestations were mucosal bleeding in 31.6%, menorrhagia in 27.5%, epistaxis in 5.1%, melena in two subjects, and hematemesis in a single subject. Our results revealed no significant differences in cytokine levels during the three phases of infection between subjects with and without plasma leakage (Figure 2). Nevertheless, we did observe higher viremia in those with plasma leakage (Figure 3). Cytokines profile with bleeding. Interleukin-8 expression was significantly elevated during the acute phase and at defervescence. The IL-8 median value (interquartile range) was 27.65 (range, 17.42-37.42) pg/mL for the group with bleeding during the acute phase and 17.98 (range, 13.06-25.95) pg/mL for the group without bleeding (P = 0.00, Figure 4). Similarly, at defervescence, the IL-8 median value (IQR) was 21.76 (range, 16.86-30.14) pg/mL for the group with bleeding and 16.19 (range, 11.18-25.30) pg/mL for the group without bleeding (P = 0.01, Figure 4). These data represent the detectable levels of cytokines in 96 subjects during the three phases of dengue infection. The cytokine levels are shown as the median and IQR in pg/mL. The Mann-Whitney U-test was used to determine the P value between two groups using the median value. Similarly, the levels of IL-8 were significantly higher in DHF grade II-IV patients than DHF grade I patients ( Figure 5). In addition, IL-6 levels during the acute phase were significantly elevated in DHF grade II-IV compared with DHF grade I ( Figure 5). The levels of IL-8 were also significantly higher in DF with bleeding than DF without bleeding during the acute phase of infection ( Figure 6). Cytokine profile with hepatitis. In the cytokine profile generated for severe hepatitis (AST/AST > 400 U/L), we observed that in addition to IL-8 being significantly elevated during the acute phase (P = 0.00) and at defervescence (P = 0.01), TNFα was significantly elevated during the acute phase (P = 0.01, Figure 7). The levels of IL-8 were significantly higher during the acute phase and at defervescence in patients with an increase in liver enzymes (AST/ALT ³ 1,000 U/L). The median value of IL-8 was 30.28 pg/mL (IQR, 25.18-86.58) in subjects with AST/ALT ³ 1,000 U/L, which was significantly elevated during the acute phase (P = 0.03). At defervescence, in subjects with AST/ALT ³ 1,000 U/L, the median IL-8 value was 32.29 pg/mL (IQR, 23.80-57.67), which was significantly elevated compared with subjects with AST/ALT < 1,000 U/L (P = 0.01). DISCUSSION In the present study, the levels of cytokines during dengue infection were measured, and we analyzed the profiled cytokine patterns to determine the immunopathophysiological events that occur during dengue infection leading to complications, such as bleeding and hepatitis. To the best of our knowledge, few studies have performed cytokine analysis during the three phases of infection. In particular, we designed our study to measure the cytokine levels at defervescence and identified elevated cytokine levels with complications. In viral infections, cytokines can interfere with the interferon signaling pathway, 23 which is the host's primary antiviral response to an acute infection. 24 Our study subjects had a diminished IFNγ response, which might have resulted from a secondary dengue infection, where the Th1 response is prominent in primary infection. 25 This phenomenon was previously demonstrated in vitro, where dengue virus-infected cells failed to express or demonstrated a reduced expression of IFNγ on secondary infection. 26 The cytokine response reflecting a T-helper 2 response was supported by our undetectable levels of secreted IL-2, in addition to the very low levels of IFNγ, with an increased expression of IL-10. This trend of cytokine expression also suggests immunoregulatory effects consistent with secondary infection. 27 Memory B cells are primed to produce antibodies by the expression of IL-10, 28 and with its proteolytic properties, IL-10 inhibited IFNγ expression and deregulated the expressions of IL-6, IL-8, and TNFα in our study subjects and as previously described. 29 Unlike previous reports that showed an association of IL-4 and disease severity, 30 our study failed to reveal such findings probably because IL-4 was detected in only a few cases partly because of the small number of study subjects. We demonstrated endothelial activation by the detection of IL-6, IL-8, and TNFα as previously described. 3,10 This activation of the endothelial system not only contributes to bleeding, which is common in DF and DHF, 11 but also involved in liver damage, leading to hepatitis complicating dengue. 12 Our results were consistent with previous studies in respect to higher viremia in DHF, 31 and we observed more bleeding and symptoms of systemic illness in DHF, reflecting the degree of severity when compared with DF. However, we did not find any association between viremia and bleeding as previously reported. 32 Interleukin 8 is expressed by hepatocytes in dengue infection 16 and was associated with disease severity. 17 Similar to a previous study, 33 our results regarding IL-8 levels could not distinguish DF from DHF. However, we demonstrated significantly elevated levels of IL-8 during the acute phase and at defervescence in patients with bleeding and in those with hepatitis. We also showed an association of an increased IL-8 level with bleeding during both phases. Significantly, elevated levels of IL-6 were observed during the acute phase in patients with bleeding, with a background of plasma leakage, whereas others reported elevated levels of IL-6 in DHF 34 and bleeding. 32 Other studies reported TNFα was not detectable or only detected in 30% of cases of dengue infection. 35 In our study, we detected TNFα expression in all our subjects. We highlighted other factors that contribute to bleeding in dengue, such as thrombocytopenia, 36 which can worsen bleeding in dengue infection. The liver impairment, which occurs in dengue infection, 14 impacts bleeding with a worsening of coagulation. Our results demonstrated that IL-8 and TNFα expressions were significantly higher in patients with transaminases elevated 10-fold. Furthermore, we observed that IL-8 levels were significantly elevated with a 100-fold increase in transaminase levels during both phases. The properties of IL-8 contribute to platelet activation and endothelial permeability to cause thrombocytopenia, 34 which worsens bleeding in dengue infection. We also observed that the platelet counts during both phases were significantly decreased in patients with bleeding. These elevated levels of IL-8 observed in our study might facilitate the events preceding bleeding during dengue infection. A similar observation of elevated levels of IL-8 with thrombocytopenia in dengue was described. 34 During dengue infection, infected hepatocytes express IL-8, attracting neutrophils to the liver, causing liver injury and impaired coagulation. Our results show increased levels of neutrophils during the two phases of infection with hepatitis. This increase in neutrophils with hepatitis might be a result of the elevated levels of IL-8 with hepatitis. Although a positive correlation of IFNγ and transaminase was reported, 34 we observed no similar pattern of correlations. Nevertheless, there were positive correlations between IL-6 and IL-8 and between IL-6 and IFNγ during the acute phase, and at defervescence IL-8 and TNFα were positively correlated, indicating these cytokines may act in synergy. In summary, the immunoregulatory responses observed may be implicated in the bleeding and hepatitis observed in dengue patients. CONCLUSION In this study, higher viremia in DHF with a T-helper 2 response determined by profiling the cytokine expression was demonstrated. The expression of IL-8 was significantly elevated during the acute phase in patients with bleeding in both DF and DHF, with a further increase in IL-8 levels in DHF at defervescence. IL-6 acts in synergy with IL-8 for bleeding with plasma leakage during the acute phase. Significantly, higher levels of TNFα, IL-6, and IL-8 were observed in cases of severe hepatitis. Limitations. There were some limitations in this study including the small number of total study subjects and those with severe disease.	2020-03-04T14:04:52.997Z	2020-03-02T00:00:00.000	{ "year": 2020, "sha1": "dec25f0570c37d4cfc6c214ff5d21488cc21a2f0", "oa_license": "CCBY", "oa_url": "https://www.ajtmh.org/downloadpdf/journals/tpmd/102/5/article-p943.pdf", "oa_status": "BRONZE", "pdf_src": "PubMedCentral", "pdf_hash": "f6453177114a1e26d5ccad406de7b0fd89e263c8", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
266174648	pes2o/s2orc	v3-fos-license	Radio emission as a stellar activity indicator Radio observations of stars trace the plasma conditions and magnetic field properties of stellar magnetospheres and coronae. Depending on the plasma conditions at the emitter site, radio emission in the metre-and decimetre-wave bands is generated via different mechanisms, such as gyrosynchrotron, electron cyclotron maser instability, and plasma radiation processes. The ongoing LOFAR Two-metre Sky Survey (LoTSS) and VLA Sky Survey (VLASS) are currently the most sensitive wide-field radio sky surveys ever conducted. Because these surveys are untargeted, they provide an opportunity to study the statistical properties of the radio-emitting stellar population in an unbiased manner. Here we perform an untargeted search for stellar radio sources down to sub-mJy level using these radio surveys. We find that the population of radio-emitting stellar systems is mainly composed of two distinct categories: chromospherically active stellar (CAS) systems and M dwarfs. We also seek to identify signatures of a gradual transition within the M-dwarf population, from chromospheric or coronal acceleration close to the stellar surface similar to that observed on the Sun to magnetospheric acceleration occurring far from the stellar surface similar to that observed on Jupiter. We determine that radio detectability evolves with spectral type, and we identify a transition in radio detectability around spectral type M4, where stars become fully convective. Furthermore, we compare the radio detectability versus spectra type with X-ray and optical flare (observed by TESS) incidence statistics. We find that the radio efficiency of X-ray and optical flares, which is the fraction of flare energy channelled into radio-emitting charges, increases with spectral type. These results motivate us to conjecture that the emergence of large-scale magnetic fields in CAS systems and later M dwarfs leads to an increase in radio efficiency. Introduction Radio observations are excellent tracers of supra-thermal plasma in the coronae of stars, magnetic activity of ultracool dwarfs (UCDs; including brown dwarfs), and the mangetospheres of planets.This is because plasma oscillation and electric charges moving in a magnetic field emit at radio frequencies, implying a radio detection of a stellar system is sensitive to the magnetic fields and plasma dynamics of the source.Therefore, radio observations probe phenomena on these objects that are otherwise not easily accessible in other wavelengths (Bastian et al. 1998;Güdel 2002).For example, radio emission from exoplanets provides the only known technique to directly measure the planet's magnetic field strength and topology, which is an important aspect of exo-habitability (e.g.Driscoll & Barnes 2015;Airapetian et al. 2020;López-Morales et al. 2011;Khodachenko et al. 2007;See et al. 2014;Grießmeier et al. 2015).Additionally, plasma emission from stars can be used to constrain fundamental coronal plasma parameters such as plasma density and the radial structure and dynamics of coronal plasma (Toet et al. 2021), giving insights on the impact of stellar plasma on exoplanet atmospheres. Depending on the plasma parameters at the emitter, radio emission in the metre-and decametre-wave bands can be generated via a variety of mechanisms: from incoherent ones such as free-free, gyromagnetic, and gyrosynchrotron processes (Nin-⋆ Email: yiu@astron.nldos 2020) to coherent ones such as plasma radiation (Dulk 1985) and electron cyclotron maser instability (ECMI; Wu & Lee 1979;Melrose & Dulk 1982, 1984;Treumann 2006).For the Sun, there exist two principal radio components: the unpolarised or weakly polarised broadband solar flares at gigahertz (GHz) frequencies, and the narrowband coherent solar bursts at megahertz (MHz) frequencies.The unpolarised radio emission from solar flares have generally been attributed to gyrosynchrotron radiation (Alissandrakis 1986), produced by mildly relativistic electrons (γ ≲ 2−3) either arising from the Maxwellian tail of a thermal distribution (Dulk 1985), or from a non-thermal energy distribution produced via flare acceleration and/or magnetic reconnection (Parker 1988;Sweet 1958a,b;Matthews 2019).Many stellar radio flares on isolated main-sequence stars (e.g.Bastian 1990;Güdel & Benz 1993) as well as very close binaries (e.g.Drake et al. 1989;Mutel et al. 1985) are often observed to be broadband and unpolarised.On the contrary, coherent solar bursts with narrow instantaneous bandwidths are much more luminous and more circularly polarised compared to gyrosynchrotron flares.These bursts are typically produced by plasma radiation processes (Dulk 1985;Melrose 1980Melrose , 2009Melrose , 2017;;Güdel 2002), thus emitting at plasma frequencies or its harmonics that are typically below the GHz regime.Regardless of the mechanism, all solar radio-emitting charges are expected to be created via chromospheric/coronal acceleration close to the stellar surface due to magnetic reconnection, driven by immense shear in the solar radiative-convective interface (known as tachocline; Spiegel & Zahn 1992).We refer to this broad class of acceleration mechanisms collectively as the "Sun-like radio engine". In comparison, radio observations of Jupiter show that Jupiter can generate luminous auroral radio bursts with strong circular polarisation (reaching 100%; Zarka 1998Zarka , 2004Zarka , 2007) ) at the ambient cyclotron frequency and its harmonics through the ECMI mechanism.Therefore, Jupiter's plasma dynamics are fundamentally different to the Sun due to its fast rotation rate, strong magnetic field, and low-density magnetosphere.Jovian radio emission is thus magneto-rotation-driven instead, with the magnetospheric acceleration occurring far from the planetary surface as a result of a breakdown of co-rotation between the magnetic field and plasma.An alternate acceleration mechanism known as centrifugal breakout, which also operates far from the surface, has been suggested for radio emission from hot magnetic stars with dipole-dominated magnetopspheres (Shultz et al. 2020;Owocki et al. 2020Owocki et al. , 2022)).We shall use the umbrella term "Jupiter-like engine" for acceleration in a magnetosphere far from the surface of the central body. In contrast to the Sun, the coherent radio emission of UCDs are generally interpreted as electron cyclotron maser (ECM) emission powered by Jupiter-like engines (e.g.Hallinan et al. 2007Hallinan et al. , 2008;;Kao et al. 2016;Turnpenney et al. 2017), since their highly circular-polarised radio periodic bursts share many properties with that of Jupiter and other magnetised solar-system planets (Zarka 2007).The change in radio emission mechanism between solar-like stars and UCDs suggests that there exists a transition in radio-emitting engine of stellar systems -from the Sun-like paradigm to the Jupiter-like paradigm -somewhere in the M-dwarf sequence.Finding this transition is vital in understanding the radio detectability of stars and the predominant emission mechanism of radio-bright stellar systems.It may also shed light on the effect of the transition from partially convective to fully convective interiors (i.e. the tachocline no longer exists) on stellar activity (around spectral type M4; Chabrier & Baraffe 2000;Dorman et al. 1989;Stassun et al. 2011;Williams et al. 2014). There have been previous studies that investigate and characterise the radio properties of stellar systems, from the hottest stars to the coldest brown dwarfs, in order to determine trends (see e.g.Bieging et al. 1989;Güdel & Benz 1993;Williams et al. 2014;Kao et al. 2016;Villadsen & Hallinan 2019).However, all of the above studies employed some target selection strategy, and thus could be implicitly biased towards special cases of stars and UCDs that may not reflect the general stellar population as a whole.One benefit of searching for radio-bright stellar systems in wide-field radio surveys is therefore to bypass these biases.However, this method is not without some major challenges.Helfand et al. (1999) performed the first large unbiased study of radio stars using the Faint Images of the Radio Sky at Twenty-Centimeters (FIRST; Becker et al. 1995), and concluded that FIRST astrometry alone was insufficient to avoid crippling chance-coincidence rates, especially when good proper-motion information was unavailable.Kimball et al. (2009) conducted a search for radio stars by combining FIRST with optical data from the Sloan Digital Sky Survey (SDSS), but they estimated that 108±13 out of 112 candidate radio stars were contamination, i.e. optical-faint radio quasars in chance alignment with a foreground star.Thus, they concluded a radio survey with a much higher sensitivity and resolution compared to FIRST was necessary to confidently identify radio stars.Antonova et al. (2013) conducted a 4.9-GHz volume-limited radio survey of 32 nearby UCDs with spectral types M7-T8, but failed to detect any ra-dio emission from them.They thus highlighted that low rotation rates and long-term variability of these targets might be the cause of the non-detections.Lenc et al. (2018) conducted the first lowfrequency circular-polarised all-sky survey using the Murchison Widefield Array (MWA; Tingay et al. 2013).However, with its relatively low sensitivity of ≈ 3mJy beam −1 , the survey only detected pulsars which were previously known to be radio-bright. Studies with a newer crop of radio surveys have been more successful.Pritchard et al. (2021) presented results from a circular polarisation survey for radio stars in the Rapid Australian Square Kilometre Array Pathfinder (ASKAP) Continuum Survey (RACS;McConnell et al. 2020), and they identified M dwarfs, close binaries, young stellar objects, and chemically peculiar A-and B-type stars in the radio sample.Callingham et al. (2021b) detected coherent radio emission of M dwarfs from an untargeted flux-limited low-frequency survey using LOw-Frequency ARray (LOFAR; van Haarlem et al. 2013).They highlighted the benefit of utilising circular polarisation (Stokes V) information to suppress false associations (Callingham et al. 2019).More recently, Driessen et al. (2023) identified radio stellar sources from multiple radio surveys using their proper motions provided by Gaia Data Release 3 (DR3; Gaia Collaboration et al. 2022), and showed that the transient nature of radio stellar sources makes surveys with multiple epochs more preferable for searching them. Having a wide-field radio survey with high sensitivity and angular resolution is therefore paramount in order to identify radio-bright stellar systems from the radio data, lest nondetections and extragalactic contamination dominate the sample.The ongoing LOFAR Two-metre Sky Survey (LoTSS; Shimwell et al. 2017) and the Karl G. Jansky Very Large Array (VLA; Perley et al. 2011) Sky Survey (VLASS; Lacy et al. 2020) are some of the most sensitive wide-field radio sky surveys ever conducted.Because the surveys are untargeted, they provide an opportunity to study the statistical properties of the radio-emitting stellar population in an unbiased manner.Moreover, Callingham et al. (2023) recently published V-LoTSS: the circular-polarised component of LoTSS.This provides yet another advantage as the Stokes V radio sky is orders of magnitudes sparser than its Stokes I counterpart (Callingham et al. 2023).In this paper, we present our latest efforts to identify and study the radio emission from stellar systems in these surveys as a way to understand if and how radio activity evolves with spectral type, and see whether a transition in acceleration mechanism (the so-called radio-emitting engine) exists, specifically from the aforementioned Sun-like to Jupiter-like engine as one goes from earlier to later spectral types.Additionally, to understand how stellar activity impacts radio luminosity and thus detectability, we also compare the radio detection rate in our sample to canonical activity indicators such as optical and X-ray flares. The paper is structured as follows: in Sect.2, we present details of the radio survey catalogues, Gaia Catalogue of Nearby Stars (GCNS) and the TESS/X-ray flare statistics used in our analysis.Sect. 3 contains the description of our methodology, including the radio × GCNS crossmatching and TESS-flare-rate debiasing procedure.We present our results in Sect. 4 and conclude in Sect. 5.This paper contains many acronyms, and thus we include a table of acronyms (Table D.1) in the Appendix for clarity.Notes.The symbols ν survey , AA, t exp , and Ω survey represent radio survey frequency coverage, absolute astrometric accuracy, exposure time, and the sky coverage respectively.Note that the final uncertainty in the radio position must also factor the astrometric precision that depends on the signal-to-noise ratio (SNR) of the detection.For example, LoTSS source with <1 mJy has an average astrometric precision of 0.5 ′′ (Callingham et al. 2019).The Ω survey of V-LoTSS is slightly larger since Callingham et al. (2023) searched for sources further down the primary beam of single LoTSS pointing, where the noise would have been higher than if the data was mosaicked. Radio sky surveys In order to draw conclusions regarding the radio-bright stellar population as a whole, we begin by crossmatching different radio catalogues to optical positions of known stars in the solar neighbourhood.The following section details the radio sky surveys used in this work.A summary of the important parameters of these surveys can be found in Table 1. LOFAR Two-metre Sky Survey The LOFAR Two-metre Sky Survey (LoTSS; Shimwell et al. 2017) is an ongoing low-frequency (120-168 MHz) radio survey of the Northern Sky.Each LoTSS pointing is observed for 8 hours, reaching a median 1σ rms noise of around 83 µJy/beam. Here we use the LoTSS radio catalogue from the second data release (DR2; Shimwell et al. 2022) covering 27% of the Northern Sky (5634 square degrees) containing around 4 million Stokes I sources.The unprecedented depth of LoTSS allows us to reveal a low-frequency radio stellar and substellar population neverbefore-seen as previous searches below ∼300 MHz generally lacked the required sub-mJy sensitivity necessary to detect the general population (Callingham et al. 2021b;Vedantham et al. 2020Vedantham et al. , 2023)).However, using the LoTSS Stokes I catalogue alone to identify radio stellar system candidates is not straightforward due to the potentially crippling rate of chance-coincidence associations (i.e. the so called false-positive matches; more details in Sect.3.1.1).To circumvent this, we also include the available circular polarisation (Stokes V) information of LOFAR-detected stellar sources, described in the following section. Circular-polarised sky of LoTSS The source density of the Stokes V radio sky is > 5 orders of magnitude lower than the Stokes I radio sky because most radio sources are extragalactic objects powered by the synchrotron mechanism (Begelman et al. 1984), and therefore do not have a significant degree of circular polarisation (Rayner et al. 2000;Beckert & Falcke 2002).As anticipated, only 1 extragalactic source (an active galactic nucleus) has a detectable circular polarisation (≈ 1%) in LoTSS-DR2 (Callingham et al. 2023).In rare cases where extragalactic radio sources are circularly polarised, the known radio-emission mechanisms of these objects would only allow them to have at most ≈ 1% circular polarised fraction (Saikia & Salter 1988;Valtaoja 1984;Wardle & Homan 2003), which matches with observations (e.g.Macquart et al. 2003;Agudo & Thum 2022). On the other hand, the circular polarisation of some stellar radio sources are known to be able to reach ≈ 100% (e.g.Hallinan et al. 2006Hallinan et al. , 2007Hallinan et al. , 2008;;Williams & Berger 2015) owing to plasma radiation processes and ECMI (Dulk 1985;Vedantham 2021).Callingham et al. (2023) presented V-LoTSS, which consists of Stokes V maps of LoTSS-DR2 with median 140 µJybeam −1 and a resolution of 20 ′′ .As we expect the radio emission of stellar systems to be time variable, an overall mosaic of the fields may wash out genuine detections as these sources may not be emitting in Stokes V in an adjoining field that was observed on a different date.Therefore, using V-LoTSS is again advantageous since unlike LoTSS-DR2, V-LoTSS performs the Stokes V search on individual LoTSS pointings rather than on a mosaicked image. VLA Sky Survey As the aim of this paper is to determine how much radio emission is indicative of stellar activity, we are also interested in the radio population that has a higher frequency than the LOFAR band, as different radio frequencies probe different radio emission mechanisms (Güdel 2002).For example, a flaring star can emit intense gyrosynchrotron radiation which typically falls under the decimetre-wave regime (Nindos 2020), a frequency range observable by the VLA.However, such a star might not have detectable emission in the LOFAR band due to self-absorption and brightness temperature limitations: the brightness temperature of gyrosynchrotron emissions cannot exceed inverse Compton limit of T B ∼ 10 12 K (Kellermann & Pauliny-Toth 1969).Since T B ∝ ν −2 , it can be trivially shown that gyrosynchrotron cannot produce emission in the LOFAR band detectable with the LoTSS sensitivity.Conversely, stellar systems detected in the metrewave regime are much more likely to be due to plasma radiation processes and/or ECMI (Vedantham 2021).These stars may have a large-scale magnetic field of few hundred gauss which can produce strong coherent ECM radiation that peaks at LO-FAR band but decays sharply at VLA band.Therefore, we also utilise the VLA Sky Survey (VLASS; Lacy et al. 2020) to select radio-bright stellar systems, and see how different the VLA-detected stellar population is when compared to the LOFAR-detected population.VLASS is an ongoing multi-epoch S-band (2)(3)(4) continuum radio survey covering the whole sky visible to the VLA (i.e.δ > −40 • ; 33885 square degrees) and aims to produce images with high angular resolution (≈ 2.5 ′′ ) and 1σ rms noise of ≈ 130µJy.Three epochs of observation (with two sub-epochs each) are planned and currently the first two epochs (VLASS1 & VLASS2) have been fully observed and processed.We use the latest release of VLASS which contains Epochs 1 and 2 'Quick Look' (QL) Catalogues1 , each of them containing around 3 million sources.We note that currently the VLASS data releases do not include Stokes V information. Gaia Catalogue of Nearby Stars To identify radio-bright stellar systems, we crossmatch the aforementioned radio catalogues with Gaia, the largest available star catalogue with detailed astrometric information.Previously, Callingham et al. (2021b) examined the LoTSS Stokes V maps for ≥ 4σ sources and crossmatched those detections to sources in the Gaia Data Release 2 (DR2) to search for radio counterparts to stellar systems.However, the Gaia DR2 catalogue becomes significantly incomplete for spectral types later than ∼M7 (Kiman et al. 2019).Therefore, in this work, we instead use the Gaia Catalogue of Nearby Stars (GCNS; Gaia Collaboration et al. 2021b).The GCNS is a catalogue of well-characterised objects within 100 pc of the Sun from the Gaia Early Data Release 3 (EDR3; Gaia Collaboration et al. 2021a).The GCNS contains 331312 objects, 40234 of which are within 50 pc.The fact that the catalogue is volume-complete for all objects earlier than M8 at the nominal G = 20.7 magnitude limit2 of Gaia (Gaia Collaboration et al. 2021b) is essential as our aim is to see whether radio detection rates of stellar systems evolves with spectral type across the stellar main sequence. Flare statistics As we aim to understand the correlation between the radio emission of stars and stellar activity, statistics of stellar flares occurrence are also needed in order to analyse if a higher flaring activity implies a higher chance of radio detection.The motivation for searching such a correlation is that flares may be the essential in accelerating the radio-emitting electrons.To do this, we use the study by Günther et al. (2020) for optical stellar flares from the stars observed by NASA's Transiting Exoplanet Survey Satellite (TESS; Ricker et al. 2015), and by Johnstone et al. (2021) for Xray stellar flares statistics using the NEXXUS database (Schmitt & Liefke 2004). TESS flares TESS is an all-sky survey mission designed to discover exoplanets orbiting bright nearby main-sequence stars using transit photometry.As TESS has observed numerous stars (continuously for days, some even months) and measured their light curve in order to find exoplanet transits, TESS data also contains valuable information on the incidence of stellar flares.Günther et al. (2020) performed a study of optical stellar flares for the 24809 stars observed with 2-minute cadence during the first two months of the TESS mission.Most importantly, they studied the flare rate and energy as a function of stellar type and rotation period. As seen in Figs. 4 and 5 reported by Günther et al. (2020), flares are most commonly detected on M dwarfs, especially on mid to late spectral type (beyond M4), where more than 40% of the stars showing observable flares lie.Stars of spectral type earlier than M-type have significantly lower observable flare occurrences of less than 10%.This general picture is also consistent with past findings from e.g.Kepler (Davenport 2016;Van Doorsselaere et al. 2017) and MEarth (Mondrik et al. 2019) cat-alogues of stellar flares: stars of later spectral types (especially fast-rotating, young M dwarfs) are the most likely to flare, and that their flare amplitude is independent of the rotation period (cf.Maehara et al. 2012). Although Günther et al. (2020) only present findings derived from the first two months (i.e.Sectors 1 and 2) of TESS data, here we use the data from the first 2 years (Sectors 1-26) of TESS mission instead (Günther et al., in prep.).Such a database increases the number of flaring stars detected by a factor of ≈ 13, greatly improving the flare statistics.The data for the TESS short-cadence targets from Sectors 1-26 is publicly available3 . In addition, for our analysis in Sect.4.4.1, we now also account for the average TESS flare rate ν TESS (i.e.number of stellar flares per unit time) in each spectral-type bin, instead of merely considering the fraction of flaring stars provided by Günther et al. (2020).The motivation is to compare optical flare rate to the radio detectability as a function of spectral type.If stellar activity has direct correlation to radio detectability, then it would be expected that the more frequently a star flares, the more likely it is detectable in radio within any given observation window.We describe the definition of a flaring star and the method of how TESS flares are counted in Sect.3.2. The additional data obtained from year 1 and 2 further reinforce the original conclusion by Günther et al. (2020), where they found M dwarfs of type M4-M6 dominate the TESS sample of flaring stars than any other star (Günther et al., in prep.).Note that, however, Günther et al. (2020) pointed out the several biases in their TESS flare study as a consequence of TESS only being able to observe in relative flux units.And so, a common flare energy threshold for all stars is thus necessary for an unbiased comparison of flare rates between different spectral types of stars.The several TESS flare biases and subsequently our method for debiasing the TESS flare statistics are described in Sect.3.3. X-ray flares Besides optical stellar flares like the ones observed by TESS, Xray flares are also indicative of stellar activity.The X-ray luminosity of a star is empirically determined by its mass, age, and rotation rate (Pizzolato et al. 2003).This empirical correlation implies stellar activity evolution is linked to its rotational evolution, since many activity parameters, such as magnetic dipole field strength and mass loss rate, are tightly correlated with the Rossby number R o = P rot /τ c , where P rot is the rotation period and τ c is the convective turnover time (Wright et al. 2011). Therefore, we also utilise the model by Johnstone et al. (2021) for X-ray (0.1-2.4 keV) flare rates of F, G, K, and M dwarfs, with masses between 0.1 and 1.2 M ⊙ .As seen in Fig. 19 in their work, the rates of X-ray flares above a fixed energy threshold monotonically decline with declining stellar effective surface temperature.Such a conclusion is the exact opposite of that from TESS flare study by Günther et al. (2020), in which lower mass stars seemingly flare more than higher mass ones.One reason behind such discrepancy may stem from the different definition of flares in these two studies.Johnstone et al. (2021) only includes flares of energy above an energy threshold of 10 32 erg in their analysis, unlike the TESS flare statistics where all impulsive changes in relative flux are classified as flares based on an inference framework called allesfitter (Günther & Daylan 2021) and complementary criteria such as high signalto-noise ratio (SNR).Therefore, although debiasing is required to compare TESS stellar flare rates of different spectral types (more details in Sect.3.3), there is no such need for any debiasing in the X-ray flare statistics owing to the fact that Johnstone et al. (2021) only counts flares above an energy threshold. Crossmatching method Now that we have introduced the different all-sky surveys used in this study, we outline the method we applied for crossmatching these catalogues. False alarm rate As mentioned in Sects. 1 and 2.1.1,one key challenge of using the Stokes I catalogue with Gaia alone to identify radio stellar systems is that the radio catalogue is composed mostly of galaxies.With a high density of radio sources (≈ 780 sources per square degree on average), the LoTSS sky is far denser than any previous wide-area radio survey such as the NRAO VLA Sky Survey (NVSS; Condon et al. 1998), FIRST (Becker et al. 1995), TIFR GMRT Sky Survey first alternative data release (TGSS ADR1; Intema et al. 2017), and GaLactic and Extragalactic Allsky MWA (GLEAM; Hurley-Walker et al. 2017) survey.Therefore, despite the LoTSS's high astrometric precision of 0.2 ′′ to 0.5 ′′ (Shimwell et al. 2022), true associations between dense optical surveys and radio sources cannot be confidently done by simple blind crossmatching; Callingham et al. (2019) showed that a blind search for radio-bright stellar systems in Gaia and LoTSS is dominated by false positives, and thus either additional observational or physically motivated information is needed in order to form a reliable sample of Galactic Gaia-LoTSS counterparts.For example, assuming the radio sources in the sky surveys are homogeneous, the following equation gives us an approximate estimate of false matches with the Gaia catalogue: where N false is the number of chance-coincidence associations, n radio and Ω radio are the source density and sky coverage of the radio survey respectively, and θ is the crossmatching radius.For LoTSS-Gaia DR3 crossmatching, this gives us N false ∼ 10 5 , hence highlighting the difficulty of confidently associating stellar systems with radio emission from LoTSS alone.Similarly, the VLASS catalogue also faces the same issue of false positive as it has a source density only a factor of ≈ 8 lower than that of LoTSS. LoTSS × GCNS match Nevertheless, identifying radio-bright stellar systems in the LoTSS Stokes I catalogue while keeping chance-coincidence associations small is possible.We achieve this by crossmatching with GCNS within 50 pc and by setting a small crossmatching radius of 1.4 ′′ with proper motion correction based on Gaia information.This crossmatching radius corresponds to ≈ 1 chancecoincidence association on average. V-LoTSS × GCNS match On the other hand, as mentioned in Sect.2.1.2,the circularpolarised sky of LoTSS allows us to be significantly more confident in the radio sources' association with stellar systems.Since the number of detected sources in V-LoTSS is less than a hundred, we need not restrict the crossmatching with GCNS to be within 50 pc.Moreover, the low source density also allows us to set the crossmatching radius generously to be 6 ′′ (which corresponds to a false association rate of 10 Average TESS ensemble flare rates We use the catalogue for all individual flares found in TESS year 1 and 2 (Günther et al. 2020 and in prep.).Each entry contains information on a candidate flare from a TESS short-cadencetargeted star, such as the flare's peak time, full width at half maximum (FWHM), and flare amplitude (see Günther et al. 2020). The new catalogue also contains extra information on qualitycontrol filters and a probability of the candidate being a true flare, which is computed by the convolutional neural network stella (Feinstein et al. 2020a,b).This extra information is also detailed in Feinstein et al. (2022).In this work, we follow the definition of TESS flare rate per star ν TESS by Feinstein et al. (2022): where N is the number of flares for a given star, p i is the stella probability for each flare candidate from that star, and t obs is the total observed time of the star by TESS.By binning the flare rates into effective temperature T eff bins, we obtain a relationship between stellar spectral type and the average ν TESS .This curve, when multiplied by the fraction of flaring stars in each T eff bin, gives us the average likelihood of TESS detecting a flare on a star of a particular spectral type (see Fig. 1a).We shall refer to this quantity, i.e.N flaring /N total × ν TESS , as the "average TESS ensemble flare rate".Intuitively, this quantity tells us that if one were to pick a star at random and observe it with TESS, how many flares on average would be seen on that star given its particular spectral type. Debiasing TESS flare statistics As mentioned in Sect.2.3.1, the TESS flare statistics used here are inherently biased since they defined "flaring" as a sufficient increase in flux amplitude compared to its quiescent level (i.e.sufficient increase in relative flux units), rather than a lower energy threshold above which all flares are counted.These TESS flare statistics biases are described in depth in the following sections. TESS sampling bias Firstly, TESS' short-cadence observations are biased towards brighter stars to ensure follow-up spectroscopic observations of transiting exoplanets (Collins et al. 2018).Indeed, TESS is essentially a targeted survey with a complicated target selection (for light curves) determined by the myriad of short-cadence proposals submitted by the scientific community.We can minimise this TESS sampling bias by excluding stars beyond a cer-tain distance such that TESS has observed every star within such volume; around 15 pc is where the TESS sample is volumecomplete (Günther et al. 2020 and in prep.).However, the number of flaring stars with early spectral type within 15 pc is too small to perform proper statistics.In particular, there is almost no star earlier than K0 within 15 pc in the TESS data.Therefore, we compromise on considering TESS stars within 50 pc instead, as this achieves a good balance between sample completeness and a sufficient sample size for proper statistical analysis.As shown in Fig. 1a, the orange curve -representing the average-TESSensemble-flare-rate that excludes TESS stars beyond 50 pc -has a very different shape when compared to the green curve which does not exclude any star.In particular, hotter stars with spectral type earlier than ∼K5 have significant amount of flaring events in the green curve compared to the orange curve.In addition, along with the usual mid-M peak, there is an additional peak around K0 which is absent in the orange curve.However, M dwarfs (especially around M4) still show the highest average TESS ensemble flare rate (N flaring /N total × ν TESS ) in both curves.The deviations between the curves are the manifestation of the TESS sampling bias from which the full TESS flare dataset (i.e.green curve) suffers.For example, Fig. 1a suggests that flaring solar-type stars are oversampled by TESS.One of the possible reason for this oversampling is that the scientific community is interested in studying the superflares of solar-type stars in order to obtain insights on space weather around solar analogs (e.g.Namekata et al. 2017;Maehara et al. 2017;Tu et al. 2020).In any case, we shall only consider TESS stars within 50 pc in subsequent analysis, as such a 50-pc sample achieves a good balance between minimising the TESS sampling bias and having enough flaring stars to avoid small-number statistics. TESS flare detection biases Secondly, there exists an interplay of two opposing flaredetection biases: (i) since TESS detects flares in relative flux units, a flare of a given energy is more readily detected on cooler stars because of a larger contrast between the flare and the stellar quiescent flux density; and (ii) a higher photometric noise in cooler stars decreases the SNR of their relative flare flux densities.As shown in Fig. 2f of Günther et al. (2020), the recovery rates of flares in injection tests is significantly lower for cooler stars in the M-dwarf range compared to F/G/K dwarfs.And so, in order to have a fair flare-rate comparison between stars of different spectral types, these biases must be taken into account.In the following paragraphs, we describe the procedure for the debiasing of the TESS flare statistics.Here, we start by converting the known flare completeness in relative flux units (Günther et al. 2020; in Fig. 2f), to a flare completeness in absolute flux units: Let A S be the relative flux of a stellar flare observed by TESS with respect to the corresponding quiescent stellar flux.We define A min = L ′ f,min /L ′ * as the minimum relative flux of which a stellar flare of luminosity L f,min can still be detectable from a star with an effective temperature T eff and a stellar quiescent luminosity L * , i.e.A min = A min (T eff ), with the prime symbol ′ denoting the luminosity in the TESS observing bandpass.This function encapsulates the interplay of the two aforementioned detection biases of TESS flares.The value of A min (T eff ) is already known from Fig. 2f by Günther et al. (2020).The stellar quiescent luminosity L ′ * can be expressed as where R * is the stellar radius, B * is the Planck function at the T eff of the star, and S TESS is the TESS spectral response function which is defined as the product of the long-pass filter transmission curve and the detector quantum efficiency curve (Ricker et al. 2015).To relate the stellar radius R * to its T eff , we use the Basti-IAC isochrone model (Hidalgo et al. 2018) with a solartype metallicity for stars with T eff ≳ 3000 K and the BHAC15 isochrone model (Baraffe et al. 2015) for main sequence lowmass stars down to T eff = 2600 K.We assume a stellar age of 1 Gyr (around the same order of magnitude of most stars in our galaxy, e.g.Haywood et al. 2013;Snaith et al. 2015) for these models.Fig. 1b shows R * as a function of T eff and the corresponding smooth spline interpolation. We can determine the minimum flare luminosity observable by TESS: Now, to relate this quantity to its unprimed counterpart, we assume that flares have a blackbody spectrum with the same effective temperature no matter the spectral type of the star (e.g.Shibayama et al. 2013;Davenport 2016;Yang et al. 2018;Günther et al. 2020;Gao et al. 2022).The precise value of effective temperature does not matter in subsequent analysis 4 as we only are only interested in the ratio between the flare luminosities from stars of different spectral types, not their absolute values.This assumption implies that for two stars with stellar effective temperatures T 1 and T 2 respectively, we have . Therefore, we now have L f,min as a function of minimum relative flux A min and T eff .Moreover, we assume the average duration of a stellar flare is the same for stars of all spectral type, thus , where E f (T ) is the flare energy from a star with T eff = T .Since the curve of A min vs T eff is already given by Günther et al. (2020), we can determine the minimum flare energy E f,min detectable by TESS on a particular star with T eff = T .As seen in Fig. 1c, the weakest TESS-detected flare on a early-type star is much stronger than its late-type star counterpart.For example, the E f,min of a solar-like G star is around ∼2 orders of magnitude stronger than the E f,min of an M dwarf. Lastly, we need to convert this flare energy threshold into a flare rate ν.To do this, we utilise the study by Gao et al. (2022) regarding cumulative flare frequency distributions (FFDs; e.g.Lacy et al. 1976;Hawley et al. 2014;Günther et al. 2020;Jackman et al. 2021) detected by TESS and Kepler.The cumulative FFD represents the number of flares in unit time with an energy greater than a particular value E, i.e. how often a flare of energy E f ≥ E occurs on a star.Mathematically, where α cum < 0 and β are empirically determined parameters.The debiased flare rate for a star with spectral type S is then log 10 (ν S /ν TESS,S ) = α cum log 10 (E 0 /E f,min,S ), (6) where E 0 is the common energy threshold above which all flares are counted for all stars, ν TESS,S and E f,min,S are the TESSobserved flare rate and the minimum TESS-detected flare energy on star of type S respectively.Gao et al. (2022, in Fig. 12) empirically determined that α cum from solar-type stars to mid M dwarfs is approximately consistent with a value of −1, and so we shall assume α cum = −1 in subsequent calculations.We can now convert the ratio of observed flare rates for stars of types S 1 and S 2 into a ratio of their true flare rates above some energy threshold as log 10 (ν S 1 /ν TESS,S 1 ) − log 10 (ν S 2 /ν TESS,S 2 ) = − log 10 (E 0 /E f,min,S 1 ) + log 10 (E 0 /E f,min,S 2 ).( 7) Hence, Finally, multiplying the two blue curves in Figs.1a and 1c gives us Fig. 1d, which shows the the debiased average TESS ensemble flare rate, i.e. the fraction of flaring stars N flaring /N total weighted by the debiased flare rate ν.Here, given a particular energy threshold above which flares are counted, we can see a general trend of the ensemble flare rate decreasing as we move from early-type stars towards K dwarfs.Then, at T eff ≈ 4800 K (around spectral type K3), the ensemble flare rate increases, peaking at spectral type M0 before decreasing again in the range of mid-M dwarfs.Specifically, early-type (≲G0) stars are actually around 2 − 4 times more likely to flare than early-K and mid-M dwarfs, while the ensemble flare rate around spectral type M0 is quite comparable.This debiased TESS flare trend is not exactly in line with the X-ray flare rate statistics by Johnstone et al. (2021).Regardless, we have now obtained a debiased average TESS ensemble flare rate curve, which shall be used in the TESS flare statistics analysis in Sect.4.4.1. Crossmatching Results In Tables A.1, A.2, and A.3 (all shown in Appendix A), we present the radio × GCNS crossmatching samples from the three aforementioned radio surveys.There are 22 V-LoTSS-detected sources that have optical counterparts in the Gaia Catalogue for Nearby Star (GCNS), while there are 25 LoTSS-detected sources and 65 VLASS-detected sources that have an optical counterpart in GCNS within 50 pc.Figures 2 and 3 show each of three radiodetected stellar populations in the Hertzsprung-Russell (HR) diagram. For the V-LoTSS × GCNS sample, note that out of the 22 matches, all except two have an angular separation between the V-LoTSS source (based on Stokes I astrometry) and Gaia counterpart of ≲ 1 ′′ .The remaining two sources, HAT 182-00605 and II Pegasi, have an angular separation of 1.22 ′′ and 1.83 ′′ respectively.However, since both of them are already known to be radio-bright (Callingham et al. 2021b;Toet et al. 2021), we are confident that none of the matches in our V-LoTSS × GCNS sample are false positives.Moreover, all of these matches are consistent with the stellar sample in V-LoTSS, as Callingham et al. (2023) also crossmatched V-LoTSS sources to the Gaia Data Release 2 & 3 (DR2 & DR3; Gaia Collaboration et al. 2018Collaboration et al. , 2022) ) catalogues.However, there are two stellar systems in V-LoTSS that are not in our radio sample despite being within 100 pc: DG CVn and i Boötis.They are not present in the GCNS likely due to unreliable astrometry and thus their Gaia DR2 source identifier are no longer available in Gaia DR3. Therefore, we choose to ignore these stellar systems in our analysis. Not every source in the V-LoTSS × GCNS sample makes an appearance in the LoTSS × GCNS sample, and vice versa.The latter is expected since some radio-bright stellar systems may have fractional polarisation values that makes them drop below the detection threshold of V-LoTSS.In addition, due to time variability of stellar radio sources, a genuine V-LoTSS source that appears in a particular LoTSS field may get "washed out" during mosaicing of the fields in LoTSS-DR2 (Callingham et al. 2023).The mosaicing also explains why even when a stellar system shows up in both the Stokes I and Stokes V catalogues, the quoted Stokes I flux is slightly different in each catalogue. As for the VLASS × GCNS sample, despite the fact that VLASS and LoTSS have very similar sensitivity and that the VLASS sky fully overlaps with that of LoTSS, very few LoTSS/V-LoTSS detections can be actually found in the VLASS × GCNS sample, and vice versa.The reasons for this discrepancy include the transient nature of stellar radio emission and the vastly different frequencies of the two radio surveys (see Sect. 2.1.3on the importance of survey frequency coverage).Therefore, there is no guarantee that a stellar system that emit in LOFAR band must also emit in VLA band, and vice versa. There is one source that we remove from the VLASS sample, despite a crossmatching association with GCNS.The source is a M2 dwarf named HD 9770C (Gaia DR3 5022972468944971648), which is part of the visual triple system HD 9770 (also known as BB Scl).Only the C component of this system is in the GCNS, as both the primary A star and secondary B star are not present in the GCNS due to incomplete astrometry in Gaia EDR3 and DR3 (e.g.missing proper motion value).Moreover, the Gaia colour and Gaia absolute magnitude of HD 9770C makes it deviate from the HR main sequence by > 4 mag.All of these peculiar Gaia properties are probably caused by the angular proximity of the three stars: the two stars A and B are in a well-defined 4.559-year orbit with a semi-major axis of 0.171 ′′ , and the AB × C system is in a 111.8-year orbit with a semi-major axis of 1.419 ′′ (Hirshfield & Sinnott 1985).Moreover, both A and B are themselves binaries, with B being an eclipsing binary of the BY Dra type (Watson et al. 2001).Using Gaia DR2 astrometry, we found that the radio emission is actually closer to the AB system than the C component (around 0.3 ′′ vs 1.4 ′′ ).Thus, we conclude that the radio emission most likely stems from the BY Dra variable from the secondary B component of HD 9770, and we remove HD 9770C from our analysis. As shown in Figs. 2 and 3, both the LoTSS (V-LoTSS included) and VLASS populations can be generally classified into two categories: chromospherically active stellar (CAS) systems and M dwarfs.The CAS systems can be further subdivided into RS Canum Venaticorum (RS CVn) variables and BY Draconis (BY Dra) variables.These are close stellar binaries5 that consist of late spectral types (F to M). Indicated by the presence of strong Ca II H and K emission lines, the stars in these systems have active chromospheres due to strong magnetic field generated by rapid stellar rotation (P rot ∼ days) twisting magnetic flux loops, as the binaries are generally in very close proximity (≲ 0.01 AU) and thus tidally locked.This causes extreme degrees of solar-type activity in these systems, manifested in the form of large stellar spots and magnetic interactions.In the case of the latter, they coherently accelerate electrons (Toet et al.On the other hand, M dwarfs compose the majority of the radio detections, with their spectral type ranging from M1.5 6 to M6 for the LOFAR sample, and M1.5 to M8.5 for the VLA sample.All of these M dwarfs in our V-LoTSS sample were already presented in previous publication (Callingham et al. 2021b), where it was shown that coherent radio emission is ubiquitous across the M-dwarf main sequence, as each spectral type has an equal probability of detection.This implies V-LoTSS has detected both partially convective M dwarfs and fully convective ones.The same applies to both LoTSS and VLASS as well.In particular, VLASS dives into the realm of ultracool dwarfs (UCDs) as the survey detects a few late-M dwarfs as well.This includes the M8.5V ultracool dwarf LSR J1835+3259, which is recently discovered to have a Jupiter-like radiation belt (Climent et al. 2023;Kao et al. 2023). The underlying reason why M dwarfs are more radio-bright than other isolated stars could be that M dwarfs generally exhibit not just Sun-like activity, but also Jupiter-like properties such as fast rotation, strong large-scale magnetic fields, and auroral radio emission (e.g.Hallinan et al. 2015;Callingham et al. 2021a).Therefore, unlike earlier spectral types, their radio emission could be predicated on the existence of their strong largescale stellar magnetic fields.The M dwarfs should then have a higher efficiency of converting the available energy to radio emission compared to early-type stars, thus making them easier to detect in an untargeted survey. The only LoTSS/V-LoTSS × GCNS objects that do not fall into these two categories are the prototypical α 2 Canum Venaticorum (α 2 CVn) and HD 220242, the latter of which is an isolated F5 star situated 69 pc away.This makes HD 220242 by far the furthest stellar radio detection from our sample.It is the only main-sequence star other than an M dwarf without known companion stars and is detected with high circular polarisation (78 ± 16).Generally, one does not expect a main-sequence Ftype star to harbour a large scale field strong enough to generate cyclotron maser emission at 144 MHz.As for solar-type bursts powered by plasma emission, the brightest solar burst ever recorded would only be detectable up to ≈10 pc (Vedantham 2020).Therefore, the strong circularly polarised radio emission from such a faraway F-type star is unexpected and more details about this peculiar star are discussed in Appendix B.1. However, unlike the LoTSS/V-LoTSS × GCNS population, where we mostly detect either CAS systems or M dwarfs, we see a larger variety of stellar systems in the VLASS detections (see Fig. 3).Granted, M dwarfs are still the majority of the detections, along with the chromospherically active stars such as RS CVn and BY Dra systems.However, the VLASS population also consists some unique stellar systems, including a few UCDs, one T Tauri star (TTS), the prototypical α 2 CVn, one chemically peculiar Am star, and a few K-type main-sequence stars which are probably chromospherically active.We describe more details regarding some of the more interesting stellar systems in Appendix B. One peculiarity regarding the V-LoTSS × GCNS sample is the small amount of V-LoTSS stellar detections outside of the 50-parsec sphere; only 2 out of the 22 sources are beyond 50 pc: EV Draconis which is a RS CVn system and the isolated F star HD 220242.See Appendix C for more details on the lack of detections and on the analysis of the V-LoTSS incomplete- ness.In short, there may be be some incompleteness in the V-LoTSS M-dwarfs population, but the statistical evidence for this is marginal. Pre-main-sequence stars in the VLASS × GCNS sample Another peculiarity regarding the VLASS × GCNS sample is the large number of stellar systems that deviate from the main sequence.Shown in Fig. 3, some stellar systems such as RS CVn variables, BY Dra variables, and M dwarfs (including a T Tauri star) are away from the general GCNS population by a few magnitudes of brightness.For the RS CVn systems, this is not surprising since these are all binary systems and some RS CVn systems even host subgiants or giants as one of the components.These make them much more luminous than their main-sequence non-binary counterparts, thus branching off the main sequence.Yet, the fact that so many M dwarfs also exhibit the same phenomenon is unexpected, considering that nothing of such can be seen in the LOFAR sample.A naive explanation could be that these M dwarf systems are unresolved binaries with another star of very similar spectral type, causing the observed luminosity to be up to twice of a typical isolated M dwarf.This alone, however, cannot explain more than half of the M dwarfs with a deviation from the main sequence by more than 0.753 mag, which corresponds to an unresolved binary of two identical stars (see Fig. 4).Since these are M dwarfs, they also cannot be in an evolved state as a typical M-dwarf lifespan far exceeds the current age of the Universe (Laughlin et al. 1997). An alternative possibility is that these are all young active stars similar to a T Tauri star, and belong to known kinematic groups such as stellar associations, stellar nurseries, and young moving groups (YMGs).Ultimately, we found that around 80% of the stars above the red line shown in Fig. 4 are indeed known members of YMGs and stellar associations, including Beta Pictoris (β Pic) Moving Group (e.g.Shkolnik et al. 2017), Octans-Near Association (e.g.Zuckerman et al. 2013), AB Doradus (AB Dor) moving group (e.g.Malo et al. 2013), and TW Hydrae association (TWA; e.g.Neuhäuser et al. 2010).These stars are all very young (<100 Myr) and are considered to be in the stage when they have not yet reached the main sequence, i.e. pre-main-sequence stars.Therefore, these young stellar objects (YSOs) most likely have extremely fast rotation as they have yet to experience significant rotational spin-down through stellar winds, making them even more magnetically and chromospherically active than the typical M dwarfs due to their convective envelope enabling dynamo processes (Bouvier et al. 2014).This leads to highly variable yet consistent radio emission from YSOs via mechanisms such as gyrosychrotron and ECMI. The remaining question is why such a YSO population seen in the VLA sample is nowhere to be found in the LOFAR samples.The most likely explanation is that most of the nearby young moving groups (AB Dor, TWA, β Pic, etc.) are located in the Southern Sky and/or the galactic plane, both of which lie outside of the current LoTSS footprint.Another possible explanation is that the predominant radio emission mechanism for these YSOs have most of its power emitted in the decimetre-wave (VLA band) regime rather than metre-wave (LOFAR band).This could happen in the case where most of their radio emissions are gyrosynchrotron emission stemming from flares (Nindos 2020). Radio evolution with spectral type As mentioned in Sects.3.1.3and 3.1.4,the vast majority of isolated stars detected in the radio surveys are M dwarfs, so one might be tempted to claim that there is a radio evolution with spectral type and the radio-engine (Sun-like vs Jupiter-like) transition occurs in the realm of M dwarfs.However, M dwarfs are also the most numerous stellar type, and so a proper statistical test is necessary to ascertain this apparent evolution of radio detectability with spectral type.As a statistical test, we compute the cumulative distribution functions (CDFs) with respect to Gaia colours for the 4 populations: the GCNS "background" population, the LoTSS × GCNS population, the V-LoTSS × GCNS population, and the VLASS × GCNS population.We only consider sources within 50 pc of our Solar System in the following analyses since spectral types later than M9 start to become incomplete in GCNS beyond 50 pc (Gaia Collaboration et al. 2021b).The number of sources in GCNS thus decreases to ≈ 40k, and the two sources beyond 50 pc from V-LoTSS × GCNS are removed.The LoTSS × GCNS and VLASS × GCNS samples remain unchanged as we only crossmatch VLASS catalogue with GCNS within 50 pc. There are many statistical tests widely used for quantifying the discrepancy between two CDFs.Here, we choose the Cramér-von Mises test (CvM test; Anderson 1962) to test the null hypothesis: the two samples being compared are drawn from the same distribution.The CvM test computes a p-value in a two-sample statistical test, which represents how likely the null hypothesis is true.For our analysis, a small p-value signifies a deviation between the radio × GCNS population and the background population, suggesting a transition of radio detection rate with spectral type.We choose the classic threshold of p-value < 0.05 as the condition for a rejection of the null hypothesis. In the following section, we consider the cases for both the inclusion and exclusion of CAS systems in our CDF analyses. GCNS background vs radio-detected samples In Fig. 5, we show the CDFs of the three radio-bright stellar populations (V-LoTSS, LoTSS, and VLASS) compared to the GCNS background CDF, both with and without inclusion of the chromospherically active stars (i.e.RS CVn and BY Dra).The justification for removing the CAS systems is that they likely have a different mechanism powering the radio emission as opposed to isolated stars.We note that with this removal, the GNCS background is practically unchanged due to the rarity of such systems. We see from the top panel of Fig. 5 that the VLASS × GCNS sample is inconsistent (> 3σ) with the GCNS background distribution, specifically between the spectral types G0 and M3.This shows that RS CVn and BY Dra systems are much more luminous in radio wavelengths compared to any other stellar systems of similar spectral type, likely due to magnetic interactions, tidal locking, and high chromospheric activity (Dulk 1985; Slee et al. 2008;Toet et al. 2021).Indeed, despite constituting less than 1% of the stellar population (Eker et al. 2008), these CAS systems are so radio-luminous that they dominate the radio source counts.On the other hand, there is agreement between the GCNS background distribution and the LOFAR-detected population within uncertainties.However, since LoTSS and V-LoTSS both detect a significant amount of CAS systems despite their small number (≲200 in our solar neighbourhood), this agreement should only be treated as a mere coincidence due to large uncertainties from the small LOFAR-detected sample size.We conclude that in terms of stellar systems, radio surveys are much more likely to detect these CAS systems than any other stellar systems. The bottom panel of Fig. 5 shows the radio populations after removal of the known CAS systems.Now, the conspicuous "G0−M3 bump" in the VLASS × GCNS sample is absent, leaving instead a sudden rise in radio detections around the spectral type M3 to M5.The same rise can be seen in the LOFAR sample, albeit with larger uncertainty.Visually, this region is where the VLASS population deviates from the background distribution the most.The CvM test agrees with the discrepancy as well; the p-value for VLASS vs GCNS is 0.0039, whereas the p-values 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 Fig. 6: Zoomed-in version of the bottom panel of Fig. 5, with V-LoTSS × GCNS sample omitted for the sake of clarity. for V-LoTSS vs GCNS and for LoTSS vs GCNS are 0.039 and 0.198 respectively. M4 transition in stellar radio engine The low p-values demonstrate an evolution in radio properties with spectral type.Specifically, Fig. 6 shows where the largest discrepancy between the VLASS (grey line) and background population (blue line) is located.As one can see, while the error bar for the LoTSS population (orange region) is too large to make any definite statements, the grey line deviates from the blue line the most around spectral type M3-M4.This is approximately the spectral type at which most theoretical stellar evolution models predict the transition from partially convective regime (i.e. with a radiative-convective interface akin to the solar tachocline) into the fully convective regime (e.g.Dorman et al. 1989;Clemens et al. 1998;Ribas 2006;Morales et al. 2009).This so-called "M4 transition" (see Stassun et al. 2011) is also supported by observational evidence from the properties of M dwarfs -stellar parameters, activity lifetime, rotation, magnetic field strength, and topology (e.g.West et al. 2008;Donati et al. 2008;Morin et al. 2008).As such, the radio evolution with spectral type may also be a direct cause of such a regime transition.This motivates us to conjecture that the M4 transition also leads to a gradual shift from a Sun-like radio engine to a Jupiter-like radio engine.Originally, we also expect the LOFAR-detected and VLAdetected populations would be inconsistent with each other, with reasons previously mentioned in Sect.2.1.3.However, in both cases where CAS systems are included/excluded, we cannot claim any significant deviations between the radio samples as they all overlap each other within 95% confidence interval. Comparison to multiwavelength flare rates One natural question that arises from the detected radio evolution with spectral type is whether the evolution follows other known stellar activity indicators.This leads us to the investigation on the impact of stellar flare occurrence on radio detectability. GCNS × TESS flare CDF If flares are ultimately the prime source of radio energy, and if the ratio between the flare energy that goes into the optical band and that into the radio band does not change with spectral type, then we expect the radio detections and the TESS flare detection to have the same CDF shapes.By comparing the two CDF curves, we can therefore determine if stars of some spectral type are more efficient than others at generating radio emission from flaring events. To compare the CDFs of the optical flare rate and radio detectability, we start with the debiased average TESS flare rates versus T eff from Sect.3.3.2.The CDF of TESS optical flares as a function of spectral type is proportional to the multiplication of two CDF curves: the debiased average-TESS-ensemble-flarerate CDF curve (recall Sects.3.2 and 3.3), and the CDF of the GCNS background distribution. To cast every CDF curve as a function of T eff , we determine the values of T eff for each star in our sample from the TESS Input Catalog version 8 (TICv8; Stassun et al. 2018) catalogue if available, or from Gaia DR3 catalogue otherwise.If neither catalogue provides a valid T eff , then we estimate T eff from the Gaia colour G BP − G RP using a polynomial fit (from numpy.polyfit;deg = 10) between colour and temperature to the data from Pecaut & Mamajek (2013) 7 . The result is shown in Fig. 7.Note that here we did not plot the CDF without CAS systems (RS CVn and BY Dra variables) since the TESS targeting strategy is not expected to discriminate against close binaries or active flaring stars, thus there is no compelling reason to exclude them in our analysis.As mentioned 2800 3000 3200 3400 3600 3800 Fig. 8: Same as Fig. 7, but with only M dwarfs considered.Stars before spectral type M0 are excluded from this specific analysis. during the discussion of Fig. 1d in Sect.3.3.2,early-type stars (earlier than ∼G0) and late-K/early-M dwarfs are much more likely to flare compared to early-to mid-K dwarfs and mid-M dwarfs.Conversely, late-type stars (later than ∼M3) are much more numerous in the Gaia stellar population.Hence, the interplay of these two factors yields a CDF (solid blue curve) not too dissimilar to the CDF of the GCNS population (dashed blue curve), as illustrated in Fig. 7. Here, the GCNS × TESS flare CDF curve (solid blue line) is visually consistent with all three radio-detected populations at the > 95% confidence level for T eff ≳ 3500 K, i.e. most of the solid blue line is within the shaded regions of the radio × GCNS samples for T eff ≳ 3500 K.More interestingly, it is only within the realm of M dwarfs (later than spectral type ∼M3 in particular) where the GCNS × TESS flare curve CDF starts to significant deviate from all three radio-detected populations.This inconsistency is confirmed by the CvM test as well; the p-values for GCNS × TESS flare vs V-LoTSS, LoTSS, and VLASS are 0.00770, 0.00398, and 0.00013 respectively. To confirm that the deviation is indeed the most significant in the M-dwarf population seen in Fig. 6, we also compare the CDFs in the T eff range of 3850K to 2800K only, as illustrated in Fig. 8.Such a T eff range corresponds to M0-M6 dwarfs.Now, we can clearly see that the discrepancy between GCNS × TESS flare curve and the radio-detection population is the largest around spectral type M4-M5.Here, the CvM-test p-values are 0.00300, 0.00252, and 3.66 × 10 −9 , for V-LoTSS, LoTSS, and VLASS respectively.These p-values are all much smaller than their counterparts that consider the entire T eff range. Hence, we can confidently conclude that all three radiodetected population do not fully follow the TESS optical flare statistics.Specifically, the radio samples are consistent with TESS flare CDF for stars with T ≳ 3500 K, but significantly inconsistent for T ≲ 3500 K.This is somewhat unexpected, because one would expect that flares occur due to a universal underlying process which leads to a consistent fraction of energy channelled into radio-emitting accelerated charges.Our results instead suggest that the radio efficiency of optical flares -the fraction of flare energy channelled into radio-emitting chargesevolves with spectral type.In particular, Figs.7 and 8 suggest that the radio efficiency of optical flares remains roughly constant throughout the stellar population until the M dwarfs, where the M4 transition (mentioned in Sect.4.3.2) seems to manifest itself once again, this time in the form of an increase in radio efficiency of optical flares.We hypothesise that this evolution is related to the evolution of magnetic field properties with spectral type owing to the transition from a Sun-like engine to a Jupiterlike engine. We are aware that, despite our best debiasing efforts described in Sect.3.3, the TESS target selection is still inherently biased since the TESS 50-pc sample used in our analysis is not volume-complete, and thus the degree of biases depends on the topics of interest in the scientific community back during TESS year 1 and 2. To the best of our knowledge, however, a full analysis on the TESS target selection biases does not exist.Therefore, we assume that TESS target selection bias no longer plays a significant role when we only consider the TESS 50-pc sample. GCNS × X-ray flare CDF Similarly, we follow the X-ray flare rate trend presented by Johnstone et al. (2021) to construct a correlated CDF from the GCNS background distribution and X-ray flare statistics.Specifically, Johnstone et al. (2021, in bottom-most panel of Fig. 19) showed the evolution of flare rate with total emitted X-ray energies above 10 32 erg vs stellar mass for stars older than 5 billion years old.This gives us the linear relation: N(> 10 32 erg) = 1.11M − 0.12, where N is the number of stellar flares per day and M is the stellar mass in units of solar mass.From this relation, we simply construct the CDF of GCNS background distribution × X-ray flare statistics with very similar procedure from Sect.2.3.1 for the GCNS × TESS flare, but in stellar mass rather than effective temperature.In fact, the procedure is even simpler as there is no need to debias the X-ray flare statistics as previously stated in Sect.2.3.2. We present the result in Fig. 9.As expected, the GCNS × Xray flare CDF has quite a different shape compared to the GCNS × TESS flare CDF, i.e. the X-ray flare CDF follows a more gently rising trend not seen in the TESS flare CDF.One reason for such a discrepancy could be of astrophysical origin.Audard et al. (2000) found that the rate of X-ray flares is almost linearly proportional to X-ray luminosity among stars of all spectral types.Therefore, solar-mass stars appear to flare in X-ray more frequently than low-mass stars as the former are significantly more X-ray luminous on average.On the other hand, stellar activity is also strongly correlated to rotation period.M dwarfs generally rotate much faster than early-type stars since the latter experience more angular momentum loss owing to stronger stellar winds, causing a spin down of the star.This may imply that there is no universal flare energy partition between optical and X-ray bands, thus leading to disparate observed flare rates between the two wavelengths. Back to the comparison of X-ray flare CDF and radiopopulation CDF: Fig. 9 shows that early-type stars get significantly boosted in the CDF since early-type stars have higher rate of X-ray flares than the low-mass stars.We can see that visually the GCNS × X-ray flare sample is completely inconsistent with all three radio-detected populations.Using the CvM test, we determine the p-values for GCNS × X-ray flare vs V-LoTSS, LoTSS, and VLASS are 2.72 × 10 −4 , 1.07 × 10 −4 and 8.03 × 10 −8 respectively.Therefore, we can conclude that the radio efficiency of X-ray flares -the fraction of X-ray flare energy channelled into radio-emitting charges -evolves with spectral type as well, similar to the conclusion drawn from the TESS flare CDF analysis in Sect.4.4.1. Güdel-Benz relationship We also investigate if our radio-detected samples adhere to the empirical quasi-linear relationship which correlates quiescent gyrosynchrotron 5-GHz radio spectral luminosity L ν,rad and soft X-ray luminosity L X .This is known as the canonical Güdel-Benz relationship (GBR; Güdel & Benz 1993;Benz & Güdel 1994), and the L X ∝ L 0.73 ν,rad law holds for over 10 orders of magnitude in L ν,rad , from the energetic RS CVn systems all the way down to nanoflares from the Sun.Despite their vastly different emission mechanisms, where thermal Bremsstrahlung in the coronal plasma is responsible for the X-ray emission, and incoherent gyrosynchrotron emission from relativistic electrons gyrating in the coronal magnetic field for the radio emission, the presence of this relationship implies there must exist a mechanism that deposits a consistent fraction of flare energy into accelerating charges that emit in the radio and X-ray bands. However, there are also stellar systems that violate GBR.For example, Hallinan et al. (2008) found that the 5-GHz mildly circularly polarised quasi-quiescent radio emission of UCDs, including brown dwarfs, do not obey GBR.They argued that the radio emission for these systems stems from ECMI instead of gyrosynchrotron, making them radio-overluminous (with respect to GBR) as ECMI is more efficient and capable of creating luminous radio bursts (e.g.Yu et al. 2011).Alternatively, it could be that the energetics of these systems or their magnetospheric structure do not support a stable thermal corona to form, which leads to their X-ray underluminosity (Berger et al. 2010;Williams et al. 2014).Callingham et al. (2021b) found that the highly polarised radio emission at 144 MHz from M dwarfs also violates GBR; these objects are very unlikely to be operating in an incoherent mechanism that is gyrosynchrotron due to their high degree of circular polarisation, and thus a coherent emission mechanism must be responsible for 144-MHz radio emission of these objects. And so, if one were to accept the explanation that the 5-GHz radio emission from the stellar systems that established the original GBR stems from gyrosynchrotron in the coronal magnetic field, one must expect our V-LoTSS sample to depart from the GBR.Conversely, the VLASS sample should obey GBR, except for stars of very late spectral type -into the realm of UCDswhere their corona starts disappearing. GBR vs radio-detected samples Here, we plot the soft X-ray luminosity L X against radio spectral luminosity L ν,rad of the LOFAR-detected and VLA-detected populations, as shown in Figs. 10 and 11 respectively, to see which stellar system obeys GBR and which deviates from it.As previously stated, none of the V-LoTSS sample should obey GBR as these stellar systems operate in coherent mechanisms rather than gyrosynchrotron processes.As expected, such is the case for most M dwarfs and the F-type star; most of these stellar systems deviate from the relationship by a few orders of magnitude.However, the chromospherically active stellar systems in our V-LoTSS sample surprisingly adhere to GBR.This peculiar phenomenon was already discussed by Vedantham et al. (2022), where they report that highly polarised 144-MHz radio emission from RS CVn/BY Dra systems and other high-activity stars unexpectedly follows the GBR.In the other two populations (LoTSS and VLASS), CAS systems also consistently obey GBR, though this is less surprising as previously mentioned. As for the Güdel-Benz (GB) deviators, we can see from Figs. 10 and 11 that virtually every single M dwarf in all the radio-survey population are consistently radio-bright/X-ray-dim, and most of them by a few orders of magnitude as well.In particular, the M4 transition also seems to manifests itself in the GBR plot, as M dwarfs beyond ∼M4 significantly deviate from the rest of the population on average.Intriguingly, there exist two stellar sources (LSR J1835+3259 and HD 43162C) that deviate from GBR by more than 5 orders of magnitude, and both of them are later than M4 as well.In Fig. 10, there is a clear gap between the early-M-dwarf and the late-M-dwarf populations, especially in V-LoTSS.The deviation from the GBR with respect to spectral type is better illustrated in Fig 12 .Here, we can see that the radio detection becomes significantly more radio-overluminous/Xray-underluminous than the values predicted by GBR in the region of M4-M6.This gap might also signify a bimodality in the dynamo of the M-dwarf population (e.g.McLean et al. 2012;Cook et al. 2014;Williams et al. 2014), supported by geodynamo simulations and Zeeman-Doppler imaging (ZDI) analysis (Morin et al. 2010(Morin et al. , 2011)). To confirm the monotonic inverse relationship between the L X residual from GBR and spectral type for each of our radio Mdwarf sample in a statistical manner, we calculate the Kendall's τ (Kendall 1938), which is a measure of the relationship between two variables -L X / GB fit and G BP −G RP in our case.This allows us to perform a τ hypothesis test on each of our radio-detected samples, where the alternative hypothesis is that the rank correlation is less than zero, i.e. a negative correlation between L X / GB fit and G BP − G RP .To be succinct, we obtain a p-value much smaller than 0.05 and a negative τ coefficient (around −0.3 to −0.4) for all three radio M-dwarf samples.We thus conclude there exists a negative correlation between goodness of GB fit to the radio data and spectral type in M dwarfs from all three radio surveys.(Williams et al. 2014), where L X is the soft (0.1-2.4 keV) X-ray luminosity and L ν,rad is the 144-MHz total-flux radio luminosity for the LOFAR population.The radio sources without a detectable X-ray luminosity are indicated with a downward-pointing arrow and shown as 3σ upper limits.Colours and symbols are as in Fig. 2, with the additional classification of M dwarfs beyond M4 (represented by yellow circles) motivated by the M4 transition mentioned in Sect.4.3.2. Rotation-related activities Even though the M dwarfs are consistently radio-bright or Xray-dim with respect to Güdel-Benz relationship, there seems to be no obvious new scaling law that can be established as these GBR violators generally scatter around the L X against L ν,rad plot in Figs. 10 and 11.Such was the original conclusion from Callingham et al. (2021b); the highly polarised radio-emitting M dwarfs do not establish a clear new relationship even when accounting for chromospheric activity indicators such as Hα luminosity, rotation period, and Rossby number.Therefore, they suggest that the chromospheric activity has limited influence on the detected radio luminosity.In contrast, previous GHz-frequency studies discovered that the fastest rotating M dwarfs have a higher radio surface flux and radio detection fraction (McLean et al. 2012).Here, we attempt to apply similar investigation in our VLA-detected sample. As mentioned in Sect.4.2, the majority of the non-CAS stellar objects are members of young moving groups (YMGs).This already indicates that they are most likely very fast-rotating stars since they have yet to lose significant angular momentum due to stellar wind.The implication, therefore, is that perhaps the fastest rotators do have a higher radio detection fraction in the 3-GHz band, similar to previous findings by McLean et al. (2012).Unfortunately, finding literature values of rotation period P rot for the VLASS-detected M dwarfs proves to be difficult, as the studies of most of these M dwarfs are lacking in general since they are otherwise run-of-the-mill objects of little interest to the scientific community.Additionally, most of them are too distant and/or too active to be traced by detailed M-dwarf surveys such as CARMENES 8 and MEarth (Irwin et al. 2009).However, for the stellar systems which have measured rotation period (e.g.GJ 3237, GJ 4185, V374 Peg, and V1274 Her), they indeed rotate very fast (typically P rot < 1 day).Therefore, although it is currently not possible to make any definite statements, we believe that rotation period plays an important role on the radio detectability of these stellar objects. Conclusion We conducted a search for radio-bright stellar systems using existing radio sky survey catalogues -LoTSS, V-LoTSS, and VLASS -and crossmatched them with the Gaia data along with proper-motion correction.The high sensitivities and resolutions of these radio surveys provided datasets to isolate an untargeted sample of radio stellar systems such that the statistical properties of the population could be determined accurately.We ascertained our stellar radio samples are not dominated by chancecoincidence associations by using the Gaia Catalogue of Nearby Stars.To the best of our knowledge, this is the first untargeted search for ALL stellar radio sources down to sub-mJy level, and thus the first statistical analysis of radio activity trend with spectral type using an unbiased radio-emitting stellar population. We found that both the LoTSS (V-LoTSS included) and VLASS populations can be generally classified into two categories: CAS systems and M dwarfs.In particular, the VLAdetected sample also contains many more peculiar stellar systems such as YSOs, seemingly isolated K dwarfs, and chemically peculiar stars.The reason why these systems are not present in the LOFAR-detected samples is most likely due to the 3 different survey parameters between LoTSS and VLASS.Firstly, the survey sky coverage: VLASS covers much more of the sky than LoTSS-DR2 (33885 deg 2 vs 5634 deg 2 ), thus it is currently easier to find rare stellar objects in VLASS compared to LoTSS.In addition, VLASS also covers the entire galactic plane visible to the VLA (δ > −40 • ), while LoTSS-DR2 focused on imaging regions at high Galactic latitude and thus deliberately avoids the galactic plane, in which most of the nearby young moving groups (AB Dor, TWA, β Pic, etc.) are located.Secondly, the survey frequency coverage: these two surveys are probing very different radio regimes, as LoTSS is more likely to 11: Same as Fig. 10, but with the VLASS × GCNS population instead.Colours and symbols are as in Fig. 3, with the additional classification of M dwarfs beyond M4 (represented by yellow circles) motivated by the M4 transition mentioned in Sect.4.3.2.The one CAS system that significantly deviates from GBR is a BY Dra system called V1274 Her, which is a mid-M-dwarf binary.Note that two M dwarfs from the VLASS sample are not shown in this figure for the sake of clarity, since they have an X-ray luminosity of ≪ 10 26 erg s −1 .These are LSR J1835+3259 and HD 43162C, with spectral type M8.5 and M5 respectively.capture coherent emission compared to VLASS, as mentioned in Sects.2.1.3and 4.2.The predominant radio emission mechanism for these peculiar systems may have most of its power emitted in the decimetre-wave (VLA band) regime rather than metrewave (LOFAR band).This could happen in the case where most of their radio emissions are gyrosynchrotron emission stemming from flares (Nindos 2020).Thirdly, the survey exposure time: each LOFAR pointing in LoTSS is observed for a total of 8 consecutive hours, whereas VLASS creates 3-minute images.Hence, VLASS is much more susceptible to stellar radio sources with high variability. Using CDF visualisation and the CvM statistical test, we demonstrated an evolution in radio properties with spectral type.The discrepancy between the radio-detected and the GCNS CDF curves is the largest for the VLASS × GCNS population, simply due to the fact that the VLASS sample size is larger -more than twice the sample size of LoTSS/V-LoTSS.In particular, we showed that radio-detected populations deviate the most from the GCNS background population around spectral type M3-M4.This is remarkable due to its location at the so-called "M4 transition", where most theoretical stellar evolution models predict the transition from partially convective into the fully convective regime.This transition drastically changes the dynamo of stars that is responsible for the generation of magnetic field.As such, we suggest that the radio evolution with spectral type may be related to the changing nature of the stellar dynamo. We also studied the evolution of radio efficiency with spectral type via comparison to other known stellar activity indicators.In particular, we tested the impact of stellar flare occurrence on radio detectability using the TESS and X-ray flare statistics.The flare statistics in these two wavelengths do not show the same trend: given a particular energy threshold above which flares were counted, we saw the general trend of flare rate decreasing as we moved towards stars of later spectral type in the X-ray flare statistics, and yet the trend seen in TESS flare statistics is much more complicated.Despite that, the radio detectability does not seem to follow neither of these trends.To test if the ratio between the flare energy that goes into the optical/X-ray band and that into the radio band also changes with spectral type, we compared the CDF curves of the radio detection and the TESS/X-ray flare detection.We discovered that the GCNS × X-ray flare curve is clearly inconsistent with all Fig. 12: Residuals from the Güdel-Benz relationship versus Gaia G BP − G RP colour.For the sake of clarity, the V-LoTSS × GCNS sample is omitted, as well as the two M dwarfs from the VLASS sample (grey circle) with residuals ≪ 10 −3 .They are LSR J1835+3259 and HD 43162C, with spectral type M8.5 and M5 respectively.The radio sources without a detectable Xray luminosity are indicated with a downward-pointing arrow and shown as 3σ upper limits.The top axis of the CDF plots indicates the nominal stellar spectral types (Pecaut & Mamajek 2013). three radio-detected populations at the ≫ 95% confidence level, both visually and statistically.The inconsistency in the GCNS × X-ray flare curve is much more subtle: the GCNS × TESS flare curve agrees with the radio populations throughout most of the stellar population until we reached the M-dwarf population.There, the TESS flare CDF significantly deviates from the radio-population CDFs, especially around spectral type M4-M5.Therefore, the radio efficiency of flares in both optical and Xray -the fraction of flare energy channelled into radio-emitting charges -evolves with spectral type, with later M-dwarfs being more radio-efficient than the rest of the stellar population.We hypothesise that such an evolution may be owing to two factors (or a combination thereof): (i) the efficiency with which charges are accelerated and/or the lifetime of accelerated charges may itself increase for stars of later spectral types; and (ii) the emergence of large-scale axisymmetric magnetic field in fully convective stars may assist more efficient radio emission mechanisms such as ECMI and/or provide new magnetospheric acceleration mechanisms in addition to flare-related processes.We also found that nearly every single M dwarf in all the radio-survey population are consistently and significantly more radio-overluminous or X-ray-underluminous with respect to GBR.In particular, the M4 transition is alluded once again, as M dwarfs beyond ∼M4 significantly deviate from the rest of the population on average.This gap might be signifying a bimodality in the dynamo of the M-dwarf population (e.g.McLean et al. 2012;Cook et al. 2014;Williams et al. 2014), supported by geodynamo simulations and ZDI analysis (Morin et al. 2010(Morin et al. , 2011)).We also confirmed a negative correlation between goodness of GB fit to the radio data and spectral type in M dwarfs from all three radio surveys, i.e. late-M dwarfs deviate from GBR more than early M dwarfs.This was done by performing the Kendall's τ test.We propose follow-up ZDI observations on our radio stel-lar population to see if their radio emission is dominated by the large-scale axisymmetric magnetic field.Additional light curve analysis of the M dwarfs mentioned in Sect.4.2 to get their rotation periods is also vital in helping us prove our conjecture of rotation period playing an important role on the radio detectability of the stellar objects in VLA band. The fact that radio-detected stellar population is dominated by the CAS systems and M dwarfs regardless of the radio surveys implies there must exist some similarities between these two classes of stars such that they are consistently radio bright.Therefore, we postulate that all of these radio-bright stellar systems exhibit strong large-scale stellar magnetic field.CAS systems and M dwarfs (especially young ones) can generate magnetic field strength up to kilogauss (e.g.Giampapa et al. 1983;Saar et al. 1986;Petit et al. 2004;Kochukhov et al. 2013for CAS systems, Johns-Krull & Valenti 1996;Reiners et al. 2009;Morin et al. 2010;Afram & Berdyugina 2019;Kochukhov 2021 for M dwarfs, andGregory et al. 2012 for YSOs).If the hypothesis of radio predominantly tracing magnetic field dynamics is true, then the predominant radio emission mechanism is likely to be ECMI.Since electron cyclotron maser emits mostly at the fundamental and the second harmonic of the cyclotron frequency: ν B = 2.8B MHz, where B is the local magnetic field strength of the radio emission region (Melrose & Dulk 1982), the main difference between the stellar population in LoTSS and VLASS would be that they have different magnetic field strength; B ∼ 50 G would peak at the LOFAR band, while B ∼ 1 kG at VLA band. Currently, the LOFAR-detected sample size is severely limited by the incomplete sky coverage of LoTSS/V-LoTSS, as mentioned in Sect 4.2.This makes interpreting the results of statistical test and drawing conclusions from the LOFAR sample more difficult due to small number statistics.By the completion of the LoTSS survey of the Northern Sky, this will effectively increase the current LoTSS/V-LoTSS sample size threefold.Having the new radio stellar detections will most certainly give us definite conclusion on e.g.whether there exists a discrepancy between the LOFAR-detected and VLA-detected stellar population.Moreover, as mentioned in Sect.2.1.2,looking at the Stokes V sky for radio stellar sources is statistically motivated, since the chance-coincidence association is effectively zero in a circularpolarised radio sky survey.Looking at the Stokes V sky is also physically motivated: circular polarisation information may provide a measure of coronal density for stars, and magnetic field strength for brown dwarfs and exoplanets.Therefore, an equivalent of V-LoTSS for VLASS would be very useful for understanding the underlying radio emission mechanism of the VLA radio-bright population. (α For A) is a F-type solar-age slow-rotating subgiant Fuhrmann & Chini (2015) with a long orbital period of 280 ± 11 years (Izmailov 2019).The angular separation between α For A and B is 4.05 ± 0.08 ′′ (Izmailov 2019).We clearly identify α For B as the radio-loud component in this system because the radio source is < 1 ′′ from B and > 4 ′′ from A in both epochs of VLASS (see Figure B.1). Fuhrmann & Chini (2015) discovered that α For B shows significant chromospheric activity with strong Hα and Ca II H&K lines, whereas nothing alike can be seen on α For A. They argued that since α For A is about solar age and its slow-rotating inactive nature is confirmed by spectroscopy, this means α For B cannot be young and thus has reignited its activity, possibly due to some mass transfer or merger events (Fuhrmann & Chini 2015).Finally, Fuhrmann et al. ( 2016) followed up this system and found a close companion (dubbed α For Bb) around the K dwarf (dubbed α For Ba) using radial velocities.They determined α For Bb to be a white dwarf since there is significant barium enhancement in α For Ba compared to the primary A component.Barium is a red-giant nucleosynthesis product (Jorissen et al. 2019), and thus the material must have been deposited on the K dwarf during a mass-transfer episode that occurred back when the now white dwarf was a giant.This mass transfer likely reignited the K dwarf's activity.At a distance of d = 14.1 pc and with a period of 3.75 days (Fuhrmann et al. 2016), this makes the subsystem α For Ba-Bb one of the nearest white dwarf in a close binary. An important question, therefore, is the role of the white dwarf (WD) in generating the radio emission of α For B. Currently, there are only two known radio-bright WD systems, excluding a class of variable stars known as cataclysmic variables (Mason & Gray 2007).The first system discovered is AR Scorpii (AR Sco; Marsh et al. 2016), which is a WD/M-dwarf close binary with a period of 3.56 hours.The emission from AR Sco pulses over a broad range of wavelengths (from radio to X-ray) on a period of 1.97 min, corresponding to the beat frequency between the 1.95-min spin period of the white dwarf and the 3.56-hour orbital period, thus earning the moniker "white dwarf pulsar".Very recently, Pelisoli et al. (2023) discovered another such system: J191213.72-441045.1 (J1912-4410) consists of a WD and an M dwarf in a 4.03-hour orbit, exhibiting similar pulsed emission with a period of 5.30 mins.Despite a large number of proposed models (e.g.Geng et al. 2016;Katz 2017;Garnavich et al. 2019) to understand its peculiar observed behaviour and follow-up observations of this system (e.g.Marcote et al. 2017;Stanway et al. 2018), however, the exact radio emission mechanism of AR Sco remains unknown.If α For B turns out to be a WD-pulsar system, this makes it an extremely unique system since localising the radio emission region within the α For Ba-Bb subsystem and detecting its orbital motion is possible with VLBI telescopes, unlike AR Sco and J1912-4410.This in turn would allow greater understanding of the underlying radio emission mechanism, validate the theoretical models, and show whether this is an unique phenomenon or rather pulsing radio emission is ubiquitous for close binaries that contain a WD. Lastly, α For B being a radio source is not new - Güdel et al. (1995) observed this star with the VLA at 8.4 GHz in two different epochs (1993 and 1995; respective legacy ID: AG394 and AG458).There is a clear detection in each epoch despite strong variability: around 0.6 mJy in 1993 and 0.3 mJy in 1995, which is ∼20σ and ∼10σ respectively.More importantly, the radio emission shows strong circular polarisation (∼ 30%; Güdel et al. 1995) and no significant linear polarisation (< 1%), both of which are very similar to that seen in AR Sco (Stanway et al. 2018).Previously, Güdel et al. (1995) failed to ascertain if the radio emission stemmed from α For A or B. With Gaia astrometry and proper-motion values now available, we reimaged the data and can confidently conclude that the α For B is the radio emitter rather than A.Moreover, the K dwarf also exhibits strong X-ray luminosity (Hünsch et al. 1999;Boller et al. 2016) compared to a typical isolated K dwarf (e.g.Fleming et al. 1994), yet another similarity to AR Sco (Takata et al. 2018). Here, α 2 CVn is present in both LoTSS (with no significant circular polarisation) and VLASS in both epochs, but it was previously detected in other radio frequencies as well (e.g.Helfand et al. 1999;Drake et al. 2006;Leto et al. 2021).Hajduk et al. (2022) already investigated the radio emission of α 2 CVn detected in LoTSS.They modelled its radio spectrum to constrain the distance at which co-rotation of plasma with the largescale stellar magnetosphere breaks down, and confirmed that the spectral behaviour is consistent with gyrosynchrotron mechanism in α 2 CVn.This is also consistent with previous studies by Leto et al. (2021); Owocki et al. (2022); Shultz et al. (2022), where centrifugal breakout is thought to drive magnetic reconnection, which is responsible the acceleration of radio-emitting gyrosynchrotron.However, the centrifugal breakout model has been challenged due to the lack of Hα emission or shell absorption in α 2 CVn (Pfeffer & McSwain 2022). One intriguing point regarding α 2 CVn is that it deviates from GBR significantly in both 144-MHz and 3-GHz band; it has a radio spectral luminosity and X-ray luminosity very similar to a non-early M dwarf.This might be suggesting that its predominant radio emission mechanism may also be ECMI similar to that of UCDs.Hajduk et al. (2022) determined the circular polarisation at 144 MHz to be significantly below 60% from the LoTSS detection, so α 2 CVn may have mildly circularly polarised emission instead.UCDs have been observed to emit mildly circularly polarised quasi-quiescent radiation in the past (Berger 2002;McLean et al. 2012;Williams et al. 2014), which is thought to be ECM emission that appears only mildly polarised because of depolarisation effect during propagation in the magnetosphere (Hallinan et al. 2008).α 2 CVn deviates from GBR in both LoTSS and VLASS, but the deviation is slightly more significant in the VLA band.This may imply that its cyclotron emission could be stronger in the GHz regime.This is not surprisingly given the strong organised magnetic field of α 2 CVn (Kochukhov & Wade 2010).Stokes V information of this VLASS source would be very useful in discerning the underlying mechanism of α 2 CVn's radio emission.The contours are of levels with a increment of 4σ, i.e. 4σ, 8σ, 12σ, etc.The peak flux of the source is around 2.5 mJy for Epoch 1 and 2.7 mJy for Epoch 2. Appendix B.4: TWA 5 A The furthest radio source in our VLASS × GCNS sample (at 49.7 pc) is discovered to be a stellar system called TWA 5 A (= CD-33deg7795), which is one of the five original kinematic members of the TW Hydrae association (TWA; Kastner et al. 1997): a group of very young (∼ 5 − 15 Myr) low-mass stars and substellar objects (e.g.Neuhäuser et al. 2010).The system consists of the primary T Tauri star (TTS) and a brown dwarf companion (TWA 5 B) located approximately 2 ′′ away (Webb et al. 1999;Lowrance et al. 1999).TWA 5 B is of spectral type M8.5 and has an estimated mass between ∼ 15 − 40M J (Neuhäuser et al. 2000).TWA 5 B is not present in the Gaia catalogues.Their very far (> 100 AU) separation precludes any magnetic interactions between them.The VLASS images in both epochs show that the radio source is located much closer to A than to B (see Fig. B.2). Therefore, we conclude the TTS should indeed be the predominant radio emitter. The primary TTS itself is a spectroscopic binary (Webb et al. 1999;Muzerolle et al. 2000;Torres et al. 2000).Macintosh et al. (2001) resolved the primary into a 60-mas binary system (TWA 5 Aab) using adaptive optics.Konopacky et al. (2007) and later Köhler et al. (2013) refined the semi-major axis to be 63.7 ± 0.2 mas.Finally, Köhler et al. (2016) improved the orbital parameters and estimated the mass ratio of Aa and Ab to be M Ab /M Aa = 0.93 ± 0.13, making them very similar in mass.Therefore, these two components -Aa and Ab -are also equally likely to be radio emitters. Intriguingly, there are previous attempts at detecting radio emission from this peculiar system.Osten & Jayawardhana (2006) searched for radio emission from young brown dwarfs in X band, with TWA 5 B being one of them, and did not detect any source in the map at the positions of TWA 5 B in 2005.They also looked at an archival observation made in 2002 and found no detection either.These two non-detections imply a radio luminosity upper limit of this system L ν,rad < 2 − 3 × 10 14 erg s −1 Hz −1 at 8.4 GHz (Osten & Jayawardhana 2006).And yet, TWA 5 A has very significant radio emission (> 10σ) in both epochs of VLASS, giving a radio luminosity L ν,rad ∼ 10 16 erg s −1 Hz −1 at 3 GHz.This radio information thus gives us two models for the radio emission of TWA 5 A: (i) gyrosynchrotron emission that peaks around 3 GHz and quickly drops off beyond the turnover frequency; and (ii) ECM emission which cuts off before the X band. be recovered given the survey's flux density detection threshold.The ≈100% completeness of V-LoTSS was calculated to be 1.0 mJy by injecting and recovering point sources (Callingham et al. 2023), and thus we use this value as the detection threshold.The simulation gives us the expected ratio between detected sources within and beyond 50 pc. Finally, to determine how likely it is to only detect 1 CAS system beyond 50 pc given that there are 6 detections within 50 pc, we simply compute the p-value based on a Poisson distribution: where k and λ and the observed and expected number of radio detections beyond 50 pc respectively.Here, k = 1 and we compute λ = 13.3.This gives us a probability of 0.0022%, which is the likelihood that incompleteness can explain the lack of sources beyond 50 pc.Following the same procedures for the case of M dwarfs, however, gives a result consistent with observation: we obtain λ = 0, i.e. we expect no radio detections from M dwarfs outside of the 50-pc sphere. In addition, we also perform the V/V max statistical test to see if it agrees with the hypothesis that survey incompleteness alone cannot explain the lack of CAS detections.This test (also known as luminosity-volume test) was first introduced by Schmidt (1968) as a measure of the uniformity of the space distribution of radio sources.Originally, the test was to study the cosmic evolution prove that quasars are not at all uniformly distributed in space and the density of radio quasar evolves with distance.However, it is now also widely used also to test for volume completeness of a flux-limited population detected in a survey (e.g.Locatelli et al. 2019). The test is as follows: let V be the volume enclosed by a sphere with a radius equal to the distance d to a radio source, and V max be the volume enclosed within the maximum distance at which the source would be detectable.Then, for each source, one can calculate the ratio V/V max .If the sources are indeed uniformly distributed and their properties do not evolve with distance, the sample should have V/V max uniformly distributed between 0 and 1 with an average value ⟨V/V max ⟩ = 0.5 in Euclidean space.In a flux-limited survey, this ratio can be expressed as where N is the number of objects in the sample, S i is the flux density of the i th source, and S min is the detection threshold of the survey.This powerful test makes no assumption on the luminosity function, and thus should work on any flux-limited population.Now, we calculate the ⟨V/V max ⟩ for the V-LoTSS sample.For the non-CAS systems, we obtain ⟨V/V max ⟩ = 0.355 ± 0.075, an inconsistency at the ≈2σ level.For the CAS systems, we obtain ⟨V/V max ⟩ = 0.450 ± 0.116 which is consistent with the expected value of 0.5.The uncertainties are obtained via bootstrapping.In conclusion, there may be be some incompleteness in the V-LoTSS M-dwarfs population but the result is marginal.As such, we proceed with the assumption that we are not missing a significant population in our crossmatching.Eruptive* M3.0Ve (10) 18.46 2.44 0.80 ± 0.15 28.53 CR Dra * M1.5Ve (11) 20.15 2.17 0.83 ± 0.23 29.56 GJ 9603 * M2V (12) 21.49RS CVn F2IV+K2IV (24) 46.89 0.65 1.88 ± 0.33 30.75 2MASS J14333139+3417472 * M5V (25) 48.20 3.75 0.95 ± 0.13 27.92 (26) Notes.). (c) Stokes I flux density S I is given by LoTSS-DR2 (Shimwell et al. 2022). (d) X-ray luminosity X L of each stellar system is given by the Second ROSAT all-sky survey source catalogue (2RXS; Boller et al. 2016), unless specified otherwise.The derivation of the X L upper-limit values for some stellar systems in our sample is described in detail by Callingham et al. (2021b).† Both M-dwarf components of the binary system GJ 856 have very similar angular separation (≲ 1 ′′ ) with the LoTSS radio source.Since the GJ 856 A and GJ 856 B themselves are separated by around 1.25 ′′ , we cannot ascertain which one is (or if both are) emitting the radio signal.Their wide separation (> 10 AU) precludes direct magnetic or chromospheric interactions between them (cf.CAS systems).Eruptive* M3.5V (5) 13.35 2.69 −1.53 ± 0.14 77 ± 10 28.76 YY Gem RS CVn M1Ve+M1Ve (6) 15.08 1.93 −0.87 ± 0.13 47 ± 10 29.88 HAT 182-00605 * M4V (7) 17.90 2.92 −0.73 ± 0.11 82 ± 25 28.53 LP 212-62 * M5V (8) 18.19 3.40 −6.63 ± 0.13 75 ± 3 27.58 (23) GJ 3861 Fig. 1 : Fig.1: TESS flare statistics and debiasing.The top horizontal axis of each subfigure indicates the nominal stellar spectral types(Pecaut & Mamajek 2013).(a): The average TESS ensemble flare rate -defined as the fraction of flaring stars N flaring /N total in the TESS short-cadence year 1 & 2 observations, weighted by the average flare rate ν TESS -as a function of stellar effective temperature T eff .ν TESS follows the definition from Equation2.The green curve shows the average TESS ensemble flare rate using the entire TESS flare statistics which suffers greatly from TESS sampling bias (see Sect. 3.3.1),whereas the orange curve only considers TESS stars within 50 pc, with the blue curve as its polynomial fit (from numpy.polyfit;deg = 5) used for subsequent analysis.M dwarfs later than ∼M6 were not observed in a high enough sample size due to TESS target selection (which avoids very faint stars) and are thus excluded (represented by the red region).(b): Stellar radius R * as a function of T eff .The Basti-IAC isochrone model(Hidalgo et al. 2018) and the BHAC15 isochrone model(Baraffe et al. 2015) are combined to create our interpolated model.(c): Minimum flare energy E f,min detectable by TESS on a particular star with effective temperature T eff .The unit for the flare energy is arbitrary since we are only interested in the shape of the E f,min versus spectral type curve.(d): Same as Fig.1a, but with the debiased flare rate ν instead of TESS-observed flare rate ν TESS .This debiased average-TESS-ensemble-flare-rate curve is obtained by multiplying the two blue curves in Figs.1a and 1c. Fig. 2 : Fig. 2: Hertzsprung-Russell (HR) diagrams for the V-LoTSS × GCNS sample (left panel) and LoTSS × GCNS sample (right panel), according to Gaia G BP − G RP colour and Gaia G absolute magnitude.The radio-bright stellar systems are represented by different colours and symbols according to object classification, as shown in the legends.There exist two sources in the V-LoTSS × GCNS population that are beyond 50 pc, whereas only radio detections within 50 pc are included in the LoTSS × GCNS population to suppress chance-coincidence associations.The top axis of the HR diagrams indicates the nominal stellar spectral types (Pecaut & Mamajek 2013). Fig. 3 : Fig. 3: Hertzsprung-Russell (HR) diagram for the VLASS × GCNS sample according to Gaia G BP −G RP colour and Gaia G absolute magnitude.The radio-bright stellar systems are represented by different colours and symbols according to object classification, as shown in the legends.Green triangle represents stellar systems that are likely to be chromospherically active from literature.The radio sources here are all within 50 pc, as crossmatching the entire GCNS would lead to significant coincidence associations with VLASS sources.The top axis of the HR diagrams indicates the nominal stellar spectral types (Pecaut & Mamajek 2013). Fig. 4 : Fig. 4: Top panel: Zoomed-in version of Fig. 3 with RS CVn variables removed.The cyan line represents the median fiducial and the red line shows the same fiducial shifted by -0.753 mag, which corresponds to an unresolved binary system of two identical stars.Therefore, stellar systems that significantly above the green line cannot be explained simply by binarity.The top axis of the HR diagrams indicates the nominal stellar spectral types (Pecaut & Mamajek 2013).Bottom panel: Residuals of M G relative to the median fiducial (cyan). Fig. 5 : Fig.5: Cumulative distribution functions (CDFs) for the GCNS background distribution (dashed blue), the V-LoTSS × GCNS sample (green), the LoTSS × GCNS sample (orange), and VLASS × GCNS sample (grey), shown as a function of Gaia colour G BP − G RP .The top panel shows the two radio-detected populations with the inclusion of chromospherically active stellar (CAS) systems.The bottom panel shows the population without them.The shaded regions correspond to the 95% confidence level based on a binomial distribution.As the number of sources in GCNS is much larger than the radio-detected populations, the confidence interval of GCNS is negligible.The top axis of the CDF plots indicates the nominal stellar spectral types(Pecaut & Mamajek 2013). Fig. 7 : Fig.7: Cumulative distribution functions (CDFs) for the GCNS background distribution (dashed blue), the GCNS × TESS flare distribution (solid blue), the V-LoTSS × GCNS sample (green), the LoTSS × GCNS sample (orange), and VLASS × GCNS sample (grey), shown as a function of effective temperature T eff .The solid blue line is obtained from the product of the dashed blue line and TESS flare rate curve according to Fig.1d.The shaded regions correspond to the 95% confidence level based on a binomial distribution.Note that the stars with T eff < 2800 K (around spectral type M6; shaded red box) are excluded in our analysis due to their incompleteness in TESS.The top axis of the CDF plots indicates the nominal stellar spectral types(Pecaut & Mamajek 2013). Fig. 9 : Fig.9: Cumulative distribution functions (CDFs) for the GCNS background distribution (dashed blue), the GCNS × X-ray flare distribution (solid blue) according to the study byJohnstone et al. (2021) , the V-LoTSS × GCNS sample (green), the LoTSS × GCNS sample (orange), and the VLASS × GCNS sample (grey), shown as a function of stellar mass in units of solar mass M ⊙ .The shaded regions correspond to the 95% confidence level based on a binomial distribution.Note thatJohnstone et al. (2021) only consider stars of stellar mass between 0.1 and 1.2 M ⊙ , and so we exclude stars with less than 0.1 M ⊙ (around spectral type M6) in this specific analysis.The top axis of the CDF plots indicates the nominal stellar spectral types(Pecaut & Mamajek 2013). Fig. 10 : Fig. 10: Radio-detected populations (V-LoTSS on the left panel, LoTSS on the right panel) plotted against the canonical Güdel-Benz relationship (GBR) represented by the dashed blue line: L X = 9.48 × 10 18 L 0.73ν,rad(Williams et al. 2014), where L X is the soft (0.1-2.4 keV) X-ray luminosity and L ν,rad is the 144-MHz total-flux radio luminosity for the LOFAR population.The radio sources without a detectable X-ray luminosity are indicated with a downward-pointing arrow and shown as 3σ upper limits.Colours and symbols are as in Fig.2, with the additional classification of M dwarfs beyond M4 (represented by yellow circles) motivated by the M4 transition mentioned in Sect.4.3.2. Fig Fig. B.1: VLASS Epoch 1 (left panel) and Epoch 2 (right panel) images using CIRADA cutout service 10 , centred at the corresponding radio source.In both images, the lime dot represents the Gaia position of α For B with proper-motion correction, and the blue dot represents that of α For A. The FWHM beam size is shown in the bottom left corner of each image as a grey crossed circle.The rms noise σ is around 140 µJy in both epochs.The contours are of levels with a increment of 4σ, i.e. 4σ, 8σ, 12σ, etc.The peak flux of the source is around 3.3 mJy for Epoch 1 and 1.5 mJy for Epoch 2. Fig Fig. B.2: VLASS Epoch 1 (left panel) and Epoch 2 (right panel) images using CIRADA cutout service 11 , centred at the corresponding radio source.In both images, the lime dot represents the Gaia position of TWA 5 A with proper-motion correction, and the blue dot represents the position of TWA 5 B using astrometry by Neuhäuser et al. (2000).The FWHM beam size is shown in the bottom left corner of each image as a grey crossed circle.The rms noise σ is around 150 µJy in Epoch 1 and 130 µJy in Epoch 2.The contours are of levels with a increment of 4σ, i.e. 4σ, 8σ, 12σ, etc.The peak flux of the source is around 2.5 mJy for Epoch 1 and 2.7 mJy for Epoch 2. (a) Asterisk * denotes M dwarfs unless specified otherwise."Eruptive" denotes eruptive variable stars classified by SIMBAD.(b)Distance d (derived from the parallaxes) and Gaia colour G BP − G RP of each stellar systems are given by the Gaia Catalogue of Nearby stars(Gaia Collaboration et al. 2021b) which quotes measurements from Gaia Early Data Release 3(Gaia Collaboration et al. 2021a Notes.(a)Asterisk denotes M dwarfs unless specified otherwise."Eruptive*" denotes eruptive variable stars classified by SIMBAD.(b)Distance d (derived from the parallaxes) and Gaia colour G BP − G RP of each stellar systems are given by the Gaia Catalogue of Nearby stars(Gaia Collaboration et al. 2021b) which quotes measurements from Gaia Early Data Release 3(Gaia Collaboration et al. 2021a Table 1 : Parameters of the radio surveys utilised for identifying radio-bright stellar systems. Table D . 1 contains the definitions of all the acronyms frequently used in this work. Table D . 1: Table of acronyms frequently used in this work.Table A.1: Multi-wavelength properties of the 144-MHz detected LoTSS × GCNS population sorted by increasing distance d.The symbols G BP − G RP , S I , and log L X correspond to Gaia BP-RP colour, Stokes I flux density, and logarithmic soft (0.1-2.4 keV) X-ray luminosity respectively. Table A . 2: Multi-wavelength properties of the 144-MHz detected V-LoTSS × GCNS population sorted by increasing distance d.The symbols G BP − G RP , S V , S V /S I , and log L X correspond to Gaia BP-RP colour, Stokes V flux density, fraction of circularly polarised radio emission, and logarithmic soft (0.1-2.4 keV) X-ray luminosity respectively. Table A . 3: Multi-wavelength properties of the 3-GHz detected VLASS × GCNS population sorted by increasing distance d.The symbols G BP − G RP , S I , and log L X correspond to distance, Gaia BP-RP colour, Stokes I flux density, and logarithmic soft (0.1-2.4 keV) X-ray luminosity respectively.The last column refers to the VLASS epoch from which the radio data is used for this study, i.e. whether the value of S I is quoted from VLASS Epoch 1 or 2.	2023-12-13T06:42:42.028Z	2023-12-12T00:00:00.000	{ "year": 2023, "sha1": "8cca9420ca53075cb936e83e2630846a42299263", "oa_license": "CCBY", "oa_url": "https://www.edpsciences.org/images/stories/librarians/EDP-AA-S2O.pdf", "oa_status": "HYBRID", "pdf_src": "ArXiv", "pdf_hash": "8cca9420ca53075cb936e83e2630846a42299263", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
8975843	pes2o/s2orc	v3-fos-license	New film‐coated tablet formulation of deferasirox is well tolerated in patients with thalassemia or lower‐risk MDS: Results of the randomized, phase II ECLIPSE study Abstract Once‐daily deferasirox dispersible tablets (DT) have a well‐defined safety and efficacy profile and, compared with parenteral deferoxamine, provide greater patient adherence, satisfaction, and quality of life. However, barriers still exist to optimal adherence, including gastrointestinal tolerability and palatability, leading to development of a new film‐coated tablet (FCT) formulation that can be swallowed with a light meal, without the need to disperse into a suspension prior to consumption. The randomized, open‐label, phase II ECLIPSE study evaluated the safety of deferasirox DT and FCT formulations over 24 weeks in chelation‐naïve or pre‐treated patients aged ≥10 years, with transfusion‐dependent thalassemia or IPSS‐R very‐low‐, low‐, or intermediate‐risk myelodysplastic syndromes. One hundred seventy‐three patients were randomized 1:1 to DT (n = 86) or FCT (n = 87). Adverse events (overall), consistent with the known deferasirox safety profile, were reported in similar proportions of patients for each formulation (DT 89.5%; FCT 89.7%), with a lower frequency of severe events observed in patients receiving FCT (19.5% vs. 25.6% DT). Laboratory parameters (serum creatinine, creatinine clearance, alanine aminotransferase, aspartate aminotransferase and urine protein/creatinine ratio) generally remained stable throughout the study. Patient‐reported outcomes showed greater adherence and satisfaction, better palatability and fewer concerns with FCT than DT. Treatment compliance by pill count was higher with FCT (92.9%) than with DT (85.3%). This analysis suggests deferasirox FCT offers an improved formulation with enhanced patient satisfaction, which may improve adherence, thereby reducing frequency and severity of iron overload‐related complications. \| I N T R O D U C T I O N Transfusion and iron chelation therapy can be a lifelong requirement for many patients with transfusion-dependent anemias. Compliance with iron chelation therapy can influence the frequency and severity of iron overload-related complications, 1 with demonstrated improvement in organ dysfunction and survival in patients compliant with iron chelation therapy. [2][3][4][5][6] The once-daily oral deferasirox dispersible tablet (DT) formulation (Exjade ® ), available since 2005, offered an improved option over parenteral deferoxamine (Desferal ® ), providing greater compliance, patient satisfaction, and health-related quality of life. 7,8 The efficacy and safety of deferasirox DT has been well-defined through an extensive clinical trial program in adult and pediatric patients with a variety of anemias, including thalassemia, myelodysplastic syndromes (MDS), sickle-cell disease, and other rare anemias, [9][10][11][12][13] and has been used in clinical practice worldwide for over a decade. Nonetheless, barriers to optimal patient acceptance of treatment still exist with deferasirox DT, including palatability, the need to take the drug in a fasting state (ie, not being able to take with food), and drug-related side effects, notably gastrointestinal (GI) tolerability. 14 An improved filmcoated tablet (FCT) formulation of deferasirox (US, Jadenu ® ; EU, Exjade ® ) has therefore been developed for oral administration. Both deferasirox FCT and DT are once-daily, oral iron chelators that are dosed based on body weight. The FCT contains the same active substance, dose-adjusted to achieve comparable exposure to that achieved with the DT, 15 but excipients (lactose and sodium lauryl sulfate) have been removed. As a result of increased bioavailability of the FCT, doses required to achieve the same chelation effect are 30% lower than the DT. 15 Deferasirox DT is taken according to labeling recommendations on an empty stomach, at least 30 min before the next meal, and administration requires careful dispersion of the tablets in a glass of water, orange juice, or apple juice, and has a chalky consistency. Deferasirox FCT can be taken orally on an empty stomach or with a light meal (<7% fat content and 250 calories), offering a simpler and more convenient mode of administration, and potentially improved GI tolerability (due to a change in excipients and administration with food). >1000 ng/mL at screening. Key exclusion criteria were: creatinine clearance (CrCl) below contraindication limit as per local label (<60 mL/min or <40 mL/min); serum creatinine (SCr) >1.5 3 upper limit of normal (ULN); alanine aminotransferase (ALT) >5 3 ULN (unless liver iron concentration confirmed as >10 mg Fe/g dry weight 6 months prior to screening); urine protein/creatinine ratio (UPCR) >0.5 mg/mg; or impaired GI function. \| Study design ECLIPSE was an open-label, randomized, multicenter, two-arm, phase II study with the primary endpoint after 24 weeks of treatment (Supporting Information Figure S1). Randomization was stratified by underlying disease and previous chelation treatment. In all chelation-naïve patients, the starting deferasirox dose was 20 mg/kg/day with DT or 14 mg/kg/day with FCT. For previously treated patients, a washout period of 5 days was required before randomization; all pre-treated patients well managed on treatment with deferasirox DT, deferoxamine or deferiprone were requested to start on a DT or FCT dose equivalent to their pre-washout dose (eg, 20 mg/kg/day DT equivalent to 14 mg/ kg/day FCT equivalent to 75 mg/kg/day deferiprone equivalent to 40 mg/kg/day deferoxamine). Deferasirox DT was taken on an empty stomach, at least 30 min before the next meal; FCT was taken (no later than 12:00 pm) with or after a light meal. Dose adjustments to improve treatment response based on serum ferritin levels and investigator's judgment were recommended every 4 weeks for chelation-naïve patients and every 3 months for pre-treated patients, in increments of 5-10 mg/kg/day for DT or 3.5-7 mg/kg/day for FCT, up to a maximum dose of 40 mg/kg/day for DT and 28 mg/kg/day for FCT. Dose adjustments based on safety and dose reductions for patients unable to tolerate the protocol-specified dosing schedule were allowed at any time during the study. The study was conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki, and was approved by independent ethics committees at participating sites. Patients (or parents/ guardians) provided written, informed consent prior to enrollment. \| Outcomes The primary endpoint was overall safety of deferasirox FCT and deferasirox DT formulations, measured by frequency and severity of adverse events (AEs) and changes in laboratory values from baseline to 24 weeks. Secondary endpoints included the evaluation of both formulations on selected GI AEs (diarrhea, constipation, nausea, vomiting, and abdominal pain) during treatment, estimation of treatment compliance, evaluation of both formulations on patient satisfaction, palatability, and GI symptoms using PROs and evaluation of the PK of both formulations. Safety was evaluated by monitoring and assessing AEs, changes in laboratory parameters, and clinical observations from the start of study treatment to 30 days after the last intake of study drug. Compliance to treatment was evaluated by relative consumed tablet count (total consumed tablet count/total prescribed tablet count) and patient-reported treatment compliance using a daily compliance questionnaire. Patient satisfaction, palatability of medicine, and GI symptoms were measured for both formulations using PRO questionnaires (modified Satisfaction with Iron Chelation Therapy [SICT] and palatability questionnaire) and a GI symptoms diary. All PRO instruments used in this study (palatability, GI symptom and modified SICT questionnaires) and estimation of compliance (pill count) followed FDA Guidance to Industry for development. Comprehensive qualitative, linguistic and psychometric validation was performed within this trial; manuscripts on the qualitative validation and psychometric evaluation are in development. The modified TAHER ET AL. \| 421 SICT questionnaire used 5-point response scales to assess adherence (six questions), satisfaction/preference (two questions), and concern domains (three questions); higher scores in adherence and satisfaction/ preference domains indicated worse adherence, higher scores in concern domain indicated fewer concerns. The palatability questionnaire consisted of four items: taste and aftertaste of the medication (5-point response scale: 1 5 very good; 5 5 very bad), whether the medication was taken (ie, whether the patient vomited after swallowing medication or not) and how the patient perceived the amount of medication to be taken (not enough, just enough, or too much). The GI symptom diary consisted of six items: five items (pain in your belly, nausea, vomiting, constipation, diarrhea) rated on an 11-point scale (0 5 best, \| Statistical evaluations Standard descriptive analyses were performed for both formulation groups. No hypothesis was tested. The incidence of any AEs overall and by severity was summarized by treatment using frequency counts, percentages of patients, and 95% confidence intervals (CIs) for percentages obtained using Clopper-Pearson method. Laboratory data were summarized using absolute change from baseline by treatment arm at each post-baseline time window. The safety analysis set included all patients who received at least one dose of the study drug and was used for all safety evaluations. The PK analysis set consisted of all patients who received at least one dose of study medication and had at least one evaluable concentration measurement and was used for all PK analyses. Serum ferritin data were considered as an exploratory, non-safety outcome; absolute and relative change from baseline were summarized by treatment arm at each post-baseline visit. \| Exposure to treatment and compliance The mean actual deferasirox DT dose 6 SD received during the 24- week study was 27.5 6 7.73 mg/kg/day over a mean duration of 154.5 6 44.67 days (median 168.0 days); the mean actual deferasirox FCT dose 6 SD received was 20.8 6 5.44 mg/kg/day over a mean duration of 163.2 6 27.76 days (median 169.0 days; Table 2). More patients receiving FCT were in the longest exposure category (12 weeks; DT 89.5%; FCT 96.6%) and highest mean actual dose category (35 mg/kg/day DT/24.5 mg/kg/day FCT; DT n 5 16, 18.6%; FCT n 5 27, 31.0%; Table 2). However, post-hoc analyses identified that 23 patients on FCT (26%) were started on a dose that was higher than recommended in the protocol compared with eight patients (9.3%) on DT (not recognized or reported by the investigators as dosing error). Over 24 weeks, dose was interrupted at least once in 43 patients \| Adverse events Investigator-reported AEs regardless of relationship to deferasirox were reported in 77 patients (89.5%) on DT and 78 patients (89.7%) on FCT ( Table 3). The most frequently reported AEs were diarrhea, nausea, and abdominal pain ( Table S1). \| Post-hoc evaluation of patients with renal events (renal adverse events and abnormal renal laboratory parameters) For the purpose of this detailed evaluation of patients with renal AEs and abnormal renal laboratory parameters, a patient was classified as experiencing a renal event if one of the following criteria was met: a reported AE with the following preferred terms: renal impairment, blood creatinine increased, blood creatinine abnormal, glomerular filtra- \| Clinical PK Deferasirox pre-dose concentrations (C trough ) at steady state were similar to both DT and FCT formulations throughout the study. Geometric PK variability shown as coefficient of variation of geometric mean was smaller with FCT than DT. Results were similar with or without dose normalization, which suggest that overall exposure to deferasirox was similar for both formulations, with slightly higher post-dose concentrations with FCT. PK results observed in this study were consistent with data previously obtained in healthy volunteers (data on file). Figure 1D). The overall GI symptom scores were low for both formulations, indicating patients experienced very little trouble/concern associated with GI symptoms ( Figure 1E). Results favored FCT with patients reporting near-perfect scores for all three modified SICT domains and palatability. \| DISCUSSION The ECLIPSE study evaluated safety, PK, and PRO of the original deferasirox DT formulation and the new dose-adjusted FCT formulation, which contains the same active substance and can be swallowed without the need to disperse into a suspension, in patients with lower-risk Mean domain scores for patient-reported outcomes (adherence, satisfaction/preference, and concern) (A-C), mean palatability score (D), and mean gastrointestinal symptom scores (E). For adherence (A; scale 6-30), satisfaction/preference (B; scale 2-10), and GI symptoms (E; scale 0-50), higher scores indicate worse outcomes/symptoms. For concern (C; scale 3-15) and palatability (D; scale 0-11), higher scores indicate fewer concerns and better palatability. A-D, baseline was defined as week 2 assessment. If missing, then the week 3 assessment was considered baseline; E, baseline was defined as week 1 score. If missing, then the week 2 score was considered baseline. BL, baseline. GI disturbances are often reported during clinical evaluation of deferasirox, usually mild-to-moderate and occurring early in the course of treatment. 16 As deferasirox FCT can be taken with a light meal and also lacks the excipients lactose and sodium lauryl sulfate, both found in the original DT formulation and possibly implicated in GI AEs, it was expected that deferasirox FCT would show improved GI tolerability. Although similar numbers of patients in each arm experienced one or more GI AEs or received dose adjustments as a result of GI AEs, 12.8% and 4.6% of patients in the deferasirox DT and FCT groups, respectively, experienced severe GI AEs (including diarrhea, nausea, and abdominal pain) and four patients (DT) and one patient (FCT) discontinued treatment because of GI AEs. These results were reflected in the PROs: patients receiving FCT reported little or no concern with GI symptoms. Taken together, these results suggest that the GI tolerability profile may be improved with FCT compared with DT, which could be because of the change in excipients and/or the ability to take the medicine with a light meal. Further insight should be gained once longer-term data are available. In the current study, more patients receiving FCT experienced renal AEs or abnormal renal parameters than those receiving DT, although the number of patients with renal laboratory values in the notable/extended ranges were either similar or lower in the FCT arm. Renal laboratory changes and AEs are well characterized with deferasirox therapy and are generally mild, non-progressive, and reversible. 17 As such, the observed imbalance in reported renal events between the two treatment arms are likely attributed to a larger proportion of patients in the FCT group receiving a higher than recommended dose, as well as non-adherence to protocol-recommended dose modifica-	2017-03-06T08:46:50.166Z	2017-02-18T00:00:00.000	{ "year": 2017, "sha1": "facf66a1abb5309e41f52173359650d43bcc7f50", "oa_license": "CCBYNC", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/ajh.24668", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "facf66a1abb5309e41f52173359650d43bcc7f50", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
7361492	pes2o/s2orc	v3-fos-license	Molecular structure‐function relationship of dietary polyphenols for inhibiting VEGF‐induced VEGFR‐2 activity 1 Scope We recently reported potent inhibition of VEGF signalling by two flavanols at sub‐micromolar concentrations, mediated by direct binding of the flavanols to VEGF. The aim of this study was to quantify the inhibitory potency and binding affinity of a wide range of dietary polyphenols and determine the structural requirements for VEGF inhibition.2 Methods and results The concentration of polyphenol required to cause 50% inhibition (IC50) of VEGF‐dependent VEGFR‐2 activation in HUVECS was determined after pretreating VEGF with polyphenols at various concentations. Binding affinities and binding sites on VEGF were predicted using in‐silico modelling. Ellagic acid and 15 flavonoids had IC50 values ≤10 μM while 28 other polyhenols were weak/non‐inhibitors. Structural features associated with potent inhibition included 3‐galloylation, C‐ring C2=C3, total OH, B‐ring catechol, C‐ring 3‐OH of flavonoids. Potency was not associated with polyphenol hydrophobicity. There was a strong correlation between potency of inhibition and binding affinities, and all polyphenols were predicted to bind to a region on VEGF involved in VEGFR‐2 binding.3 Conclusion Specific polyphenols bind directly to a discrete region of VEGF and inhibit VEGF signalling, and this potentially explains the associations between consumption of these polyphenols and CVD risk. Introduction Plants synthesise a range of polyphenols via the phenylpropanoid biosynthetic pathway and a range of these are consumed as part of human diets in the form of foods and beverages. Flavonoids make up about one third and other polyphenols (mainly simple phenolics) about two thirds of the polyphenols consumed in human diets, with estimates of average daily intakes being around 1200 mg per day [1]. There has been considerable interest in the potential health protective properties of dietary polyphenols, supported by epidemiological evidence of an inverse relationship between consumption of flavonoids and other polyphenols and risk of various chronic diseases including coronary heart disease, coronary artery disease, stroke, and certain cancers [2]. In addition, there are now hundreds of reports of randomised controlled dietary interventions showing beneficial effects on CVD risk factors and biomarkers of vascular health [3,4]. There are also thousands of scientific reports providing evidence of a multitude of biological activities for polyphenols that have been demonstrated in vitro. However, the majority of the biological activities reported to date remain theoretical because direct evidence of these mechanisms occurring in vivo has not been demonstrated. In the majority of cases, the activities observed in vitro have been in response to supra-physiological concentrations and/or non-physiological compounds [5]. To date, the most scientifically robust evidence of health benefits accruing from consumption of dietary polyphenols centres on cardiovascular diseases. The biggest cause of cardiovascular disease is atherosclerosis which is a long-term process in which artery walls become thickened through the formation of numerous atherosclerotic plaques. This process is dominated by the recruitment of white blood cells to artery walls and the inflammatory responses they produce, and is driven by accumulation of low-density lipoproteins (which are taken up by the WBCs), differentiation of the white cells into macrophages (pro-inflammatory) and insufficient removal of fats and cholesterol by high-density lipoproteins. The growth of atherosclerotic plaques causes thickening of the vascular intima and a crucial consequence of this is to increase the thickness above that capable of supporting diffusion of oxygen and nutrients from the blood. This triggers angiogenesis and the formation of new blood vessels to supply the deeper parts of the plaque with oxygen and nutrients. Angiogenesis, the process resulting in the formation of new blood vessels from pre-existing ones, plays an important role in the development and destabilisation of atherosclerotic plaques [6][7][8]. Vascular endothelial growth factor (VEGF) has been shown to be the most important pro-angiogenic growth factor in humans [9][10][11] exerting its angiogenic effects by stimulating VEGF receptor 2 (VEGFR-2), which is critical for promoting proliferation and differentiation of endothelial cells [11,12]. Indeed, it has been demonstrated directly that VEGF drives the growth of existing plaques in high fat diet-fed New Zealand White rabbits and hypercholesteraemic ApoE −/− mice [13,14]. Therefore, VEGF-dependent angiogenesis is an essential process required for growth, healthy functioning of tissues and wound repair (physiological angiogenesis), but can also play a role in pathological processes including atherosclerosis (pathological angiogenesis). Recently, we reported that two flavan-3-ols (epigallocatechin gallate and a procyanidin tetramer), but not others like epicatechin, potently inhibited VEGF-induced VEGFR-2 phosphorylation in human umbilical vein endothelial cells (HUVECs), and that potent inhibition only occurred when the VEGF was pre-incubated with the polyphenols [15]. This observation, along with evidence from additional experiments, indicated that the polyphenol-induced inhibition of VEGFmediated activation of VEGFR-2 was a result of direct binding of the polyphenol(s) to the VEGF protein. This finding was particularly noteworthy because for the first time VEGF was identified as a molecular target that certain polyphenols could interact with directly, and this linked polyphenol biological activities to a receptor (VEGF receptor-2; VEGFR-2) [15]. In addition, the concentrations of polyphenols required to substantially inhibit VEGFR-2 signalling were low (in the region of 100-200 nM) and were physiologically relevant. The overall aim of the research reported here was to assess the ability of a wide range of polyphenols to bind to VEGF and consequently inhibit VEGF-dependent VEGFR-2 signalling, and to develop a better understanding of the structural and physicochemical features of polyphenols that contribute to them being potent or impotent inhibitors. We also determined binding affinities for all the polyphenol-VEGF interactions, and the location of the binding site(s) on the VEGF macromolecule. Polyphenols for study Isolated single-mass procyanidin fractions with various degrees of polymerisation (dp 2, dp 3, dp 4) were isolated from apples as previously described [16,17]. EGCG was isolated from green tea as previously described [15]. Orobol was synthesised from phloroglucinol and 3,4-dihydroxyphenylacetic acid using the general procedure of Goto et al. [18] previously used for synthesis of isoflavones. Polyphenol treatments of VEGF and estimation of IC 50 values for inhibition of VEGF-mediated VEGFR-2 activation The specific assay used to assess the potency with which polyphenols interact directly with VEGF and inhibit VEGFinduced VEGFR-2 phosphorylation has been described elsewhere [15]. This assay comprises a pre-incubation where the polyphenol and the VEGF are mixed and incubated together for 5 min, after which the residual VEGF activity is assessed www.mnf-journal.com by adding the mixed polyphenol/VEGF solution to HUVECS for 5 min and the resulting activation of VEGFR-2 is quantified using an ELISA assay based on a monoclonal antibody specific for the (Phos) Tyr1175 residue in VEGFR-2. For each concentration tested, the polyphenol (stocks dissolved in DMSO) was incubated with VEGF (25 ng/mL: human recombinant VEGFA 165 ; R&D Systems, Abingdon, UK) in endothelial basal medium (EGM-2 with no serum or growth factors) for 5 min (final concentration of DMSO ࣘ 0.1%). Towards the end of the 5 min incubation period, confluent HUVECs were washed two times with warm PBS in preparation for the addition of the mixture of VEGF and polyphenol. Control incubations included a vehicle control (equivalent concentration of DMSO, ࣘ 0.1%) and VEGF (25 ng/mL and final concentration of DMSO ࣘ 0.1%) alone. After 5 min exposure of the HUVECs to the pre-incubated treatments, cells were washed two times with ice-cold PBS and then lysed with RIPA buffer (radio-immunoprecipitation assay buffer) containing protease and phosphatase inhibitors. Lysates were transferred to eppendorf tubes and kept on ice for 15 min with periodic vortexing, and subsequently the lysates were clarified by centrifugation at 13 000 g for 10 min at 4ЊC. Supernatants were stored at -80ЊC until analysis. The total protein content of lysates was determined using a commercially available BCA assay (Sigma, Poole, UK). Initially, polyphenols were tested at a higher concentration (100 M) to assess if they had any significant inhibitory activity. Those polyphenols that significantly decreased the phosphorylation of VEGFR-2 in HUVECs at 100 M were pre-incubated with VEGF at a range of concentrations (0.025-200 M). The specific concentrations selected for each polyphenol were initially estimated from the extent of inhibition at 100 M. If the inhibition of VEGFR-2 activation was always less than 50% or always more than 50% then additional assays were conducted such that final datasets for IC 50 estimations included at least 4 and up to 10 different polyphenol concentrations that spanned above and below the final estimated IC 50 value. Phosphorylated VEGFR-2 ELISA Quantification of phosphorylated VEGFR-2 in lysates was determined using a PathScan Phospho-VEGFR-2(Tyr1175) sandwich ELISA kit (Cell Signalling Technology, Hitchin, UK), following the manufacturer's instructions. The half inhibitory concentrations (IC 50 ) and their confidence intervals were determined by using the log (inhibitor) versus normalised response -variable slope analysis tool in the Graph-Pad Prism software. Prediction of polyphenol-binding sites on VEGF The crystal structure of VEGF was obtained from the RCBS protein data bank (PDB code: 2vpf, [19]). Structures of the ligands (polyphenols) used for docking were obtained from the PubChem chemical library [20]. All ligands were subject to binding to VEGF using AutoDock Vina in the PyRX 0.8 Virtual Screening Tool [21]. For each ligand the conformer with the lowest free binding energy was taken as the optimal docking conformation. Results and discussion We have previously reported that EGCG from green tea and a tetrameric procyanidin oligomer from apple are potent inhibitors of VEGF-induced VEGFR-2 signalling, and achieved this by tightly binding to the VEGF protein and reducing its binding to the VEGFR-2 receptor [15]. The polyphenol-induced inhibition of VEGF-induced VEGFR-2 activation occurred at nanomolar concentrations for these two polyphenols, which may be achieved through diet (IC 50 values estimated as 88 nM for EGCG and 280 nM for the procyanidin tetramer, Table 1). To further evaluate the potential for polyphenols to inhibit VEGF-dependent VEGFR-2 activation through direct interaction with VEGF, we first expanded our investigation into a range of flavanols with different structures and determined what structural features were compatible with potent inhibition. Subsequently, we extended the investigation of structure-activity relationships to include a range of polyphenols and phenolics with different chemical and structural characteristics. This allowed us to define the key chemical and structural features of polyphenols associated with potent inhibition of VEGF-dependent VEGFR-2 activation. Concentration-dependent inhibition of VEGF-mediated VEGFR-2 activation and influence of flavan-3-ol structure Representative examples of the dose-dependent inhibition of VEGFR-2 activation by polyphenol pre-treated VEGF are shown in Fig. 1. It is immediately obvious from such plots of percent inhibition (y-axis) plotted against log polyphenol concentration (x-axis) that some flavanols are very potent inhibitors of VEGF-mediated VEGFR-2 activation while others are essentially inactive. The galloylated monomeric catechins epigallocatechin gallate, catechin gallate and epicatechin gallate were all potent inhibitors with estimated IC 50 values of 0.088, 0.094 and 0.16 M, respectively. Epigallocatechin and (+)-catechin were very weak inhibitors (42.9 and 215 M, respectively) and (−)-epicatechin was not an inhibitor at even the highest concentration tested (Table 1). For procyanidins, the tetrameric procyanidin was the most potent, the trimeric dp3 slightly less potent, whereas the procyanidin dimer was a weak inhibitor and the monomers were extremely weak inhibitors. These data show that inhibitory activity increases with increasing degree of polymerisation at least up to dp4, while larger oligomers have not been directly www.mnf-journal.com Table 1. tested. Hydroxylation of the B-ring also influenced inhibitory activity; e.g. the trihydroxylated B-ring in EGC conferred stronger inhibitory activity compared to the corresponding dihydroxylated epicatechin. Galloyl esterification of catechins 3-Galloyl esters of (+)-catechin and (−)-epicatechin were potent inhibitors of VEGF-mediated VEGFR-2 activation, whereas the corresponding non-galloylated compounds were very weak inhibitors (ß2000-fold higher IC 50 values), suggesting an important role for the gallic acid ester group in monomeric catechins. It has been reported that gallic acid esters of flavanols such as epigallocatechin (giving rise to EGCG) are potentially very unstable and degrade rapidly in physiological buffers [22]. In order to test whether the potent inhibitory nature of (epi)catechin gallic acid esters is a feature of the labile galloyl ester which can potentially release a galloyl group, we tested methylgallate and gallic acid. However, no inhibitory activity could be detected for either gallic acid or methyl gallate. These data suggest that the potent inhibitory activity conferred by the presence of galloyl-esterification of the 3-hydroxy group in (epi)catechins contributes to increase the inhibitory potency of the intact molecule rather than facilitate the release of a galloyl group which subsequently interacts with VEGF. Table 2). The presence of the C2═C3 double bond of flavonoids contributes strongly to the inhibition of VEGF-induced VEGFR-2 activation, and hydrogenation of the C2═C3 double bond reduces VEGF inhibitory activity significantly. The molecules with saturated C2═C3 bonds (flavanones, flavanols, flavanonols) permit more twisting of the B-ring in relation to the C-ring, whereas a C2═C3 double bond increases the p-conjugation of the bond linking the B-and C-rings, which favors near-planarity of the two rings. Molecules with near-planar structures have been shown to more easily enter hydrophobic pockets in proteins [23,24]. Furthermore, quercetin exhibited a >250-fold lower IC 50 value compared with (+)-catechin, whose structure does not contain either the C2═C3 double bound or the 4-oxo group (Table 2). Therefore, the C2═C3 double bond in conjugation with the 4-oxo group of flavonoids also makes a significant contribution to the inhibition of VEGF-induced VEGFR-2 activation. It appears that planarity of the C-ring in flavonoids maybe important for binding interaction with proteins. Differences in the C-ring Although neither cyanidin or (+)-catechin possess either the C2═C3 double bound or the 4-oxo group in the C-ring, the IC 50 value for cyanidin was 200-fold lower than (+)-catechin ( Table 2). This is likely to be due to the presence of the positive charge on the C-ring of cyanidin which is the major difference between the two structures, and if so it suggests that cyanidin may interact with a negatively charged region/residue on the protein which contributes to stabilising the cyanidin-VEGF complex. The same trends were observed for other compounds sharing the same OH group substitution patterns but with differences in the C-ring. For example, luteolin was a more potent inhibitor of VEGF-induced VEGFR-2 activation than eriodictyol, which is equivalent in structure except that the C2═C3 double bond is lacking (saturated) ( Table 2). Further, myricetin inhibited VEGF activity with an IC 50 >300-fold lower than EGC, with the main structural difference being the lack of the C2═C3 double bound in conjugation with the 4-oxo group compared with myricetin (Table 1). Isoflavones are structural isomers of flavones whose B-ring is positioned on C3 instead of C2 in the C-ring. This change -which orientates the B-ring at an angle approximately 60°around the C-ring towards the keto group -significantly alters the inhibition of VEGF-induced VEGFR-2 activation. For example, the isoflavone orobol exhibited a 2-fold lower IC 50 value compared with the flavone luteolin which has an otherwise equivalent structure (Table 1). Therefore, positioning of the B-ring on C3 of the C-ring produces more potent VEGF inhibitors than structures where the B-ring is positioned on C2 of the C-ring. One plausible explanation is that this serves to position one or more hydroxyl groups in the B-ring closer to specific amino acid side chains in VEGF and enhances the binding affinity. Hydroxylation on B-ring of flavonoids Within the same class, the number of OH groups on the Bring of polyphenols affected the VEGF-inhibitory activity. For example, tri-hydroxylated B-ring flavan-3-ols such as EGCG (5,7,3 ,4 ,5 , 3-gallate) showed IC 50 values 2-fold lower than the dihydroxylated B-ring equivalent ECG (5,7,3 ,4 , 3-gallate), and the IC 50 for tri-hydroxylated EGC (3,5,7,3 ,4 ,5 ) was 42.9 M while di-hydroxylated epicatechin (3,5,7,3 ,4 ) . VEGF inhibitory activity was affected not just by the number of hydroxyl groups in the B-ring but also by the position of hydroxylation. For example, although quercetin (3,5,7,3 ,4 ) and morin (3,5,7,2 ,4 ) differ only in the position of the two hydroxyl groups on the B-ring, the IC 50 of morin was about 2-fold higher than that of quercetin (Table 1). This observation is consistent with (i) a hydroxyl group in the 3 -position more strongly interacting (e.g. through Hbonding) with VEGF than a hydroxyl group in the 2 -position, or (ii) the chemistry of the catechol group being important for binding to VEGF. Since in all cases increased hydroxylation of the B-ring conferred increased inhibition of VEGF-mediated VEGFR-2 activation, and the position of hydroxylation was also important, hydroxylation of the B-ring clearly plays an important role in the interaction with the VEGF protein. Hydroxylation on A-ring of flavonoids In contrast to hydroxylation of the B-ring, the effects of hydroxylation of the A-ring on VEGF activity depended on the polyphenol class. For flavonols, the IC 50 value decreased significantly (and potency increased) as the number of OH group substitutions on the A-ring increased. For example, quercetagetin (3,5,6,7,3 ,4 ) was a sevenfold more potent VEGF inhibitor than quercetin (3,5,7,3 ,4 ). There was also a strong effect of position of A-ring hydroxyl groups. For example, luteolin (5,7,3 ,4 ) with OH groups in a meta arrangement potently inhibited VEGF activity (IC 50 = 7.46 M) while 3 ,4 ,7,8-tetrahydroxyflavone with OH groups in an ortho position only inhibited VEGF activity by 63.2 % at 100 M (Table 1) Hydroxylation on C-ring of flavones Hydroxylation of position 3 of the C-ring of flavones enhanced the inhibition of VEGF-induced VEGFR-2 activation. For example, the IC 50 value for quercetin (3,5,7,3 ,4 ) was 10 times lower than that for luteolin (5,7,3 ,4 ) (Table 1). Similarly, galangin (3,5,7) was a more potent inhibitor than chrysin (5,7) (85% at 50 M and 58.6% at 100 M, respectively), and 3-hydroxyflavone caused 52.8 % inhibition at 100 M while flavone was ineffective at the same concentration ( Table 1). The consistent observation of increased potency of inhibition for 3-hydroxylated flavonoids strongly suggests that the 3-hydroxy group is directly involved in binding to VEGF. Research to establish direct evidence of interactions between specific groups on flavonoids with VEGF warrants further investigation. Total OH groups Across the range of compounds studied, there were three major causes of differences in the total number of hydroxyl groups: (i) variations in the number of hydroxyl substitutions on monomeric flavonoids, (ii) 3-galloyl esterification of flavan-3-ols, and (iii) oligomerisation of flavanols. There was a positive correlation (R 2 = 0.9873) between the total number of hydroxyl substitutions and the potency of a flavonoid in terms of inhibition of VEGF-induced VEGFR-2 activation within the non-gallated monomeric flavonoids (Fig. 2a) except for (+)-catechin and rhamnetin. The lack of the C2═C3 double bound or the 4-oxo group in the C-ring of (+)-catechin and the methylation on position 7 of rhamnetin have a greater effect on the IC 50 value than the total number of OH groups ((+)-catechin: 5 OH groups, IC 50 = 214.7 M; rhamnetin: 4 OH groups, IC 50 = 0.547 M). As discussed above, 3galloylation substantially increases the potency of inhibition of flavan-3-ols and this may be related to the additional three hydroxyl groups this adds to the structure. Increasing the degree of polymerisation of flavan-3-ols increases the total number of hydroxyl groups and also increases their inhibition potency (Fig. 2b). However, it is not possible to conclude that the increased potency is predominantly due to the increased total number of hydroxyl groups because there is also a substantial change in the total molecular mass as the dp increases. Conjugation of hydroxyl groups: methylation and glycosylation Polyphenols are often found with substitution of hydroxyl groups. For example, in plants flavonoids are almost always found O-linked to sugars through one or more hydroxyl positions on the A, B or C-rings. Further, during absorption in humans, flavonoid glycosides are deglycosylated by human ␤-glucosidases in the gastrointestinal tract [25][26][27] and then undergo phase-2 conjugation in the enterocytes and the liver [28,29]. For some flavonoids such as quercetin (onions, apples, tea), hesperitin (oranges) and (−)-epicatechin (cocoa, dark chocolate and apples), conjugation is extensive and many phase-2 conjugates of the flavonoids are found (Oglucuronides and O-sulphates of the aglycone and methylated derivatives) [30][31][32][33]. In contrast, for other flavonoids and phenolics such as epigallocatechin gallate (green tea) and ellagic acid (pomegranate, raspberries, walnuts), phase-2 conjugation is hardly detectable and the predominant form of the polyphenol in blood is the aglycone [34]. We therefore evaluated whether conjugation and 3-O-glycosylations of hydroxyl groups impacted on the ability of the compounds to inhibit VEGF-induced VEGFR-2 activation. Isorhamnetin, which has a methyl group on position 3 , had an estimated 12-fold higher IC 50 value than quercetin, which has a hydroxyl group on that position (Table 1). Although IC 50 values were not calculated for the flavanones, since they were not potent VEGF inhibitors, the same trend was observed for the inhibition percentage of VEGF-induced VEGFR-2 activation when naringenin and isosakuranetin were compared at the same concentration; the inhibition percentage of isosakuranetin with a 4 -methyl substitution was sixfold lower than for naringenin which has a hydroxyl group on that position (Table 1). Thus, methylation of hydroxyl groups on the B-ring of flavonoids decreases the inhibition of VEGF-induced VEGFR-2 activation. In contrast, methylation of hydroxyl groups on the A-ring did not affect the potential for inhibition of VEGF-induced VEGFR-2 activation. Rhamnetin, which has a methyl group on position 7, possessed a similar IC 50 to quercetin, which presents a hydroxyl group on that position (Table 1). On the other hand, methylation of hydroxyl groups on the A-and B-ring in the same molecule decreased the inhibition of VEGF-induced VEGFR-2 activation (sinensetin: methyl groups on position 5,6,7,3 ,4 inhefective at 100 M; quercetagetin: hydroxyl groups on position 3,5,6,7,3 ,4 IC50 = 0.096 M). We also tested whether several 3-O-glycosylated forms (galactoside, rutinoside, rhamnoside and glucopyranoside) of quercetin, which was one of the most potent inhibitors (IC 50 : 0.754 M), modified the inhibition of VEGF-induced VEGFR-2 activation. All of the glycosides tested here (hyperoside, hirsutrin, quercitrin and rutin) were completely ineffective at 50 M. In addition, phloridzin, the 2 -glucoside form of phloretin was completely ineffective and piceid (resveratrol-3-glucoside) inhibited VEGF activity by only 28% at 50 M (Table 1). These data show that glycosylation has a strong negative impact on the effectiveness of polyphenols as inhibitors of VEGF-induced VEGFR-2 activation. There are a number of possible reasons why glycosylation has such a strong effect on VEGF inhibition activity. For example, glycosylation would be expected to introduce a substantial change to the molecular shape and possibly sterically hinder access of key polyphenol substitutions to the VEGF molecule, thus weakening binding to VEGF. Relationship with polyphenol hydrophobicity A number of polyphenols are known to interact with proteins in a non-specific manner, with binding usually driven by hydrophobic or hydrogen bonding interactions between the hydrophobic phenolic rings of the polyphenols and hydrophobic surface regions of proteins [35][36][37]. However, these non-specific interactions have only been described at high polyphenol concentrations that are orders of magnitude higher than the IC 50 values for the VEGF inhibitors reported here (Table 1). Nevertheless, we investigated whether there was a relationship between potency of inhibitory activity and the 'hydrophobicity' of the polyphenol, but show that there is no correlation between the polyphenols distribution coefficient (log D octanol/water ) and the capacity to inhibit VEGF-dependent VEGFR-2 activation (IC 50 ) (Fig. 3B) (see Supporting Information for LogD data). This observation provides further evidence that the interactions between certain polyphenols and the VEGF protein that cause inhibition of VEGF function are due to site-specific binding events rather than non-specific interactions. Polyphenol-VEGF binding affinities Binding sites and binding affinities for the lowest energy binding interactions between VEGF and all 44 polyphenols were predicted using in-silico modelling. The lowest energy poses of polyphenols with dimeric VEGF predicted that all the polyphenols bound to a similar region of VEGF, namely at the end of the groove between the two monomers ( Fig. 4 and [12]). However, there were some differences in the predicted polyphenol orientation relative to the VEGF protein. For example, the inhibitors quercetin, luteolin and cyanidin were predicted to bind in a different orientation compared to the non-inhibitor (+)-catechin (Fig. 4). In vivo, VEGF functions as a dimer, with the two VEGF molecules positioned antiparallel, and this generates two identical ends to the grooves which have been shown to contain the amino acid residues that are involved in VEGF binding to VEGFR-2 [38]. We reported previously [15] that EGCG was predicted to bind to VEGF in a groove formed between the two VEGF monomers and next to the VEGFR2 binding site which comprises Tyr36, Ile43 and Ile46 on one side, Asp63 and Glu64 on the other side, with Ser30 and Asp34 forming the bottom of the groove. Here, we report that several other monomeric flavonoids that are inhibitors of VEGF also bind in this groove which is located very close to the binding site for VEGFR2. These monomeric flavonoid inhibitors are predicted to be sufficiently close to the side chains of Phe36, Lys48, Asn62 and Asp63 to be able to interact with them. The predicted binding affinities for the binding of the various polyphenols to the ends of the groove in VEGF varied widely from -8.3 kcal/mol for catechin gallate and EGCG to -5.2 kcal/mol for methyl gallate (see Supporting Information). Remarkably, there was a very strong correlation (r = 0.9445, p<0.0001) between the predicted binding affinities and the experimentally determined IC 50 vales for the set of 15 monomeric polyphenols that exhibited significant inhibitory activity (Fig. 3A). This data provides further evidence to support the notion that the interaction of polyphenols with VEGF is a result of specific rather than non-specific binding to the protein. Concluding remarks Data presented here show that some specific polyphenols are potent inhibitors of VEGF-dependent VEGFR-2 activation, with 11 polyphenols possessing IC 50 values ࣘ1 M, while many other polyphenols are very weak inhibitors or are not inhibitors. The potency of inhibition was strongly related to: the presence of a galloyl group at 3-positition of flavan-3-ols such as EGCG, CG and ECG; the degree of polymerisation of procyanidin oligomers; the presence of a C2═C3 double bound in the C-ring, especially if conjugated with the 4-oxo group (flavones and flavonols); the total number of hydroxyl groups on the B-ring; the presence of the catechol group on the B-ring; hydroxylation of position 3 on C-ring; lack of further substitution of hydroxyl groups on the B-ring. Inhibitory activity was not associated with polyphenol hydrophobicity, and the polyphenols were predicted to bind to a similar region of VEGF which is involved in interactions with its receptor VEGFR-2, with binding affinities strongly correlated with inhibitory potency. These data are consistent with the notion that potent VEGF inhibitory activity of specific polyphenols is a consequence of specific binding of polyphenols to distinct binding sites on VEGF dimers. It is noteworthy that the polyphenols identified here to have the highest affinities to and to show the strongest inhibition of VEGF activity are those that have most commonly been reported to exhibit antiangiogenic and anti-atherosclerotic activities in a range of in vitro models [39].	2018-04-03T03:36:32.664Z	2015-09-08T00:00:00.000	{ "year": 2015, "sha1": "1174c062d1b999945735095730c177bbb0f1b8fa", "oa_license": "CCBY", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/mnfr.201500407", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "1174c062d1b999945735095730c177bbb0f1b8fa", "s2fieldsofstudy": [ "Chemistry", "Medicine" ], "extfieldsofstudy": [ "Chemistry", "Medicine" ] }
20063296	pes2o/s2orc	v3-fos-license	Genetic variants affecting cross-sectional lung function in adults show little or no effect on longitudinal lung function decline Background Genome-wide association studies have identified numerous genetic regions that influence cross-sectional lung function. Longitudinal decline in lung function also includes a heritable component but the genetic determinants have yet to be defined. Objectives We aimed to determine whether regions associated with cross-sectional lung function were also associated with longitudinal decline and to seek novel variants which influence decline. Methods We analysed genome-wide data from 4167 individuals from the Busselton Health Study cohort, who had undergone spirometry (12 695 observations across eight time points). A mixed model was fitted and weighted risk scores were calculated for the joint effect of 26 known regions on baseline and longitudinal changes in FEV1 and FEV1/FVC. Potential additional regions of interest were identified and followed up in two independent cohorts. Results The 26 regions previously associated with cross-sectional lung function jointly showed a strong effect on baseline lung function (p=4.44×10−16 for FEV1/FVC) but no effect on longitudinal decline (p=0.160 for FEV1/FVC). This was replicated in an independent cohort. 39 additional regions of interest (48 variants) were identified; these associations were not replicated in two further cohorts. Conclusions Previously identified genetic variants jointly have a strong effect on cross-sectional lung function in adults but little or no effect on the rate of decline of lung function. It is possible that they influence COPD risk through lung development. Although no genetic variants have yet been associated with lung function decline at stringent genome-wide significance, longitudinal change in lung function is heritable suggesting that there is scope for future discoveries. INTRODUCTION Reduction of FEV 1 relative to FVC defines COPD, one of the leading causes of death worldwide. Measures of lung function are also important predictors of morbidity and mortality in the general population. [1][2][3] While environmental factors, particularly smoking, impact lung function, genetic variation is also a major determinant. 4 Genome-wide association studies (GWAS) to date have identified numerous regions associated with lung function measured at a single point in time (ie, cross-sectional lung function). [5][6][7][8][9] The lung function attained at a given time point in adulthood will be influenced by factors that affect either the development of lung function in earlier life or the rate of subsequent decline in lung function or both. Both cross-sectional lung function and longitudinal change in lung function are heritable. Although heritability estimates for longitudinal decline in lung function range between 10% and 39%, 10 the individual genetic determinants have yet to be defined. Identifying the responsible genes could provide a promising route for intervention in COPD, since this is typically diagnosed well after lung function has reached its peak and modifying its further decline could prove to be a feasible therapeutic option. The objectives of our study were as follows: (1) to examine the association with longitudinal change for those regions previously identified as significantly associated with cross-sectional FEV 1 or FEV 1 /FVC and (2) to seek novel variants which reach genome-wide significance for association with longitudinal change in lung function, using a cohort with multiple lung function measurements over an extended period of up to 40 years and an imputation panel which provides dense coverage of both common and low-frequency variants. Discovery data source and study population The Busselton Health Study (BHS) is a longitudinal health survey that began in 1966 in the town of Busselton in the southwestern region of Western Australia. In 1994/1995, a crosssectional community follow-up study was undertaken where blood was taken for DNA extraction. A sample of 1468 individuals with European ancestry were genotyped using the Illumina 610-Quad BeadChip (BHS1) and subsequent genotyping was carried out on an independent group of 3407 individuals with European ancestry using Illumina 660W-Quad (BHS2). Spirometric measures of FEV 1 and FVC were assessed. From 1966 to 1978 (five surveys), FEV 1 and FVC were measured using a McDermott dry spirometer (Pneumoconiosis Research Unit, Penarth, UK) and recorded as the highest values from three measurements ( provided that two recordings were within 10% of each other). Wedge spirometers (Vitalograph, Buckingham, UK) were used in the 1981 survey and pneumotachograph spirometers (Welch Allyn, Skaneateles Falls, New York, USA) were used in 1994/1995. 11 13 From 1981 onwards, spirometric measurements met American Thoracic Society guidelines (52% of the total number of measurements). 14 15 Demographic details and mean lung function measurements for this and replication cohorts are shown in table 1. Observations prior to age 25 were excluded from both discovery and replication analyses. 16 Genotype quality control and imputation Genotype data from BHS1 and BHS2 samples were merged and quality control (QC) was undertaken for both variant quality and sample quality. Prior to imputation, variants were excluded if they had a call rate <95%, deviated from Hardy-Weinberg equilibrium ( p<10 −6 ) or had a minor allele frequency (MAF) of <1% (100 240 variants). Individuals were excluded if their call rate was <95%, if their submitted gender and gender inferred by genotype were inconsistent, if they were a duplicate or if they were an outlier for heterozygosity (169 individuals). Principal components analysis was used to exclude individuals of non-European ancestry (those at least four SDs away from the mean for at least one of the first two principal components) (27 individuals). Analysis of identity by descent (IBD) was compared with the reported pedigree; any inconsistencies were reviewed and individuals were excluded as appropriate (40 individuals; see online supplementary material for further details). QC was undertaken using PLINK V. 1 22 Variants where imputation quality (r 2 ) was <0.3 or MAF <1% were excluded. Phenotype QC and analysis Data on age and smoking were checked for consistency over time. Where possible, missing data on height or smoking at one time point were imputed to be the same as that at the nearest time point (407 data items imputed in total). Otherwise, observations with missing data on lung function, age or height or where FEV 1 was greater than FVC were excluded from the analysis. After phenotype and genotype datasets were merged, and after samples failing genotype QC were excluded, 12 863 observations for 4170 individuals and up to eight time points were available prior to further phenotype QC. A time variable was created for each observation, determined by the difference between the age of the individual at that observation and their age at the first time point for which data were available on that individual. Family was defined based on IBD, such that all individuals in a family were related (IBD >0.2) with at least one other person in the family. A full model with age, age 2 , height, height 2 , sex and time as fixed effects and an intercept varying per individual, an intercept varying per family and a slope for time varying per individual as random effects was fitted to the data for each trait (FEV 1 , FEV 1 /FVC and FVC). Additionally, another four models were fitted, excluding the random intercept varying per family, the random slope for time varying per individual and age 2 and height 2 one at a time and comparing with the full model using Akaike information criterion and Bayesian information criterion. The final model included all the terms in the full model, except for age 2 as this did not improve the fit of the model (see online supplementary table S1). Outlying observations (with residuals more than four SDs away from the mean) for any of the three traits analysed, FEV 1 , FEV 1 / FVC and FVC, were then excluded from the analysis (168 observations and three individuals excluded). This resulted in 12 695 observations for 4167 individuals and up to eight time points included in the final analysis. The mean number of measurements per participant was three (see online supplementary figure S1) and the mean length of follow-up was 15.5 years. Genetic risk scores Previously published studies identified variants in 26 regions significantly associated with cross-sectional FEV 1 and/or FEV 1 / FVC. [5][6][7][8] We compared the effect size estimates for these 26 variants in BHS with the previously published effect size estimates. A weighted risk score was calculated for the joint effect of these 26 regions on baseline FEV 1 and FEV 1 /FVC as well as on change over time in both traits. The single-nucleotide polymorphisms (SNPs) included in this analysis are shown in the online supplementary table S2. Unbiased (winner's curse-free) effect sizes, as calculated previously but excluding any data from BHS, 8 were used as weights for the 26 variants in the risk score calculations (weights, range 0.2-2.57, provided in online supplementary table S3). The number of risk alleles for each variant was multiplied by its corresponding weight and then summed across variants in order to obtain the risk score for each individual and to create the risk score variable. This risk score variable was then added to the final model described above, together with the risk score by time interaction, in order to obtain both the effect on baseline and on change over time. Genome-wide association analysis and selection of variants for follow-up analysis The effect of individual genetic variants (which included SNPs and indels) on the rate of decline in FEV 1 , FEV1/FVC and FVC was tested by adding a variant and variant-by-time interaction into the final phenotypic model described above. Using this model, genome-wide association analyses were undertaken for 29 798 550 variants assuming an additive genetic effect. Genomic control was subsequently applied (genomic inflation factors for slope change were 1.04, 1.03 and 1.06 for FEV 1 , FEV 1 /FVC and FVC, respectively). 23 Figure 1 illustrates our criteria for selection of variants for follow-up analysis. We identified variants potentially associated ( p<5×10 −6 taken as suggestive of association) with longitudinal FEV 1 , FEV 1 /FVC or FVC. We defined regions of association around the most strongly associated variant (sentinel variant) ±500 kb. We examined region plots to assess support from neighbouring variants and cluster plots for the closest genotyped variant to the sentinel variant in order to rule out associations driven by genotyping errors. Regions were selected for follow-up if they had at least one variant (either the top variant or a proxy in linkage disequilibrium with r 2 >0.2) with imputation quality >0.7 and p value <5×10 −5 . Within each region, we selected variants for follow-up analysis as follows: (1) the sentinel variant, (2) a second variant in the region with imputation quality >0.7 and p value <5×10 −5 where the sentinel variant had imputation quality <0.7 and (3) for regions in which variants showed nominal interaction with smoking ( p <0.05 for interaction term based on a Z-test between ever-smokers and never-smokers), the most significant variant in the group (ever-smokers or never-smokers) where the sentinel variant's effect size estimate was largest. Sensitivity analyses were also undertaken for the regions which met the criteria for follow-up analysis to assess whether their effect could be mediated through smoking behaviour. The same model was fitted, with an additional term for smoking status (ever-smoked or never-smoked), and effect sizes were compared. Estimations of power were also obtained for the discovery of new signals and for detection of associations with the 26 variants previously reported to be associated with cross-sectional lung function (see online supplementary materials). Replication Follow-up analyses for the 39 new regions potentially associated with longitudinal change were undertaken in the Copenhagen City Heart Study (CCHS), a prospective study of a random sample of the Danish general population, aged ≥20 years, drawn using the Danish Civil Registration System (n=9016) who had lung function measurements at up to three time points between 1976 and 1994, 24 with a further measurement in 2001-03. 25 Follow-up was also undertaken in the Lung Health Study (LHS), a North American cohort of smokers with mild airflow limitation who had annual lung function measurements for 5 years (n=3502). 26 Risk score analyses with the previously reported 26 variants were also undertaken in CCHS. 5-8 QC procedures (including inspection of cluster plots) were applied and the same model was fitted as for BHS, but without adjustment for relatedness, given that there were no related individuals. Demographic details and mean lung function measurements for CCHS and LHS are shown in table 1. All risk score and association analyses were undertaken in R using the package 'lme4' (Bates D, Maechler M, Bolker B, et al; http://CRAN.R-project.org/package=lme4). Known regions: calculation of risk scores and replication Comparison of effect sizes for cross-sectional lung function and longitudinal change and weighted risk score analyses showed that 26 variants previously identified as associated with crosssectional lung function were not significantly associated with longitudinal change in our cohort. There was a strong correlation between estimated SNP effects on baseline lung function in BHS and in published estimates [6][7][8] for both FEV 1 (r=0.76) and FEV 1 /FVC (r=0.78). However, estimated SNP effects on change in FEV 1 and FEV 1 /FVC were weakly correlated with published estimates for the respective cross-sectional trait (figure 2 and online supplementary figure S2). A weighted risk score was calculated for the joint effect of these 26 regions on baseline FEV 1 and FEV 1 /FVC as well as change over time in both traits. This showed a strong effect on baseline FEV 1 ( p=9.75×10 −12 ) and FEV 1 /FVC (p=4.44×10 −16 ) but no effect was observed on change over time for either FEV 1 ( p=0.409) or FEV 1 /FVC (p=0.160) (table 2). In silico data were available for eight known variants in CCHS and genotyping was undertaken for the remaining 18 variants. All of these passed QC procedures. They showed a joint effect similar to that seen in BHS: a strong association with baseline measurement ( p=4.45×10 −10 ) but no association with change over time in FEV 1 /FVC ( p=0.302) and strong association with baseline measurement ( p=8.92×10 −7 ) but only borderline association with change over time for FEV 1 (p=0.030) (table 2). New signals: discovery and replication Genome-wide analysis identified 56 independent regions which were associated with decline in FEV 1 , FEV 1 /FVC and/or FVC at a significance threshold of p<5x10 −6 . Thirty-nine of these regions (48 variants) were selected for follow-up based on the criteria described under 'Methods': 11 regions (13 variants (rs6502247) showed an effect of 3.93 mL/year on decline in FEV 1 ( p=3.23×10 −9 ). Effect size estimates for these variants were not attenuated after adjusting for smoking status and no variant-by-smoking interaction met a Bonferroni-corrected threshold for the number of regions tested ( p<1.3×10 −3 ). De novo genotyping of all 48 variants was undertaken in CCHS. Where genotyping failed, tags were identified, if possible. In total, 34 variants in 30 independent regions passed QC procedures (including the inspection of cluster plots) and 14 failed. The analysis included 9016 individuals with up to four time points (25 796 observations). None of these 34 variants showed replicated association with change in lung function (using a significance threshold of p<0.0016, representing a Bonferroni correction for 30 independent regions with α=0.05) (see online supplementary table S4). Thirty-one variants (in 26 of the 39 regions) for which in silico data were available were also followed up in LHS. None of these SNPs showed significant association with lung function decline in LHS (using a significance threshold of p<0.0019, representing a Bonferroni correction for 26 independent regions with α=0.05) (see online supplementary table S5). Results from both CCHS and LHS are shown in the online supplementary materials. In total, 21 variants were analysed in both follow-up studies, of these, 14 had the same direction of effect as BHS after meta-analysing results from CCHS and LHS ( p=0.095). For sentinel variants in regions previously reported to show suggestive evidence of association with longitudinal lung function (though none met genome-wide significance and replicated in previous papers), we assessed association with FEV 1 , FVC and/or FEV 1 /FVC in our data. [26][27][28] Of these 51 variants,10 showed nominal evidence of association with at least one of the lung function traits either in the whole cohort or in one of the smoking subgroups. However, none were significant after correction for multiple testing (using a threshold of p<9.8×10 −4 ) (see online supplementary table S6). DISCUSSION We analysed the joint effect of regions previously associated with cross-sectional lung function on longitudinal change in the general population, and undertook a GWAS to identify new signals associated with longitudinal change. Regions previously identified as significantly associated with crosssectional FEV 1 and/or FEV 1 /FVC 5-8 were jointly strongly associated with baseline measurements in both discovery (BHS) and replication (CCHS) cohorts. However, although many of these variants have previously shown association with COPD risk (TNS1, RARB, FAM13A, GSTCD, HHIP, HTR4, ADAM19, AGER, GPR126, C10orf11, THSD4), [29][30][31][32][33] we have shown that they are not associated with change in FEV 1 or FEV 1 /FVC over time in our cohorts. We identified novel variants associated with decline in lung function in BHS which did not replicate in either a general population cohort (CCHS) or a cohort of smokers with mild lung function impairment (LHS). A key strength of our study design is the improved coverage of both common and low-frequency variants achieved through imputation in the 1000 Genomes Project reference panel, 20 compared with previous studies of longitudinal lung function which have used HapMap reference panels. 26 27 34 Another strength is the high number of lung function measurements in BHS: up to eight measurements over a range of up to 40 years. This is longer than other published studies, including most of the individual studies in a meta-analysis by Tang et al, 34 which included a portion of our dataset. Our study also adds value as it is the first to calculate risk scores for longitudinal lung function. Nevertheless, such a long follow-up period brings some challenges. Spirometry equipment changed over time and the earlier surveys in BHS were performed before protocols for standardisation of spirometry were published. 11 14 15 In addition, repeated measurements over a number of years could train participants in optimal technique and underestimate decline in lung function. The detection of known associations with crosssectional lung function provides some reassurance regarding the extent of any potential measurement error in lung function measurement. The biggest challenge we faced was the availability of large sample sizes for well-characterised longitudinal measures of lung function. The sample size, in combination with possible measurement error, may have limited our ability to identify individual variants which reach stringent genome-wide significant levels and replicate. However, our study will have had much greater power to detect longitudinal effects of aggregate risk scores comprised of the 26 variants previously reported to be associated with cross-sectional lung function. We did not examine risk scores for FVC, as at the time of the analyses there were no published associations with FVC, and our focus was on the determinants of obstructive lung disease. It is possible therefore that there are genetic associations with change in FVC over time which we did not identify. We did not undertake analyses stratified by sex, for reasons of power. However, effect estimates were adjusted for sex. We did not adjust for smoking in the primary analysis, since the analysis also had potential to highlight novel signals for smoking behaviour. We undertook sensitivity analyses to assess whether any top signals could be mediated by smoking behaviour. The signals were not attenuated after adjustment for smoking. However, this was based on binary smoking status and there remains potential for incomplete adjustment. Our study-the first to our knowledge to examine risk scores for change in lung function over time-complements existing studies which have sought individual SNP associations, [26][27][28] including a large meta-analysis of longitudinal lung function. 34 The meta-analysis by Tang et al 34 (concurrent with our study) incorporated data from 27 349 individuals from 14 populationbased cohorts (including a subset of 1009 individuals from BHS) and identified evidence for two novel regions associated with rate of change in FEV 1 . However, these did not replicate in two further cohort studies (including LHS). The authors noted that the number of lung function measurements and duration of follow-up varied considerably between studies included in the discovery phase and may have affected their ability to detect associations. The failure to replicate novel associations may relate to lack of power or, given that the larger replication cohort in their study (LHS) was composed of smokers with mild COPD, may indicate that the genetic determinants of lung function decline in those with COPD differ from those in healthy individuals. 34 Similarly, a GWAS examining change in FEV 1 % predicted (in a cohort of smokers with mild lung function impairment from LHS) identified two regions reaching genome-wide significance which did not replicate in four general population cohorts or in a cohort with moderate-to-severe COPD. The authors suggested that regions which modify the effect of cigarette smoke on lung function decline (or determine rate of decline at different stages of COPD) may be distinct from those which influence lung function decline in the general population. 26 A small number of studies have begun to explore interaction between genetic variants and smoking in relation to lung function decline, though these have not identified any significant associations which also replicated. 35 36 The suggestion that variants which determine lung function decline may show heterogeneity across different groups is further supported by findings from an earlier GWAS which identified suggestive evidence that different regions were associated with lung function decline in asthmatic and non-asthmatic individuals. Only the signal in non-asthmatic individuals (rs9316500 in DLEU7) showed evidence of replication (at p<0.05) in largely population-based cohorts, but it did not reach genome-wide significance in discovery, replication or twostage meta-analysis. 27 None of the top variants reported in these papers reached significance after correction for multiple testing in our data. Our findings are also consistent with recent work by Lange et al 37 which identified two distinct trajectories of FEV 1 in people who developed COPD. In their study, approximately half of those diagnosed with COPD by the end of follow-up started with normal lung function in early adulthood (mean age 40) and then showed rapid decline, whereas the remaining half started with low FEV 1 in early adulthood followed by a relatively normal rate of decline. This suggests that rapid lung function decline in later life is not necessary for development of COPD. We hypothesise that the known genetic variants examined in this paper may exert much of their effect in earlier life. Of the 26 regions examined in this paper, 19 have previously been shown to have directions of effect on lung function in children (aged 7-9 years) consistent with that in adults. 8 An additional study has also shown evidence of association with lung function as early as 5-14 weeks of age for variants in 4 of the 26 loci. 38 However, further large GWAS of lung development are required to test this hypothesis. These findings emphasise the continuing public health importance of focusing on the key environmental determinants of lung function decline, particularly smoking. Nevertheless, genetic determinants of decline may remain a therapeutic target in the half of people with COPD for whom accelerated decline is important in pathogenesis. 37 A potential strategy to identify these would be to focus on older cohorts and adjust for the effect of all known variants which affect cross-sectional lung function. Large sample sizes will also be the key to help confirm or refute our findings. However, as the largest existing meta-analysis identified significant challenges posed by phenotypic heterogeneity, ensuring comparability of the participating studies' approach to measuring longitudinal change must also be a high priority. An alternative approach would be to study longitudinal lung function in large, more homogeneous populations. In summary, regions previously identified as significantly associated with cross-sectional FEV 1 and/or FEV 1 /FVC 5-8 were jointly strongly associated with baseline measurements in both discovery and replication cohorts but were not associated with change in FEV 1 or FEV 1 /FVC over time. The present study and others to date have identified no regions associated with lung function decline which reach genome-wide significance and replicate in independent cohorts. Genetic variants identified to date that influence cross-sectional lung function, while still relevant in predicting the risk of COPD, appear to have Betas provided correspond to the risk allele. Risk allele here is defined as the allele associated with decreased lung function in BHS. Variants are given in order of Chr and position. β, per-allele change in FEV 1 , FVC or FEV 1 /FVC; Chr, chromosome. little or no effect on the rate of change in adult lung function over time in our study.	2017-07-31T19:24:11.810Z	2017-02-07T00:00:00.000	{ "year": 2017, "sha1": "bd255253a4c0d03aa09bb19c01a575b6f8c29913", "oa_license": "CCBY", "oa_url": "https://thorax.bmj.com/content/thoraxjnl/72/5/400.full.pdf", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "bd255253a4c0d03aa09bb19c01a575b6f8c29913", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
17214079	pes2o/s2orc	v3-fos-license	Asian-variant intravascular lymphoma in the African race Intravascular Large B-cell lymphoma (IVLBCL) is an exceptionally rare form of non-Hodgkin lymphoma (NHL) distinguished by the preferential growth of neoplastic cells within blood vessel lumen. Challenging to detect and deemed disseminated at diagnosis, this condition is characterized by a highly aggressive, inconspicuous course with a high mortality rate. We describe the case of a 48 year-old African-American female presenting with a two month history of low-grade fevers and malaise. Laboratory data was notable for anemia, thrombocytopenia, elevated liver function tests, and hematuria. An extensive work-up for infectious, rheumatologic and malignant causes was negative. Her symptoms progressed and within two weeks, she was admitted for disseminated intravascular coagulation (DIC). Her course was complicated by diffuse pulmonary hemorrhage and ultimately, care was withdrawn. Autopsy identified widespread CD-20 positive intravascular large B-cell lymphoma with significant hepatosplenic involvement, characteristic of the Asian variant IVLBCL. This case uniquely highlights development of the Asian variant IVLBVL in a previously undescribed race. Identified by its intraluminal vascular growth pattern, IVLBCL generally spares lymphatic channels. Diagnosis and differentiation of this condition from other hematological malignancies via skin, visceral and bone marrow biopsy is imperative as anthracycline-containing chemotherapies may significantly improve clinical outcomes. This article outlines the common presentation, natural course, and treatment options of IVLBCL, along with the histopathology, immunohistochemistry, and chromosomal aberrations common to this condition. Intravascular Large B-cell lymphoma (IVLB-CL) is an exceptionally rare form of non-Hodgkin lymphoma (NHL) distinguished by the preferential growth of neoplastic cells within blood vessel lumen. Challenging to detect and deemed disseminated at diagnosis, this condition is characterized by a highly aggressive, inconspicuous course with a high mortality rate. We describe the case of a 48 year-old African-American female presenting with a two month history of low-grade fevers and malaise. Laboratory data was notable for anemia, thrombocytopenia, elevated liver function tests, and hematuria. An extensive workup for infectious, rheumatologic and malignant causes was negative. Her symptoms progressed and within two weeks, she was admitted for disseminated intravascular coagulation (DIC). Her course was complicated by diffuse pulmonary hemorrhage and ultimately, care was withdrawn. Autopsy identified widespread CD-20 positive intravascular large Bcell lymphoma with significant hepatosplenic involvement, characteristic of the Asian variant IVLBCL. This case uniquely highlights development of the Asian variant IVLBVL in a previously undescribed race. Identified by its intraluminal vascular growth pattern, IVLBCL generally spares lymphatic channels. Diagnosis and differentiation of this condition from other hematological malignancies via skin, visceral and bone marrow biopsy is imperative as anthracycline-containing chemotherapies may significantly improve clinical outcomes. This article outlines the common presentation, natural course, and treatment options of IVLBCL, along with the histopathology, immunohistochemistry, and chromosomal aberrations common to this condition. Case Report In June of 2008, a 48 year-old African-American woman presented to her primary care physician describing a three week history of low-grade fevers and malaise. Antibiotics were administered, with no resolution of symptoms. She was subsequently admitted for further work-up. On admission, her laboratory data was notable for anemia and thrombocytopenia. Initial testing included a broncheoalveolar lavage (BAL) and lumbar puncture, which were non-diagnostic. A liver biopsy was notable for a mononuclear infiltrate consistent with chronic hepatitis. Laboratory data was notable for a normal Coombs test, autoimmune and vasculitic panel, and hepatitis panel. Computed tomography of the chest, abdomen and pelvis (CT-CAP) was unremarkable. She was discharged one month later with no formal diagnosis. Within two weeks, she was re-admitted for progression of symptoms. Admission laboratory data was notable for severe anemia (Hgb 6.0 g/dL), leukopenia (2.5 g/dL); thrombocytopenia (10¥10 9 ), and an elevated lactate dehydrogenase (LDH; 1900 U/L; reference range 122-222 U/L). Aspartate transaminase (AST; 77 U/L), alanine aminotransferase (ALT; 46 U/L), total bilirubin (8.8 mg/dL) and direct bilirubin (6.2 mg/dL) were all noted to be elevated. Haptoglobin was low. Prothrombin time (PT), activated partial thromboplastin time (APTT) and d-dimer levels were elevated. A peripheral blood smear showed schistocytes and polychromasia. Findings were deemed consistent with microangiopathic hemolytic anemia and disseminated intravascular coagulation (DIC). She was treated with plasma exchange and high dose steroid administration (methylprednisolone 125 mg/day) without effect. An extensive work-up for viral, bacterial and fungal pathogens was again, non-revealing. Magnetic resonance cholangiopancreatography (MRCP) of the liver showed hepato splenomegaly with intra-organ hemosiderin deposition. Repeat CT-CAP defined diffuse hazy opacities in both lungs, hepato splenomegaly and mild thickening of the walls of the cecum and ascending colon ( Figure 1). On repeat CT of the chest three days later, there was interval increase in the bilateral nodular opacities, air bronchograms, subcutaneous anasarca and findings consistent with pulmonary edema ( Figure 2). Bronchoalveolar lavage (BAL) identified diffuse pulmonary hemorrhages. On hospitalday 5, the patient rapidly developed acute respiratory distress syndrome (ARDS) and pulseless electrical activity (PEA). Despite successful resuscitation, she remained unresponsive and care was ultimately withdrawn. Bone marrow biopsy results performed earlier that week were available following her demise. Samples showed evidence of hyercellular (80-90% cellular) with trilineage hematopoiesis and a subtle atypical mononuclear cell infiltrate composed of large cells with slightly irregular nuclear contours and multiple visible nucleoli. An autopsy identified diffuse alveolar hemorrhage, bilateral renal congestion with acute tubular necrosis (ATN), hepatomegaly and a single enlarged perisplenic lymph node. Histological examination of multiple organs confirmed diffuse intravascular large B-cell type lymphoma (IVLBCL). Histologic andimmunohistochemical findings Histological examination showed capillary wall congestion with large; atypical lymphocytes ( Figure 3). On higher magnification, lymphocytes displayed vesicular nuclei, prominent nucleoli and frequent mitotic figures ( Figure 4). Some cells showed coarse nuclear chromatin or irregular and indented nuclei ( Figure 4). Staining with common leukocyte antigen and B cell marker CD-20 exhibited strong uptake by neoplastic cells ( Figure 5). Discussion First characterized by Pfleger and Tappeiner in 1959, 1 IVLBCL has emerged as an elusive malignancy characterized by a covert presentation and highly aggressive course. It is defined by the World Health Organization (WHO) as a rare and highly aggressive lymphoma variant, identified by the proliferation of large tumor cells within the lumen of small to mediumsized vessels. 2 We present the first reported case of the Asian variant IVLBCL in an African-American individual and proceed to review its unique presentation, progression and response to treatment. Since its first description more than half-acentury ago, literature on IVLBCL has continued to accumulate. Until recently, the relative paucity of cases has limited epidemiologic research. However, it is currently recognized as a disease primarily of the elderly, with a median age in the sixth decade of life and a male-to-female ratio exceeding 3:1. 3 The wide variability in IVLBCL's clinical presentation makes detection challenging with the majority of patients present late within the disease process. Despite variability in presentation, specific patterns have been successfully traced to a surprising common entity: the geographic distribution of patients. Appropriately reflecting their origins, two main subtypes IVLBCL are recognized: the western variant and the Asian variant. The western or classical variant remains the most common sub-type identified in North America and Europe. It is characterized by a prominence of neurologic and dermatologic findings, both at initial presentation and throughout its course. Neurologic symptoms may include alterations in consciousness and peripheral neuropathy. Cutaneous features may manifest as telangiectasias, diffuse erythematous streaks, purpuric macules and nodules. 4 The cutaneous variant of the western variant class is a unique subtype which is less aggressive than all other forms of IVLBCL, most commonly affecting pre-menopausal females. The Asian or hemophagocytic variant, like it's counterpart, has a title reminiscent of its origin and is characterized by anemia (66%), thrombocytopenia (58%), hemophagocytosis (61%), bone marrow involvement (75%), respiratory symptoms (34%), B-symptoms (fever, night-sweats; 76%), hypoalbuminaemia (<30 g/L; 61%) and hepatosplenomegaly. 5,6 In contrast to the Western variant, neurological symptoms (27%) and cutaneous lesions (15%) are rare and, when present, suggest an alternative diagnosis. 5 High lactate dehydrogenase (LDH; 98%) and disseminated intravascular coagulation (DIC; 25%) are common features. 3,6 Most Asian variant IVLBCL cases have been reported in Southwestern Japan. A few case reports have described the hemophagocytic variant within the Western hemisphere. However, the majority of these have involved either immigrants from eastern countries or a known family history of Asian descent. Our case uniquely describes an African-American within the western hemisphere, presenting with the Asian variant. Theories regarding the source of this the observed geographical difference remain disputed. More recent conjectures have included potential ethnic differences involved in inflammatory marker production such as sIL2R which appears markedly increased in Asian variant patients, in comparison to western variants. 3 Other theories speculate towards an environmental component given that the majority of Asian variant cases have been documented in Southwestern Japan where human T-cell lym-photropic virus type-1 (HTLV-1) is prominent. Helminthic infectious may also represent an environmental trigger. In contrast to other hemophagocytic-associated lymphomas, Epstein-Barr virus (EBV) has been reported in only a few IVLBCL cases and is not thought be causative. The genetic alterations in IVLBCL remain under investigation. Aberrations in chromosomes 1, 6, 18, 11 and 13 have all been postulated as sources. 3,7-11 However, in a retrospective analysis of cytogenetic abnormalities, Shimada et al. identified that only 57% of 84 patients displayed cytogenetic abnormalities. 3 Determination of the progenitor cell lines remains an area of intense investigation. Most cells express CD 19, CD20, and CD79a, consistent with B-cell origin. In addition to B cells, both intravascular T cell and natural killer (NK) cell lines have been identified and should not be considered a variant of IVLBCL. 12 It is plausible that Asian variant IVLBCL represents a unique variant of the germinal center B cell-like (GCB) form of CD5 + Diffuse Large B Cell Lymphoma (DLBCL). 5,13 Accordingly, CD5lines may be indicative of the western variant. The paucity of extravascular migration remains IVLBCL's most perplexing feature. Defective interactions between luminal walls and lymphoma cells may play a role. 14 In healthy individuals, lymphocytes bind to high endothelial venules (HEV) on vessel walls via the lymphocyte homing receptor, thereby allowing transverse movement across the walls into peripheral regions. Western variant IVLB-CL, in particular, appears to congregate in regions with low HEV, such as the brain and skin. 14 HEV deficiency within vessel lumens may thus prevent extravasation into the periphery. Other possibilities include deficien-Case Report cies in essential transvascular migration molecules such as CD29 (b1-Integrin), CD54 (intracellular adhesion molecule 1), and CD18 (Integrin b2) superficially expressed by lymphoma cells. 15,16 In the Asian variant, the mechanism of hemophagocytosis remains poorly understood, but may be influenced by macrophage colony stimulating factor and IL-6. 9 It is possible that hemophagocytosis is linked to cellular extravasation as studies have shown that the presence of hemophagocytosis correlates with disease dissemination (along with B-symptoms and elevated levels of soluble IL-2 receptor). 5 Notably, although the majority of IVLBC cases involve small vessels, at least seven case reports have identified IVLBCL in large vessels. 17 The mechanism of this is unknown. As stated previously, IVLBCL may present with wide clinical variability. Characteristic physical exam findings associated with lymphoma, such as lymphadenopahthy or tumor burden, are often absent thus requiring greater reliance on laboratory data. Suspicion should be raised in patients describing headaches, skin abnormalities, hepatosplenomegaly, or constitutional complaints. Initial laboratory tests should include a complete blood count, serum LDH, b-2 microglobulin, monoclonal serum protein, a complete hepatic and renal panel and thyroid function studies. Cytogenetic studies should also be considered. The gold standard of diagnosis remains organ biopsy. Diffuse spread of the disease allows for physician discretion in determining the biopsy site. For Asian IVLBCL, bone marrow (BM) biopsy offers the most reliable results as intrasinusoidal tumor involvement is frequently encountered (up to 96%). 6,9 Serial BM biopsies have been shown to enhance diagnostic accuracy. 18 Despite infrequent cutaneous features, Asian-variant IVLBCL is commonly isolated from skin biopsies and if positive, may prove a simple and useful diagnostic procedure. 6 Transbronchial biopsy may be performed if pulmonary symptoms are present. Hepatic and splenic biopsies are useful in the setting of hepatosplenomegaly and show sinusoidal involvement in up to 89% of patients. 9 Tumor cells are large with scant basophilic or amphophilic cytoplasm and moderately dispersed chromatin. Frequent mitotic figures and prominent vesicular nucleoli are common. Cells are rarely detected in peripheral blood. Sinusoidal distribution often occurs within the bone marrow, liver and spleen. As mentioned, erythrocyte-hemophagocytic cells are detected in the vast majority of BM biopsies. However, in the appropriate clinical context, the presence of classic findings including thrombocytopenia, fever and hepatosplenomegaly may be considered diagnostic even in the absence of hemophagocytosis. Computed tomography (CT) scans are rarely diagnostic. However, reports regarding the use of 18-Fluorodeoxyglucose-Positron Emission Tomography (FDG-PET) have been promising with characteristic uptake in the bone marrow and involved organs. 19,20 The overall prognosis for most IVLBCL cases remains poor. Historically, most patients have been diagnosed with stage IV disease, or post-mortem. Post-mortum diagnosis appears less common in Asian variants (36%) versus the Western variants (64%). 5 Prognostic comparisons between the Western and Asian variants are similar, as are responses to treatment. 5 The rarity of IVLBCL has made randomized controlled trial assessment difficult. Historically, combination chemotherapy, typically CHOP, has been utilized with dismal results. However, the addition of Rituximab, a monoclonal antibody to CD-20, has led to substantial improvements in recovery, nearly doubling survival length. 21 European studies have determined an 81% survival rate at 3 years with combined immunochemotherapy, 22 versus just 33% in the absence of rituximab. 23 To date, no studies have evaluated the addition of Rituximab specifically in the Asian variant IVLBCL. Autologous stem cell support (ASCT) has been evaluated in a number of small studies and shown efficacy, primarily in individuals undergoing their first remission. 3,23 However, the advanced age of most patients excludes them from this option. Allogenic peripheral blood stem cell transplant has been utilized with variable results in the setting of failed chemotherapy. 24 With treatment, Asian variant IVLBCL survival times range 12.5 months to 22.5 months. 21,25 Cases exclusive to the skin appear to have a better response rate (3-year survival 56%). 26 Median survival with anthracycline therapy is estimated to be 13 months. 5 Etoposide has been effective in the setting of hemophagocytic syndrome. 27 Conclusions Despite more than 50 years of research, IVL-BCL remains a highly elusive malignancy typified by limited treatment options and a gaurded prognosis. We describe the first recorded occurrence of Asian variant IVLBCL in an African-American within the western hemisphere. This case is of particular interest as descriptions of the Asian variant IVLBCL in non-Asian populations is limited. No obvious mechanism has yet surfaced explaining this unique phenomenon. However, environmental factors characteristic to certain topographies may play a lesser role than once previously considered. Alternatively, African and Asian races may share particular genomic sequences predisposing them to the Asian variant. Regardless of the cause, it remains imperative that continued research be promoted to better understand this unique malignancy.	2014-10-01T00:00:00.000Z	0001-01-01T00:00:00.000	{ "year": 2012, "sha1": "cea9791f0948030d18671996f0de292df377b71d", "oa_license": "CCBYNC", "oa_url": "https://journals.sagepub.com/doi/pdf/10.4081/rt.2012.e10", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "cea9791f0948030d18671996f0de292df377b71d", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [] }
13858631	pes2o/s2orc	v3-fos-license	Autophagy counteracts apoptosis in human multiple myeloma cells exposed to oridonin in vitro via regulating intracellular ROS and SIRT1. AIM To explore the mechanisms underlying the oridonin-induced apoptosis and autophagy in human multiple myeloma cells in vitro. METHODS Human multiple myeloma RPMI8266 cells were used. The cell viability was assessed using MTT assay. Morphological changes of apoptosis and autophagy were observed under transmission electron microscope. TUNEL and annexin V-FITC/PI dual staining assays were used to measure apoptosis. Autophagy was analyzed using Western blot analysis and immunofluorescence staining with a QDs(605 nm)-Anti-LC3 fluorescent probe. Intracellular ROS was estimated with flow cytometry using DCFH-DA fluorescent probe. Protein levels of active caspase 3, Beclin 1 and SIRT1 were determined with Western blot analysis. RESULTS Exposure to oridonin (1-64 μmol/L) inhibited the proliferation of RPMI8266 cells in a concentration-dependent manner with an IC(50) value of 6.74 μmol/L. Exposure to oridonin (7 μmol/L) simultaneously induced caspase 3-mediated apoptosis and Beclin 1-dependent autophagy of RPMI8266 cells. Both the apoptosis and autophagy were time-dependent, and apoptosis was the main effector pathway of cell death. Exposure to oridonin (7 μmol/L) increased intracellular ROS and reduced SIRT1 nuclear protein in a time-dependent manner. The blockade of intracellular generation of ROS by NAC (5 mmol/L) abrogated apoptosis, autophagy and the decrease of SIRT1 in the cells exposed to oridonin (7 μmol/L). The inhibition of autophagy by 3-MA (5 mmol/L) sensitized the cells to oridonin-induced apoptosis, which was accompanied by increased intracellular ROS and decreased SIRT1. CONCLUSION Oridonin simultaneously induces apoptosis and autophagy of human multiple myeloma RPMI8266 cells via regulation of intracellular ROS generation and SIRT1 nuclear protein. The cytotoxicity of oridonin is mainly mediated through the apoptotic pathway, whereas the autophagy protects the cells from apoptosis. Introduction Oridonin, an active diterpenoid compound isolated from Rabdosia rubescens, has various pharmacological and physiological effects (eg, anti-tumor, anti-inflammation, and anti-bacterial), and it has been widely used for the treatment of various human diseases [1][2][3] . Both apoptosis and autophagy are essential cellular homeostatic mechanisms, which are important in multicellular organisms for development, tissue turnover, and host defense. Autophagy is a conserved, genetically controlled process that leads to the degradation of cytoplasmic components within lysosomes and has recently gained much attention for its paradoxical relationship with apoptosis. Some studies suggest that the inhibition of autophagy enhances apoptosis. In contrast, others have suggested that autophagy acts as a cell death pathway, termed programmed cell death (PCD) type II, in cooperation with apoptosis [4][5][6][7] . Although oridonin has been shown to induce apoptosis and autophagy in some types of tumor cells in both in vitro and in vivo studies [8][9][10] , the relationship between the two processes is unclear. Furthermore, the molecular mechanisms underlying oridonininduced apoptosis and autophagy in RPMI8266 cells remain to be determined. Multiple myeloma (MM) is an untreatable hematological disease characterized by the synthesis of excess immunoglobulin (Ig), which forms endoplasmic reticulum-localized unfolded or misfolded proteins that are potentially toxic to MM cells. Hoang et al [11] found that the level of autophagy that occurs in MM cells was significantly higher than that in normal plasma cells. Additionally, the inhibition of autophagy www.nature.com/aps Zeng R et al Acta Pharmacologica Sinica npg in MM cells using chloroquine or Beclin 1-siRNA leads to cell death. This observation suggests that autophagy has a protective effect on MM cell viability. Furthermore, the proteasome inhibitor bortezomib, a clinical drug used for the treatment of myeloma, can induce MM cell death and autophagy. However, the treatment of MM cells with an autophagy inhibitor in combination with bortezomib resulted in an antagonistic response in vitro [11] . Consequently, the relationship between autophagy and cell death remains complicated and requires further investigation in multiple myeloma. Reactive oxygen species (ROS) are generally small, shortlived and highly reactive molecules. While ROS generation is a consequence of basal cellular respiration, increased ROS generation is associated with several pathological conditions (eg, hypoxia, ischemia, and anti-tumor agents). Some studies have demonstrated that ROS generation activated caspase cascades through the mitochondrial permeability transition to mediate apoptosis [12,13] . In addition to apoptosis, ROS generation has recently been reported to mediate autophagy under certain conditions [14][15][16] . Sirtuin1 (SIRT1) is a NAD + -dependent deacetylase that is involved in a diverse set of physiological functions, including gene silencing, stress resistance, apoptosis, inflammation, senescence, and aging. Studies have indicated that SIRT1 activity can be positively or negatively regulated by intracellular oxidative stress [17][18][19] . SIRT1 was suggested to have anti-apoptotic functions by a wide range of in vivo and in vitro studies [20,21] . However, SIRT1 was shown to promote autophagy [22,23] . Therefore, we hypothesized that the generation of intracellular ROS and SIRT1 activity could underlie the effects of oridonin treatment in RPMI8226 cells. Methods and materials Reagents and antibodies Oridonin, thiazolyl blue tetrazolium bromide (MTT), the 2', 7′-dichlorofluorescein diacetate (DCFH-DA) fluorescent probe, dimethyl sulfoxide (DMSO), the 3-methyladenine (3-MA) autophagy inhibitor and the N-acetylcysteine (NAC) free radical scavenger were purchased from Sigma-Aldrich. The purity of oridonin was confirmed by HPLC to be greater than 99%. Oridonin was dissolved in DMSO to make a stock solution. The DMSO concentration was kept maintained 0.1% in all cell cultures, and it did not exert any detectable effect on cell growth or cell death. Anti-active caspase 3 (Abcam, ab2302) and anti-LC3 (Abcam, ab48394) were purchased from Abcam. Anti-Beclin 1 (sc11427) and anti-SIRT1 (sc74504) were purchased from Santa Cruz Biotechnology. Cell culture and treatments Human multiple myeloma RPMI8226 cells were purchased from American Type Culture Collection (ATCC). The cells were maintained in RPMI-1640 medium (GIBCO, 31800-022) supplemented with 10% fetal bovine serum (FBS) (TBD Biotechnology Development, TBD0022HLY) without antibiotics at 37 ºC in a 5% CO 2 humidified atmosphere. After the cells reached a steady-state of exponential growth in normal media, they were exposed to oridonin for 0, 6, 12, or 24 h prior to the analysis. To inhibit intracellular ROS generation and autophagy, cells were pre-incubated with NAC or 3-MA, respectively, at a concentration of 5 mmol/L for 1 h prior to oridonin treatment. MTT assay A 100-µL suspension of RPMI8226 cells were seeded on 96-well plates with or without oridonin at various concentrations (1,2,4,8,16,32, and 64 µmol/L) at a density of 1×10 5 cells per well. After incubation for a designated period of time, MTT was added to each well at a final concentration of 0.5 mg/mL for 3 h, and the resulting formazan crystals were dissolved in DMSO. Optical density was measured at 490 nm with background subtraction at 630 nm using a plate microreader (TECAN SPETRA). The growth inhibitory ratio was calculated as follows: Transmission electron microscopy (TEM) analysis After treatment, cell pellets were fixed with 2.5% glutaraldehyde in 0.1 mol/L cacodylate buffer, pH 7.4 at 4 ºC for at least 30 min. After fixation, the specimens were thoroughly washed in 0.1 mol/L cacodylate buffer and then fixed with 1% osmium tetroxide in the same buffer at room temperature (RT) for 1 h. The specimens were dehydrated in a graded series of ethanol and then embedded in Epon. Thin sections (0.1 µm) were cut, stained with uranyl acetate/lead citrate and viewed using a Hitachi H-300 TEM. Analysis of apoptosis using the TUNEL assay and FCM of AV/PI dual staining In this study, several approaches were used to detect apoptosis quantitatively and qualitatively, including (I) the terminal deoxynucleotidyl transferase mediated X-dUTP nick end labeling (TUNEL) assay and (II) annexin V-FITC (fluorescein isothiocyanate)/PI (propidium iodide) staining for FCM. The TUNEL assay was performed using a commercial kit (BOSTER Biological Technology, MK1020) according to the manufacturer's protocol. Briefly, 1×10 6 cells/mL that were treated with 7 µmol/L oridonin for 0 h or 24 h were collected and fixed in 4% paraformaldehyde at 4 ºC. The fixed cells were then incubated with the TUNEL reaction mixture for 1 h at 37 ºC, followed by the addition of a peroxidase-conjugated detection antibody. DNA fragments were stained using diaminobenzidine (DAB) as a substrate for the peroxidase. Positive staining was identified using a light microscope as brown granules. The apoptosis rate was calculated as follows: apoptotic rate (%)=number of positively stained cells/total number of cells×100% (at least 500 cells were counted under a light microscope). For annexin V-FITC/PI dual staining, cells were processed with an Annexin V-FITC kit (Keygene, KGA108) following treatment according to the manufacturer's instructions. Next, the samples were analyzed using the FACScan flow cytometer www.chinaphar.com Zeng R et al Acta Pharmacologica Sinica npg (Becton Dickinson) to quantify the apoptotic rate. Different subpopulations were distinguished using the following criteria: Q1, annexin V-negative, but PI-positive (ie, necrotic cells) indicating autophagic cell death in this study; Q2, annexin V/ PI-double positive (ie, late apoptotic cells); Q3, annexin V/ PI-double negative (ie, live cells); Q4, annexin V-positive, but PI-negative (ie, early apoptotic cells). The apoptotic rate was determined as the percentage of Q2+Q4. Immunofluorescence analysis of LC3 localization using a QDs 605 nm (quantum dots 605 nm)-Anti-LC3 fluorescent probe To prepare the QDs 605 nm -Anti-LC3 fluorescent probe, coreshell QDs (ZnS-capped CdSe) were synthesized by the College of Chemistry and Molecular Sciences, Wuhan University. We used a 1.5-mL solution of high-quality oil-soluble coreshell QD 605 nm to synthesize water-soluble QDs according to a previously developed procedure [24] . These activated dots modified with thioglycolic acid were dissolved in PBS (0.08 mol/L, pH 7.4) containing 50 mmol EDC (1-(3-dimethyl-aminopropyl)-3-ethylcarbodiimide hydrochloride, >98%) and 5 mmol NHS (N-hydroxy-succinimide, >98%). Next, the QDs were incubated with 20 µL of a mouse anti-human monoclonal LC3 antibody at RT in a shaking incubator for 2-4 h. The final QD bioconjugates were purified by centrifugation at 6000×g for 10 min, and the suspension was dialyzed for 8-12 h. The resulting QD 605 nm -Anti-LC3 probes were stored at 4 ºC. For the immunofluorescence analysis, cells were collected following treatment and fixed in 4% paraformaldehyde for 1 h at 4 ºC. Next, the fixed cells were immobilized on a gelatin-covered (0.1% gelatin and 0.01% chromium potassium sulfate) slide and dried under sterile conditions at RT for 1 h. The specimen was permeabilized in phosphate buffer solution (PBS) containing 0.1% Triton X-100 and sodium citrate at RT for 10 min. Then, the specimen was incubated with the QD 605 nm -Anti-LC3 probes at a final concentration of 1×10 7 mol/L at RT for 4 h. After incubation, the slides were washed three times with PBS and observed using fluorescence microscopy (Olympus) with an excitation wavelength of 605 nm to determine LC3 localization. Under normal conditions, LC3-II is uniformly distributed and has a diffuse localization pattern. During autophagy, LC3-I is processed to LC3-II and translocates to autophagosome membranes, which appear as red fluorescent punctae. Because LC3-II localization is used as a marker for autophagy, the percentage of fluorescent punctae-positive cells compared with the total number of cells was calculated as follows to quantify the amount of autophagy: the percentage of fluorescent punctae-positive cells (%)=(the number of fluorescent punctae-positive cells/the total number of cells)×100% (at least 500 cells were counted using a fluorescence microscope). Measurement of intracellular ROS generation Intracellular ROS generation was estimated by FCM using the DCFH-DA fluorescent probe. Briefly, the treated cells were collected and washed twice with RPMI-1640. The level of intracellular ROS was determined by incubating the cells with the DCFH-DA working solution (25 µmol/L) at 37 ºC for 30 min. After incubation, cells were washed twice with RPMI-1640 and then analyzed using FCM to determine the DCF fluorescence intensity at excitation and emission wavelengths of 488 nm and 525 nm, respectively. Protein extraction and Western blot analysis Total cellular protein was harvested by washing cells with ice-cold PBS and incubating them in lysis buffer (10 mmol/L Tris, pH 7.4, 20 mmol/L NaCl, 5 mmol/L MgCl 2 , 0.5% NP-40, and 0.1 mmol/L PMSF). The extracts were centrifuged, and the clear supernatants containing total protein were collected. Cellular nucleic proteins were extracted using a commercial kit (DBI Bioscience, DBI1017) according to the manufacturer's protocol. After isolation, the protein concentration was determined using the Bio-Rad protein assay, and an equal amount of protein was subjected to SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Hybond-C extra, GE Healthcare Life Sciences, RPN203E). After blocking with 5% non-fat milk, the membranes were probed with the designated primary and secondary antibodies, developed with the enhanced chemiluminescence method (Pierce, 32106) and visualized using the Kodak Image Station. The band density was quantified using the Image J image processing program. Because the extent of LC3 conversion is correlated with the level of autophagy [25] , LC3-I and LC3-II were detected by Western blot analysis. Statistical analysis All data are presented as mean±standard deviation (SD). Statistical significance was determined using Student's t-test with P-values <0.05 representing significance. Results Oridonin inhibits the proliferation of RPMI8226 cells RPMI8226 cells were treated with oridonin at various concentrations (1,2,4,8,16,32, and 64 µmol/L) for 6, 12, or 24 h. As shown in Figure 1, oridonin induces remarkable inhibition of cell proliferation in a time-and dose-dependent manner with an IC 50 of 6.74 µmol/L in RPMI8226 cells at 24 h. The TUNEL assay, FCM analysis of annexin V-FITC/PI dual staining and TEM were performed to detect apoptosis in cells treated with 7 µmol/L oridonin for 0, 6, 12, or 24 h. As shown in Figure 2A, a significant increase in the number of TUNELpositive apoptotic cells was detected in cells treated with oridonin for 24 h. The apoptotic rate of cells exposed to oridonin was increased to 41.4%±1.42% while the apoptotic rate of the control group was only 5.78%±0.56%. Annexin V-FITC/PI dual staining demonstrated that oridonin induces a significant time-dependent increase in the apoptotic rate [(Q2+Q4)%]. As shown in Figure 2C, the apoptotic rates of cells at the indicated time points were 5.00%±1.55% (0 h), 10.47%±0.68% (6 h), 14.70%±0.60% (12 h), and 17.53%±0.68% (24 h). Apoptotic morphology was also observed using TEM in cells treated with oridonin for 24 h ( Figure 3B). To explore the mechanism underlying the time-dependent increase in apoptosis induced by oridonin, the level of active caspase 3 was determined using Western blot analysis of cells treated with 7 µmol/L oridonin for 0, 6, 12, or 24 h. As shown in Figure 2B, oridonin also results in a time-dependent activation of caspase 3. These results indicate that oridonin induces caspase 3-mediated apoptosis in RPMI8226 cells. Oridonin induces Beclin 1-mediated autophagy in RPMI8226 cells Detection of autophagosomes, in addition to LC3 conversion and localization in cells treated with 7 µmol/L oridonin for 0, 6, 12, or 24 h was performed using TEM, Western blot analysis, and immunofluorescence using the QDs 605 nm -Anti-LC3 fluorescent probe, respectively. There were a greater number of autophagosomes, which is indicative of autophagy, in cells exposed to oridonin for 24 h as compared with untreated controls ( Figure 3C). LC3-II was detected using Western blot analysis. As shown in Figure 4A, the level of LC3-II protein increased in a time-dependent manner after treatment with oridonin. As shown in Figure 4B, oridonin induces a remarkable increase in red fluorescent punctae at 6, 12, and 24 h, which is indicative of LC3-II localization in autophagosomes in cells. The percentage of red fluorescent punctae-positive cells increased in a time-dependent manner during treatment with oridonin for 24 h. The percentages of red fluorescent punctae-positive cells at the indicated time points were 14.33%±4.04% (0 h), 35.67%±4.16% (6 h), 61.00%±3.61% (12 h) and 91.00%±2.65% (24 h) ( Figure 4B). These results suggest that oridonin induces autophagy in a time-dependent manner in RPMI8226 cells. Because Beclin 1-independent autophagy has been reported, the levels of Beclin 1 in cells treated with 7 µmol/L oridonin for 0, 6, 12, or 24 h were determined using Western blot analysis to investigate its role in autophagy induced by oridonin. As shown in Figure 4C, Beclin 1 expression followed the same time-dependent pattern as LC3 conversion and localization. These data indicate that autophagy induced by oridonin is canonical Beclin 1-mediated macro-autophagy. Intracellular ROS generation mediates apoptosis and autophagy induced by oridonin through negative regulation of SIRT1 activity To investigate whether ROS and SIRT1 are involved in oridonin-induced apoptosis and autophagy, the levels of intracellular ROS and SIRT1 nuclear protein in cells treated with 7 µmol/L oridonin for 0, 6, 12, or 24 h were assessed by the FCM analysis of DCF fluorescence intensity and Western blot analysis, respectively. As shown in Figure 5A, oridonininduced apoptosis and autophagy, was associated with a time-dependent increase in DCF fluorescence intensity. DCF fluorescence intensities at the indicated time points were 23154.67±1332.90 (0 h), 28150.00±716.45 (6 h), 32574.33±1908.46 (12 h), and 44410.67±2478.12 (24 h). In contrast, oridonin treatment results in decreased nuclear SIRT1 protein in a timedependent manner ( Figure 5B). These results suggest that the induction of apoptosis and autophagy by oridonin may be positively regulated by intracellular ROS generation and negatively regulated by SIRT1 activity. NAC, a general free radical scavenger, was used to block ROS generation to further confirm the roles of intracellular ROS generation and SIRT1 activity in the induction of apoptosis and autophagy by oridonin. RPMI8226 cells were pre-incubated with 5 mmol/L NAC for 1 h prior to exposure to 7 µmol/L oridonin for 24 h, and then, the apoptosis, autophagy, and the SIRT1 activity analyses were repeated. As shown in Figures 5A and 6A, NAC completely inhibited the oridonin-induced increase in DCF fluorescence intensity (22932. 67±2715.65 versus 44410.67±2478.12, P<0.01). An analysis of apoptosis and autophagy, including the FCM analysis of annexin V-FITC/PI dual staining, Western blot analysis of active caspase 3 and LC3-II, localization of LC3-II, and Western blot analysis of Beclin 1, were repeated in cells pre-incubated with NAC. While exposure to NAC alone for 24 h did not affect apoptosis and autophagy in cells (P>0.05), NAC completely abrogated the oridonin-induced increase in apoptosis ( Figures 2B, 2C, 6B, and 6C, P<0.01) and autophagy (Figures 4, 6D, 6E, and 6F, P<0.01). The SIRT1 nuclear protein level of cells pre-incubated with NAC was measured and compared with cells treated with oridonin alone. As shown in Figure 5B and Figure 6G, NAC reversed the oridonin-induced decrease in SIRT1 nuclear protein levels (P<0.01). These data suggest that SIRT1 activity was negatively regulated by intracellular ROS generation. Therefore, we conclude that intracellular ROS generation mediates the induction of apoptosis and autophagy by oridonin, possibly through the negative regulation of SIRT1 activity. Inhibition of autophagy sensitizes RPMI8226 cells exposed to oridonin to apoptosis mediated by intracellular ROS generation and SIRT1 downregulation We used 3-MA, a well-known inhibitor of autophagy, to explore the role of autophagy in cell death and the relationship between autophagy and apoptosis. Autophagy and apoptosis of cells pre-treated with 5 mmol/L 3-MA for 1 h and then exposed to 7 µmol/L oridonin for 24 h were analyzed using above mentioned methods. 3-MA suppresses oridonin-induced autophagy, but not baseline autophagy. As shown in Figures 4A, 7A, 7B, and 7C, the amount of LC3-II, the percentage of red fluorescent punctae-positive cells and Beclin 1 protein expression in cells pre-treated with 3-MA decreased markedly compared with cells treated with oridonin alone (P<0.01). In contrast, a comparison of these autophagic parameters between cells exposed to 3-MA alone and cells under normal conditions showed no significant differences (P>0.05). In addition, 3-MA enhances oridonin-induced apoptosis. As shown in Figures 2B and 7D, 3-MA augments the oridonin-induced increase in the apoptotic rate (24.30%±1.80% versus 17.53%±0.68%, P<0.01). Measurement of active caspase 3 levels in cells pre-treated with 3-MA revealed the same results as the analysis of the apoptotic rate ( Figure 2C and Figure 7E). In conclusion, apoptosis, but not autophagy, is the major effector pathway of oridonin-induced cytotoxicity and autophagy protects cells against apoptosis. Because the results above demonstrated that intracellular ROS generation mediates oridonin-induced apoptosis through negatively regulating SIRT1 activity, we determined whether intracellular ROS generation and SIRT1 were also involved in the pro-survival effect of autophagy. The intracellular ROS generation and SIRT1 protein level in the nuclei of cells pretreated with 5 mmol/L 3-MA for 1 h and then exposed to 7 µmol/L oridonin for 24 h was detected by an FCM analysis of DCF fluorescence density and Western blot analysis, respectively. As shown in Figures 5B, 7F, and 7G, pre-treatment with 3-MA further increases DCF fluorescence intensity (57385.55±3935.33 versus 44410.67±2478.12, P<0.01) and decreases SIRT1 nuclear protein level (P<0.01) compared with cells treated with oridonin alone for 24 h. In conclusion, we propose that intracellular ROS generation contributes to the autophagic pro-survival mechanism. Inhibition of autophagy sensitizes cells to apoptosis through promotion of ROS generation and SIRT1 downregulation induced by oridonin. Discussion Oridonin, a potential drug for tumor treatment, has been reported to simultaneously induce apoptosis and autophagy in some tumor cell lines, including HeLa and A431 cells, among others [9,26] . Studies investigating the relationship between autophagy and apoptosis are complicated and depend on the cellular context and stimulus. Jia and colleagues demonstrated that the induction of autophagy was essential to TNF-α-induced apoptosis in a T-lymphoblastic leukemia cell line, which indicated that execution of apoptosis is preceded by and depends upon autophagy in this context [27] . In other cellular settings, autophagy antagonized or delayed apoptosis, and the inhibition of autophagy increased the sensitivity of the cells to apoptotic signals [26,28] . In our study, we found that oridonin could simultaneously induce caspase 3-mediated apoptosis and Beclin 1-dependent autophagy in RPMI8226 cells. The inhibition of autophagy by 3-MA sensitized the cells to apoptosis, which suggests that oridonin cytotoxicity is mainly the result of apoptosis and that autophagy acts as a pro-survival mechanism in cells exposed to oridonin. Intracellular ROS generation plays a significant role in physiological and pathological processes, and a high level of ROS is closely associated with apoptotic cell death. Recently, many studies have reported that ROS could induce apoptosis in a variety of malignant cells [12,13] . Additionally, other studies have shown that oxidative stress could induce autophagy under certain conditions. H 2 O 2 and 2-methoxyestradiol, a well-known generator of ROS, could induce autophagy, which contributed to cell death in HEK293, U87, and HeLa cells [29] . Our study demonstrated that oridonin induces a time-dependent increase in intracellular ROS generation accompanied by increases in apoptosis and autophagy. The complete inhibition of intracellular ROS generation by NAC abrogated oridonin-induced apoptosis and autophagy. These data indicate that intracellular ROS generation is required for the induction of autophagy and apoptosis by oridonin. Hasegawa et al originally demonstrated that excess ROS positively regulates SIRT1 activity and serves as a compensatory mechanism to protect cells against apoptosis [18] . Recently, some studies have shown that SIRT1 activity was negatively regulated by intracellular ROS generation in lung epithelial cells, endothelial cells, and macrophages in response to cigarette smoke extract in vitro and in the lungs of patients with COPD (chronic obstructive pulmonary disease) [19,30,31] . In this study, we demonstrate that oridonin induces a time-dependent The LC3-II and LC3-I protein levels of treated cells were detected using Western blot assay. γ-Tubulin was used as an equal loading control. The bands were quantified by densitometric analysis. The LC3-II were corrected relative to γ-tubulin. (B) After designed experiment measures, the LC3 localization in cells was determined by immunofluorcesence using a fluorescent probe of QDs 605 nm -Anti-LC3, and the localization of LC3-II at autophagosome membrane were indicated as red fluorescent punctate dots. The representative images are shown. Autophagy was quantitated by the percentage of fluorescent punctate-positive cells in total cells. the percentage of fluorescent punctate-positive cells was calculated as follows: the percentage of fluorescent punctate-positive cells (%)=number of fluorescent punctate-positive cells/number of total cells×100% (at least 500 cells were counted under a fluorescence microscope). (C) The protein level of Beclin 1 of treated cells was detected using Western blot assay. γ-Tubulin was used as an equal loading control. The bands were quantified by densitometric analysis. The values for Beclin 1 were corrected relative to theγ-tubulin and shown in histogram. All statistical significance was determined by Student's t-test. Columns, mean of three independent experiments; Mean±SD; c P<0.01. www.nature.com/aps Zeng R et al Acta Pharmacologica Sinica npg ROS generation in RPMI8226 cells exposed to oridonin. The anti-apoptotic function of SIRT1 was self-explanatory [20,21,32] . Accordingly, we demonstrated that oridonin decreases SIRT1 nuclear protein levels and increased apoptosis whereas 3-MA enhances oridonin-induced apoptosis and the decrease in SIRT1 nuclear protein levels. However, the role of SIRT1 in autophagy is controversial. Some studies have shown that SIRT1 promotes autophagy by down-regulating mTOR signaling [23] ; however, other studies have shown that the inhibition of SIRT1 activity augmented autophagy [22] . In this study, we found that the increase in autophagy was associated with a decrease in SIRT1 nuclear protein levels and that the inhibition of autophagy by 3-MA further reduces SIRT1 nuclear protein levels and increases intracellular ROS generation. This result suggests that autophagy is negatively regulated by SIRT1 activity. In summary, generation of intracellular ROS mediates apoptosis and autophagy in RPMI8226 cells exposed to oridonin and may be associated with a negative regulation of SIRT1 activity. Apoptosis, but not autophagy, was the major effector of oridonin-induced cytotoxicity. Autophagy protects against apoptosis mediated by intracellular ROS generation and SIRT1 activity. Our results provide new mechanistic information related to oridonin-induced apoptosis and autophagy, in addition to new MM therapeutic targets to enhance oridonin cytotoxicity.	2017-11-08T00:55:06.315Z	2011-12-12T00:00:00.000	{ "year": 2012, "sha1": "c1cbd44fdf91687ff7f85fdc1b61747eb112f172", "oa_license": null, "oa_url": "https://www.nature.com/articles/aps2011143.pdf", "oa_status": "BRONZE", "pdf_src": "PubMedCentral", "pdf_hash": "c1cbd44fdf91687ff7f85fdc1b61747eb112f172", "s2fieldsofstudy": [ "Biology", "Medicine", "Chemistry" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
118871493	pes2o/s2orc	v3-fos-license	Perfect fluid solutions of Brans-Dicke and f(R)cosmology Brans-Dicke cosmology with an (inverse) power-law potential is revisited in the light of modern quintessence and inflation models. A simple ansatz relating scale factor and scalar field recovers most of the known solutions and generates new ones. A phase space interpretation of the ansatz is provided and these universes are mapped into solutions of f(R) cosmology. Introduction There are currently many theoretical and experimental investigations of possible deviations of gravity from Einstein's theory in cosmology, black holes, gravitational waves, and the dynamics of galaxies and galaxy clusters [1]. The standard ΛCDM model of cosmology requires the introduction of a completely ad hoc dark energy accounting for 70% of the energy content of the universe [2]. An alternative to dark energy consists of modifing gravity at large scales, which has led to contemplating many theories of gravity, and especially the so-called f (R) class [3,4,5,6]. These are essentially scalar-tensor theories, and scalar-tensor gravity is the prototypical alternative to General Relativity (GR) which introduces only an extra scalar degree of freedom. The simplest scalartensor gravity was proposed by Brans and Dicke in 1961 [7] and was later generalized [8,9,10] and is still the subject of active research. In any case, although Solar System tests do not show deviations from GR, gravity is tested poorly in many regimes while it is not tested at all in others [11,12], and there is plenty of room for deviations from GR. Apart from the motivation arising from cosmology, attempts to unify gravity and quantum mechanics invariably produce deviations from GR in the form of extra degrees of freedom, higher order field equations, and extra tensors in the Einstein-Hilbert action, so it is expected that eventually GR fails at some energy scale. Indeed, the simplest bosonic string theory reduces to a Brans-Dicke theory with coupling parameter ω = −1 in the low-energy limit [13,14]. Motivated by the flourishing of cosmological models in alternative gravity and especially in f (R) and other scalar-tensor gravities, we revisit the simplest incarnation, Brans-Dicke cosmology with a scalar field potential. There are now over five decades of research on this subject but not many analytical solutions are known which describe spatially homogeneous and isotropic cosmology (see [15,16] for partial reviews). Contrary to the original Brans-Dicke theory, in which the extra gravitational scalar field was free and massless, we study the situation in which it acquires a power-law or inverse power-law potential, which has now been included in a large number of cosmological scenarios related to inflation or to the present acceleration of the universe [17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,15,16,6]. As shown in the following sections, most of the known solutions of spatially homogeneous and isotropic Brans-Dicke cosmology can be derived from a simple ansatz, which allows one to uncover new solutions of this theory with exponential scale factor and scalar field. A geometric interpretation of this ansatz in terms of the geometry of the phase space of the solutions is proposed in Sec. 3. The old and new solutions of Brans-Dicke cosmology with potential are then mapped into solutions of f (R) cosmology in Sec. 4. We begin with the Brans-Dicke action 1 [7] where R is the Ricci scalar, φ is the Brans-Dicke scalar field, ω is the constant Brans-Dicke coupling, V (φ) is a potential for the Brans-Dicke field, and S (m) is the matter action. Here we assume a power-law or inverse power-law potential with V 0 and β constants and V 0 ≥ 0. This form of the potential is motivated by large bodies of literature on inflation [17,18,19,20,21,22,23,24,25,26] and quintessence [27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53]. The Brans-Dicke field equations in the Jordan frame are where ∇ a is the covariant derivative operator, ≡ g ab ∇ a ∇ b , T ab is the energy-momentum tensor of ordinary matter, and T ≡ T a a is its trace. In the following we assume that ω = −3/2 and that matter consists of a perfect fluid with stress-energy tensor where u a is the fluid 4-velocity) and with barotropic, linear, and constant equation of relating the energy density ρ and pressure P . A cosmological constant can be introduced in the theory by considering a linear potential V = Λφ. In fact, since the vacuum action of GR contains the combination R − Λ, and the Brans-Dicke field φ multiplies R in the Brans-Dicke action, the natural way of introducing a cosmological constant in Brans-Dicke theory is through the combination φ (R − Λ), which is equivalent to introducing a linear potential V = Λφ. Or, considering the Brans-Dicke field equation (1.3), it is obvious that adding a term Λg ab to the left hand side is equivalent to inserting a linear potential V = Λφ in the right hand side. The parameters of the theory are (ω, β, V 0 , γ). We now specialize to spatially homogeneous and isotropic Brans-Dicke cosmology, with the geometry given by the Friedmann-Lemaître-Robertson-Walker (FLRW) line element in comoving coordinates, where k is the curvature index and dΩ 2 (2) = dθ 2 + sin 2 θ dϕ 2 is the line element on the unit 2-sphere. The equations of Brans-Dicke cosmology with the perfect fluid (1.5) and (1.6) consist of the Friedmann, acceleration, and scalar field equations respectively, where H ≡ȧ/a is the Hubble parameter and an overdot denotes differentiation with respect to the comoving time t. In addition, the covariant conservation equation ∇ b T ab = 0 yieldsρ + 3H (P + ρ) = 0 (1.11) which, using the equation of state (1.6), is immediately integrated to where ρ 0 is a non-negative integration constant. New and old solutions Since the early days of scalar-tensor gravity, various authors have looked for FLRW solutions of this class of theories in the power-law form, a(t) ∝ t q and φ(t) ∝ t s . Here we search for solutions satisfying the ansatz φ(t) = φ 0 a p . Scaling and power-law relations are ubiquitous in physics [60], biology [61,62], geophysics and glaciology [63,64,65,66], and various natural sciences, and it is rather natural to investigate such relations in cosmology. The fairly large literature studying cosmological power-law solutions a(t) ∼ t q , φ(t) ∼ t r is mostly well motivated from the physical point of view. The ansatz φ = φ 0 a p reproduces almost all these power-law solutions. The physical meaning of this ansatz resides in the fact that the effective gravitational coupling strength becomes G ef f ∼ φ −1 ∼ a −p and the ansatz relates directly the strength of gravity with the cosmic scale factor. If p > 0, the effective gravitational coupling decreases as the universe expands, while G ef f increases if p < 0. These two behaviours are separated by GR, which corresponds to p = 0. The assumption φ(t) = φ 0 a p can be rewritten in a covariant way as u c ∇ c φ/φ = −p Θ/3, where Θ is the expansion of the congruence of observers comoving with the cosmic fluid, which have timelike 4-tangent u c . According to the comoving time associated with these observers,Ġ ef f /G ef f = −φ/φ. The ansatz φ = φ 0 a p offers a self-consistent scenario realizing the assumptionĠ/G ∼ H used in analyses of the variation of the gravitational coupling. This assumption is often made on a purely phenomenological basis and corresponds to the (rather vague) idea that G varies on a cosmological time scale, in order to place experimental or observational constraints on the variation of G (cf., e.g., Refs. [67,68]). If the evolution of G ef f and that of the scale factor are not tied together directly, as in our ansatz in the context of scalar-tensor gravity, it is difficult to see how the desired phenomenological relationĠ/G ∼ H can be obtained, and how it can be obtained in a covariant way. In our study we first want to recover the known power-law solutions, hence we begin by assuming that where q(ω, β, γ) and p(ω, β, γ) are exponents to be determined as functions of the parameters of the theory and a 0 , φ 0 , and ρ 0 in Eq. (1.12) are constants. We require that p = 0 because otherwise one has a constant scalar field which reduces Brans-Dicke theory to GR, which is a trivial situation in our context. The form (2.13) and (2.14) of the solutions of Brans-Dicke cosmology to which we restrict is sufficiently general to allow us to recover a host of classic solutions. Later on, we will relax the assumption (2.13) but we will keep the ansatz (2.14) finding new exponential, instead of power-law, solutions for spatially flat FLRW universes. An interpretation of the ansatz (2.14) in the phase space of the solutions is given in Sec. 3. There are two possible approaches to the solution of Eqs. (1.8)-(1.10). In the first approach, the first step of the solution process consists of finding exponents q(ω, β, γ) and p(ω, β, γ) that solve all three field equations (1.8)-(1.10). This is obtained by matching the powers of the comoving time t in these equations. The Friedmann equation (1.8) gives Matching the powers of t in each term yields the following relations: The second possible approach to solving Eqs. (1.8)-(1.10) consists of noting that some of the four terms in the Friedmann equation (1.8) could balance each other, without having to match all the powers of t. However, the acceleration and field equations impose further constraints and, in practice, this method does not lead to new solutions with respect to those obtained with the first method (the details of this second approach are presented in A). Let us continue, therefore, with the first solution method. The second step of this process consists of taking the functions q(ω, β, γ) and p(ω, β, γ) found in the previous step (if they exist) and of determining the various integration constants a 0 , φ 0 , ρ 0 as functions of (ω, β, γ, V 0 , p, q). Using again the Friedmann equation (1.8), computer algebra provides the value of the integration constant for the density 2 ρ 0 as a function of the other two integration constants φ 0 and a 0 as Then the scalar field equation (1.10) provides the integration constant φ 0 appearing in the Brans-Dicke field as for V 0 = 0. The acceleration equation (1.9) provides another such relation for the integration constant a 0 : . If instead V 0 = 0, there is no such constraint on φ 0 and the expression In spite of the simplicity introduced by the assumptions (2.13) and (2.14), the field equations (1.8)-(1.10) are still non-linear and quite involved and it is convenient to analyze the various possibilities separately. In this vacuum none of the constraints (2.16)-(2.18) between p, q, β, and γ apply. The Friedmann equation (1.8) becomes simply and, in a non-static universe, it provides the values of p The acceleration equation (1.9) then gives . (2.24) Using the values (2.23), (2.24) and Eq. (2.14), one concludes that . This is the classic O'Hanlon and Tupper vacuum solution of Brans-Dicke cosmology with free scalar field and ω > −3/2, ω = −4/3, 0, describing a spatially flat FLRW universe [69]. In this case p and q depend only on the Brans-Dicke coupling ω, while the constants a 0 and φ 0 are not constrained. In this non-vacuum case, only the constraint (2.17) between p and q must hold, and this equation is all the information that can be obtained by matching powers of t in the field equations. Equation (2.14) then yields 28) and substituting this in the acceleration equation (1.9), one obtains an algebraic equation for p with roots The other field equation (1.10) must also be satisfied, and it is satisfied by the root p + but not by p − . Therefore, using Eq. (2.17), one concludes that the only solution of the desired form corresponding to a spatially flat FLRW universe with free Brans-Dicke scalar and with perfect fluid is This is recognized as the Nariai solution [70,71]. The power q of the scale factor a(t) ≃ t q is independent of the Brans-Dicke coupling ω if γ = 2 or if γ = 4/3. The constants a 0 and φ 0 are not constrained and ρ 0 is given in terms of them and of ω, γ, p by Eq. (2.28). Exponential solutions For k = 0, V 0 = 0, ρ 0 = 0, there are expanding/contracting de Sitter spaces with exponential scalar fields. Assuming H = const., the Friedmann equation (1.8) becomes and it can be satisfied for ρ 0 = 0 and constant H only if which yields . (2.36) In order to satisfy the Friedmann and acceleration equations, it must also be The second value of γ, however, does not satisfy the scalar field equation and is discarded. The remaining value of γ gives the solution . Ordinary matter has γ ≥ 1, which corresponds to p = −3γ < 0 and G ef f ∼ a −p increases on a cosmological time scale as the universe expands. Of the three constraints, only (2.18) must be satisfied in this vacuum. One finds and, therefore, while a 0 is not constrained, φ * = φ 0 a p 0 and (2.44) Exponential solutions For k = 0, ρ 0 = 0, and V 0 = 0, there are exponential solutions which are expanding or contracting de Sitter spaces with H = const. andφ/φ = pH = const. Assuming H = const., the Friedmann equation (1.8) reduces to where φ * = φ 0 a 1 ω+1 0 . Contrary to GR, the scalar field of these de Sitter spaces is not constant. These solutions are well known attractors in the phase space of Brans-Dicke cosmology, with an attraction basin which is wide but does not span all of the phase space [72,73,74]. 2.5 k = 0 , V 0 = 0 , ρ 0 = 0 In this non-vacuum case, it is necessarily q = 1 and p = 2 − 3γ. The FLRW solution is Here φ 0 is arbitrary (but positive) and the other integration constants are The parameters k, ω, γ, of course, must lie in a range such that a 0 > 0 and ρ 0 > 0. If k = −1 this universe is just Minkowski space in a foliation with time-dependent 3metric and the line element (1.7) can be reduced to the Minkowski one by an appropriate coordinate transformation (see, e.g., [75]). It is not a trivial Minkowski space because the effective stress-energy tensor of the free Brans-Dicke scalar cancels out the fluid stress-energy tensor in the field equations (1.3) to produce flat spacetime. If k = 1, spacetime is a genuine positively curved FLRW manifold. A phase space interpretation We now provide a geometric interpretation of the ansatz φ(t) = φ 0 a p (t) in the phase space of the solutions. For simplicity, we restrict to the simplest situation, which corresponds to the parameter values k = 0 and ρ 0 = 0, in which case the dimensionality of the phase space reduces to three, thus allowing for an intuitive graphical interpretation (a generic description of the phase space of Brans-Dicke cosmology was given in Ref. [76]). When k = 0 and in the absence of matter, the scale factor a(t) enters the cosmological equations (1.8)-(1.10) only through the combination H =ȧ/a and one can choose as variables the Hubble parameter and the Brans-Dicke scalar (H(t), φ(t)). Then the phase space reduces to H, φ,φ . The Friedmann equation (1.8), which is of first order, then acts as a constraint which forces the orbits of the solutions to lie on the analogue of the "energy surface" with equation (1.8), effectively reducing the phase space accessible to these orbits to a 2-dimensional subset 3 of the 3-dimensional space H, φ,φ . Let us examine this "energy surface" in the simple case k = 0, ρ 0 = 0. The constraint equation (1.8) becomes We can regard it as an algebraic equation forφ and expressφ as a function of the other two variables H and φ, if ω = 0 . (3.80) In general, for ω = 0, one or more regions of the (H, φ) plane correspond to a negative argument of the square root in Eq. (3.80). In this case, the corresponding regions in the "energy surface" H, φ,φ(H, φ) are "holes" which are avoided by the orbits of the solutions (in the sense that no real solutions of the dynamical system (1.8)-(1.10) exist whose orbits enter these regions). These holes can be infinite or semi-infinite. Further, the "energy surface" is composed of two sheets, corresponding to the positive or negative signs in Eq. (3.80), which will be denoted "upper sheet" and "lower sheet" (in keeping with the terminology of Ref. [76]). These two sheets join on the boundaries of the "holes", which are identified by the equation ∆ = 0. Looking for solutions satisfying the ansatz φ = φ 0 a p means intersecting the "energy surface" (3.80) with the surface of equatioṅ expressing the assumption that the effective gravitational coupling varies asĠ ef f /G ef f = −pH. However, the constant p is not assigned a priori. The problem consists of finding simultaneously values of p for which these intersections exist and the intersection curves themselves, which are the orbits of the solutions satisfying the desired ansatz. As an example, consider the parameter values ω = 55 and β = 4 and use units in which V 0 is unity. Then the region forbidden to the orbits of the solutions is given by The two sheets composing the "energy surface" have equationṡ Upper sheet, lower sheet, and the "energy surface" are plotted in Figs. 1-3. The boundary of the hole, where the two sheets join, is the curve whose points have coordinates The intersections between the energy surface (3.84) and the surfaceφ (H, φ) = pHφ change as p changes. However, only the value of p given by Eq. (2.41), that is p = −4/115 for this example, corresponds to actual orbits of the solutions of the dynamical system (1.8)-(1.10). This "ansatz surface" and its intersection with the "energy surface" are plotted in Fig. 4 and Fig. 5, respectively. 4 Solutions of metric f (R) gravity f (R) theories of gravity [77,78] are a subclass of scalar-tensor gravity with action where f (R) is a non-linear function of the Ricci scalar R. This action is equivalent to a Brans-Dicke one. By defining the scalar field φ = f ′ (R), it can be shown that the action (4.86) is equivalent to [3,4,5,6] where and where R = R(φ) is now a function of φ = f ′ (R) usually defined implicitly [3,4,5]. This theory has Brans-Dicke coupling ω = 0 and the potential (4.88) for the Brans-Dicke scalar. The Jordan frame solutions of Brans-Dicke cosmology reported in the previous sections can be seen also as solutions of some f (R) cosmology. This is true if ω = 0 and The functional form f (R) = µR n , where µ and n are constants, satisfies these requirements provided that 4 β = n n − 1 , (4.90) for n = 1 (if n = 1 this f (R) theory reduces to GR). It must be n > 1 to guarantee that V 0 > 0. In practice, the value of n is severely constrained by Solar System experiments, which require that n = 1 + δ with δ = (−1.1 ± 1.2) · 10 −5 [80,81,82,83,84,85,86,87,88,89]. At the same time, any f (R) theory must satisfy f ′ > 0 in order for the graviton to carry positive energy and f ′′ ≥ 0 to guarantee local stability [3,4,5,90]. These constraints are satisfied if n = 1 + δ with δ ≥ 0. In spite of the experimental bounds on the exponent n, R n gravity has been the focus of much work aiming at exploring the possible phenomenology of f (R) gravity and many phase space analyses for R n cosmology are available in the literature [91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109] (see [113] for a phase space picture of general f (R) cosmology analogous to that of the previous section). Moreover, in strong curvature regimes in the early universe, in which the present-day Solar System constraints do not apply, Starobinsky-like inflation [114] corresponding to f (R) = R + µR 2 is well approximated by f (R) ≃ µR 2 . When the conditions (4.90) and (4.91) are satisfied, the FLRW solutions of Brans-Dicke gravity with power-law potential reported in the previous sections are also solutions of R n gravity with or without a perfect fluid, which are added to the relatively scarce catalogue of exact solutions of these theories. Conclusions The simple ansatz φ = φ 0 a p recovers most of the known solutions of Brans-Dicke cosmology and generates new ones in the presence of a power-law or inverse power-law potential, which is well-motivated in cosmology and particle physics. These solutions include power-law and exponential dependence of the scale factor a(t) and of the Brans-Dicke field φ(t) from the comoving time t. This ansatz has a fairly simple geometric interpretation in the phase space of the solutions as the simultaneous search for a curve generated by the intersection of two surfaces and for the number p. The geometry of the phase space, however, can be complicated for various choices of the (inverse) powerlaw potential V (φ) = V 0 φ β and if the spatial sections of the Brans-Dicke cosmology are curved. Details such as the integration constants appearing in the classic solutions as functions of the parameters of the theory have been provided by new general formulas, which are missing in the literature probably because of the non-availability of computer algebra at the time when these solutions were discovered. We have now a more unified and comprehensive view of analytical solutions of Brans-Dicke cosmology. Prompted by the huge literature on f (R) cosmology as an alternative to dark energy, it is natural to try to relate the old and new solutions of Brans-Dicke cosmology to corresponding solutions of f (R) cosmology. It turns out that this is possible, and indeed relatively straightforward, for the family described by the choice f (R) = µR n . The search for new f (R) cosmologies will be continued in the future. (1.94) Equating to zero the terms in parenthesis separately yields If we substitute these ρ 0 and k values into the acceleration equation (1.9), we obtain (1.97) This equation can be satisfied in two ways: the first one consists of setting the first parenthesis to zero, while the second way consists of setting the second parenthesis to zero. However, the second possibility is already discussed in Sec. 2.7. Setting the first parenthesis to zero gives Now we need to satisfy the scalar field equation (1.10), which gives requiring one to set the powers of t equal to each other. Therefore, we conclude that balancing these two pairs does not produce any new solution. • Balancing the terms (2) and (3) gives 2q = q(p + 3γ), whereas balancing (1) and (4) In order to satisfy the scalar field equation (1.10), we substitute the values of V 0 , k, and γ to obtain (1.106) Finding a solution without setting the powers of t equal to each other requires, at a minimum, to set β = 1, which makes the power of t infinite. Therefore, this choice of balancing terms give no reasonable solution without setting the powers of t equal to each other. These three cases show that making different matches of the terms in Friedmann equation (1.92) does not yield new solutions.	2018-02-27T18:01:05.000Z	2017-11-10T00:00:00.000	{ "year": 2018, "sha1": "cebfc8b492a60b8b1887b1040f718ee512076435", "oa_license": null, "oa_url": "http://arxiv.org/pdf/1711.04026", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "cebfc8b492a60b8b1887b1040f718ee512076435", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
210072454	pes2o/s2orc	v3-fos-license	Anti-Photoaging Effect of Korean Mint (Agastache rugosa Kuntze) Extract on UVB-Irradiated Human Dermal Fibroblasts Ultraviolet B (UVB) irradiation-induced photoaging leads to wrinkles, dryness, and skin roughness. UVB irradiation activates the production of reactive oxygen species (ROS) and stimulates the mitogen-activated protein kinase (MAPK)/activator protein-1 (AP-1) signaling pathway, which promotes expression of matrix metalloproteinases (MMPs) and inflammatory cytokines. The current study aimed to assess the anti-photoaging activity of Agastache rugosa extract (ARE) on UVB-treated human dermal fibroblasts. ARE treatment reduced the overproduction of ROS and promoted mRNA expression of anti-oxidant enzymes. ARE treatment significantly inactivated the MAPK/AP-1 signaling pathway, which downregulated the expression of MMPs. Moreover, ARE promoted the production of type-I procollagen and upregulated mRNA expression of collagen genes. Additionally, ARE suppressed the expression of inflammatory cytokines, including interleukin (IL)-1β, IL-6, and IL-8, by preventing expression of nuclear factor-kappa B. Collectively, our findings show that ARE could be a potential candidate for anti-photoaging treatment. INTRODUCTION Skin aging involves intrinsic and extrinsic aging (Ramose-Silva et al., 2013). Intrinsic aging occurs with the passage of time, whereas extrinsic aging occurs owing to exposure to external factors, such as harmful chemicals, air pollutants, and ultraviolet B (UVB) irradiation (Kohl et al., 2011). Extrinsic aging resulting from long-term exposure to UVB irradiation is called photoaging. During this process, UVB irradiation penetrates into the dermis and induces breakdown of skin components; this is associated with skin inflammation, DNA damage, and oxidative stress (Matsumura and Ananthaswamy, 2004). In particular, excessive generation of reactive oxygen species (ROS), a major element involved in stimulating oxidative stress, damages skin cells, which degrades and disorganizes extracellular matrix (ECM) components (Kim et al., 2015). UVB irradiation-induced ROS production activates mitogen-activated protein kinases (MAPKs), subsequently stimulating to formation of the activator protein-1 (AP-1) (Rabe et al., 2006;Heng, 2013;Lu et al., 2016). AP-1 increases the expression of matrix metalloproteinases (MMPs), which stimulate breakdown of ECM compo-nents, especially collagens (Watson et al., 2014). Degrading collagens, main components involved in supporting dermal layers, lead to the destruction and collapse of skin structure (Quan et al., 2004;Chen et al., 2015). Therefore, downregulating MMP expression is an effective strategy to prevent and delay the development of photoaging-related symptoms, such as wrinkle formation and thickening in UVB-exposed skin (Bae et al., 2010). In terms of new collagen formation in the ECM, AP-1 complex blocks type-I procollagen production and downregulates collagen gene expression in UVB-exposed skin cells (Kammeyer and Luiten, 2015). Thus, photoaging is closely associated with both stimulating collagen breakdown and blocking collagen formation by suppressing type-I collagen synthesis and collagen gene expression (Sun et al., 2016). Additionally, exposure to UVB irradiation enhances translocation of nuclear factor kappa-B (NF-B) to the nucleus where it is mainly involved transcription of inflammatory cytokines (Pillai et al., 2005). The inflammatory responses induced by activation of NF-B in UVB-exposed skin cells lead to MMP overexpression and collagen degradation (Bai et al., 2015). Agastache rugosa Kuntze, also known as Korean mint, is mainly found in Korea, Japan, and China. A. rugosa has been used as a food source and in traditional medicines to cure disease because it has varied bioactivities, which encompass anti-oxidant, anti-melanogenic, anti-atherogenic, anti-inflammatory, and anti-fungal properties (Hong et al., 2001;Lee et al., 2017). Previously, A. rugosa leaf extract was shown to reduce photoaging by activating glutathione and superoxide dismutase (SOD) in human keratinocytes (HaCaT) (Oh et al., 2016). However, the antiphotoaging effect of A. rugosa extract (ARE) and the underlying mechanisms in human dermal fibroblasts (HS68) have not yet been proven. Here, we determined the attenuating effect of ARE on photoaging in UVB-treated human dermal fibroblasts by examining the underlying molecular mechanisms. Preparation of ARE ARE was obtained from Cosmax NBT (Seoul, Korea). The dried aerial parts of A. rugosa were extracted with water at 95 o C for 4 h. ARE filtrates were evaporated, and the final yield of ARE was 10% (w/w). Cell culture and UVB irradiation HS68 cells were purchased from the American Type Culture Collection (Manassas, VA, USA). The cells were maintained in Dulbecco's modified Eagle's medium (HyClone Laboratories, Inc., Logan, UT, USA) containing 120 U/mL penicillin and 75 g/mL streptomycin (Invitrogen, Grand Island, NY, USA), and 10% fetal bovine serum (HyClone Laboratories, Inc.) in a humidified atmosphere of 5% CO 2 at 37 o C. To induce photoaging, the cells were exposed to UVB irradiation (15 mJ/cm 2 ) by using a UV crosslinker CL-1000M (Ultra-Violet Products Ltd., Cambridge, UK). Cell viability After UVB irradiation, HS68 cells were treated with several concentrations of ARE dissolved in serum-free medium for 24 h. After 24 h, the cells were incubated with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich Co., St. Louis, MO, USA) solution (0.5 mg/mL) for 4 h in 5% CO 2 atmosphere at 37 o C. After the removal of MTT solution, the absorbance of insoluble formazan dissolved in dimethyl sulfoxide was recorded using a VERSAmax tunable microplate reader (Molecular Devices Inc., Sunnyvale, CA, USA) at 540 nm. No significant cytotoxicity of ARE (up to 20 g/mL) was observed (data not shown). Therefore, ARE (≤20 g/ mL) was used for further experiments. Measurement of ROS production After treatment with ARE and UVB exposure, cells were incubated with 2,7-dichlorofluorescein diacetate (Sigma-Aldrich Co.) in a CO 2 incubator at 37 o C for 30 min. Subsequently, the cellular fluorescence was recorded using a SpectraMax Gemini EM microplate spectrofluorometer (Molecular Devices Inc.). Western blot analysis Cells were lysed with NP40 lysis buffer (Elpis-Biotech, Daejeon, Korea) supplemented with 0.2% protease inhibitor cocktail (Sigma-Aldrich Co.). After centrifugation at 16,000 g at 4 o C for 15 min, the concentration of protein in the supernatant was evaluated by Bradford solution (Bio-Rad Laboratories, Inc., Hercules, CA, USA) with bovine serum albumin (bioWORLD, Dublin, OH, USA) used as a standard. Equal amounts of proteins in each group were subjected to 8∼10% sodium dodecyl sulfatepolyacrylamide gel electrophoresis for 90 min at 110 V. The separated proteins were transferred onto nitrocellulose blotting membranes (GE Healthcare, Piscataway, NJ, USA) for 1 h at 110 V. After nonspecific sites were blocked with 5% skimmed milk (Difco Laboratories Inc., Detroit, MI, USA), membranes were immunoblotted with primary antibodies against NF-B, phospho (p)-p38, pextracellular signal-regulated kinase (p-ERK), p-c-Jun Nterminal kinases (p-JNK), and c-Fos (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA.); and p-c-Jun, c-Jun, ERK, JNK, p38, and -tubulin (Cell Signaling Technology, Danvers, MA, USA) at room temperature for 4 h. After washing three times in Tris-buffered saline (Dynebio, Gyeonggi, Korea) containing Tween 20, membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit or anti-mouse secondary antibodies (Bethyl Laboratories, Montgomery, TX, USA) at room temperature for 1 h. Enhanced chemiluminescence solution (Amersham BioSciences UK Ltd., Little Chalfont, UK) was used to develop the membranes. The band of protein was visualized using a G:BOX EF imaging system (Syngene, Cambridge, UK) and the GeneSys program. Measurement of type-I procollagen contents After treatment with ARE and UVB exposure, the type-I procollagen content of the medium was evaluated using a human procollagen I alpha 1 enzyme-linked immunosorbent assay kit (Abcam, Cambridge, MA, USA), according to the company's protocol. Absorbance was recorded using a VERSAmax tunable microplate reader (Molecular Devices Inc.) at 450 nm. Reverse transcription-polymerase chain reaction (RT-PCR) RNAiso Plus (Takara Bio Inc., Kusatsu, Japan) was used for isolation of total RNA from cells. The quantity and quality of the isolated mRNA were spectrophotometrically determined using a NanoDrop Lite spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). The mRNA was reverse-transcribed to cDNA with RT PreMix (Elpis Biotech, Daejeon, Korea). Next, the cDNA of target genes was amplified using specific primers and HiPi PCR PreMix (Elpis Biotech). The list of primer sequences is shown in Table 1. Amplification was carried out over 25∼30 cycles; denaturation for 30 s at 94 o C, annealing for 1 min at 56∼60 o C, and extension for 1 min at 72 o C. PCR experiments were carried out using a SimpliAmp Thermal Cycler (Applied Biosystems, Hercules, CA, USA). The amplified cDNA was stained with Loading star (Dynebio) and separated for 30 min in a Mupid electrophoresis chamber (Advance, Tokyo, Japan) on 1.5% agarose gel. The bands of target genes were visualized with G:BOX EF imaging system (Syngene) and GeneSys program. Statistical analysis All experiments were carried out three or more times. Data are shown as the mean±standard deviation (SD) and were analyzed using one-way analysis of variance (ANOVA). Intergroup differences were evaluated by Scheffe's test using SPSS 24.0 (SPSS Inc., Chicago, IL, USA). Statistical significance is indicated by P<0.05 or P<0.01. Hash tags and astericks denote significant differences compared to control cells and UVB-treated cells, respectively. ARE suppresses ROS production and increases anti-oxidant enzyme expression. Excessive production of ROS is a major cause of photoaging in UVB-irradiated skin, and can further induce oxidative stress and MMP expression (Park et al., 2014). Thus, inhibition of UVB-induced ROS production could be an effective strategy to prevent photoaging. The ability of ARE to scavenge free radicals, and thereby exert anti-oxidant effects, was previously reported (Desta et al., 2016). Moreover, the phytochemicals, acacetin, and rosmarinic acid, which are present in ARE, have been reported to possess anti-oxidant activities (Chen et al., 1990;Kim et al., 1999). Based on these studies, we evaluated the anti-oxidant properties of ARE in UVB-exposed human dermal fibroblasts by examining total levels of intracellular ROS. UVB irradiation markedly increased ROS generation; however, ARE treatment dose-dependently alleviated ROS production by 92% (10 g/mL) and 142% (20 g/mL), compared with UVB-treated HS68 cells (Fig. 1A). These results indicate that ARE exhibits anti-oxidant activity by suppressing UVB-induced ROS production. We also examined mRNA expression of anti-oxidant enzymes in UVB-treated HS68 cells. It was reported that UVB-induced ROS led to cellular senescence in dermal fibroblasts (Yang and Li, 2015). As anti-oxidant enzymes protect skin from UVB-induced thickening and wrinkle formation, it may be necessary to enhance the activities of these enzymes to prevent photoaging (Draelos, 2007). According to a previous study, linarin, an active compound in ARE, markedly increased expression of the major anti-oxidant enzyme catalase in lipopolysaccharide-stimulated mouse lung tissues (Han et al., 2018). We evaluated mRNA expression of the anti-oxidant enzymes catalase, SOD, and glutathione peroxidase (GPx) in cells treated with ARE. UVB irradiation reduced expression of all three transcripts; however, mRNA expression levels of these enzymes were markedly elevated after ARE treatment (Fig. 1B). Taken together, the reduction of UVB-stimulated ROS production and elevated expression of anti-oxidant enzymes suggest that ARE has potent anti-oxidant properties. ARE suppresses UVB-induced MMP expression through the downregulation of the MAPK/AP-1 pathway. UVB irradiation stimulates MAPKs, which promotes heterodimerization of proteins in the AP-1 complex. Enhancement of the MAPK/AP-1 signaling pathway promotes upregulation of MMP expression, which subsequently results in breakdown of ECM components, such as elastic fibers, glycosaminoglycans, and collagens. Consequently, this signaling cascade results in photoagingrelated phenotypes, including wrinkle formation, epidermal thickening, and skin dryness (Chiang et al., 2012). It was reported that UVB exposure upregulated the pro-MMP-2 and -9 expression; however, probiotic-fermented ARE decreased the MMP expression in UVB-induced HaCaT cells (Shin et al., 2018). Based on this effect of ARE, we evaluated the expression of MMPs, MAPK, and AP-1 in UVB-irradiated HS68 cells. UVB treatment markedly increased MMP mRNA expression; however, treatment with ARE significantly decreased their expression ( Fig. 2A). In this study, the MAPK signaling pathway was activated in the UVB-exposed cells. However, ARE significantly decreased MAPK expression, compared with UVB-irradiated HS68 cells (Fig. 2B). In comparison with untreated cells, UVB irradiation significantly upregulated protein expression of the AP-1 components p-c-Jun and c-Fos, whereas ARE downregulated p-c-Jun and c-Fos (Fig. 2C). Collectively, these data indicate that ARE decreases UVB-induced MMPs expression by inactivating the MAPK/AP-1 signaling pathway. ARE increases type-I procollagen production and collagen expression. UVB irradiation stimulates collagen breakdown by increasing the expression of MMPs and downregulates expression of collagen genes by activating the AP-1 complex (Chen et al., 2015). These responses alter the ECM ar-chitecture, resulting in wrinkle formation, elastosis, and skin dryness (Kammeyer and Luiten, 2015). Since ARE was revealed to reduce MMP expression by inhibiting MAPK/AP-1 cell signaling in UVB-treated HS68 cells, we evaluated the effect of ARE on regulation of genes related to collagen synthesis and production of type-I procollagen. The expression of the collagen genes [collagen type I 1 (COL1A1), collagen type III 1 (COL3A1), collagen type IV 1 (COL4A1), and collagen type VII 1 (COL7A1)] was significantly reduced in UVB-treated HS68 cells. However, ARE markedly upregulated mRNA expression of collagen genes (Fig. 3A). ARE treatment prevented UVB irradiation from reducing type-I procollagen production in a dose-dependent manner. Specifically, at a dose of 20 g/mL, ARE increased type-I procollagen content by 27% (Fig. 3B). Consistent with these current results, A. rugosa recovered skin barrier function by increasing expression of collagens in atopic dermatitisinduced mice (Seo, 2014). This study demonstrates that ARE attenuates skin photoaging by activating type-I procollagen production and improving collagen synthesis in UVB-irradiated HS68 cells. ARE inhibits UVB-induced inflammation. Previous study showed that UVB irradiation elevated inflammatory responses by activating NF-B and stimulating expression of NF-B-mediated inflammatory cytokines (Bradley, 2008). UVB-induced inflammatory responses upregulate MMP expression, resulting in skin dehydration, psoriasis, and erythema (Schnegg et al., 2012). Here, mRNA expression of inflammatory mediators was examined in UVB-exposed HS68 cells. UVB irradiation elevated protein expression of NF-B, which was inhibited by ARE (Fig. 4A). In addition, mRNA expression of NF-B-stimulated inflammatory cytokines was decreased in calls treated with ARE, compared with scription-polymerase chain reaction analysis of the expression of collagen type I alpha 1 (COL1A1), collagen type III alpha 1 (COL3A1), collagen type IV alpha 1 (COL4A1), and collagen type VII alpha 1 (COL7A1) after ARE (10 and 20 g/mL) treatment. (B) Type-I procollagen content after ARE (10 and 20 g/mL) treatment. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a housekeeping gene. Data are expressed as mean±SD of three independent experiments. ## P <0.01 compared with the control group. ** P <0.01 compared with the UVB-treated cells. UVB-irradiated HS68 cells (Fig. 4B). The anti-inflammatory effect of ARE was reported in an osteosarcoma cell line, accompanied by downregulation of interleukin-1 and tumor necrosis factor- (Oh et al., 2005). Consistent with this report, our data showed that ARE exhibited strong anti-inflammatory effects, thereby playing a critical role in anti-photoaging property by suppressing expression of NF-B and pro-inflammatory cytokines. Taken together, these results suggest that ARE could be used as an anti-photoaging agent owing to its alleviation of UVBinduced skin inflammatory responses. In this study, the anti-photoaging effect of ARE was investigated in UVB-irradiated human dermal fibroblasts. ARE suppressed UVB-induced oxidative stress by reducing production of ROS and by increasing expression of anti-oxidant enzymes. In addition, ARE attenuated UVBactivated MAPK/AP-1 signaling, resulting in downregulation of MMP expression. Moreover, ARE both upregulated expression of collagen genes and increased production of procollagen. Further, ARE alleviated UVB-stimulated inflammatory responses by suppressing expression of inflammatory cytokines. Collectively, our findings suggest that ARE could be a potential candidate for skin anti-photoaging treatment. To develop ARE as a novel anti-photoaging agent, potentially as an ingredient in functional foods or nutraceuticals, the anti-photoaging effect of ARE should be verified in animal studies and clinical trials.	2020-01-09T14:47:23.094Z	2019-12-01T00:00:00.000	{ "year": 2019, "sha1": "2134eb250f26f1d1f4a89c285bab915292aacee6", "oa_license": "CCBYNC", "oa_url": "https://europepmc.org/articles/pmc6941729?pdf=render", "oa_status": "GREEN", "pdf_src": "PubMedCentral", "pdf_hash": "2134eb250f26f1d1f4a89c285bab915292aacee6", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Medicine", "Chemistry" ] }
10400421	pes2o/s2orc	v3-fos-license	Derived categories of irreducible projective curves of arithmetic genus one We investigate the bounded derived category of coherent sheaves on irreducible singular projective curves of arithmetic genus one. A description of the group of exact auto-equivalences and the set of all t-structures of this category is given. We describe the moduli space of stability conditions, obtain a complete classification of all spherical objects in this category and show that the group of exact auto-equivalences acts transitively on them. Harder-Narasimhan filtrations in the sense of Bridgeland are used as our main technical tool. Introduction The purpose of this paper is to study the structure of the bounded derived category D b coh (E) of coherent sheaves on a singular irreducible projective curve E of arithmetic genus one. In the smooth case, such structure results are easily obtained from Atiyah's description [2] of indecomposable vector bundles over elliptic curves. However, if E has a node or a cusp, some crucial properties fail to hold. This is illustrated by the following Despite these difficulties, the main goal of this article is to find the common features between the smooth and the singular case. A list of such can be found in Remark 5. 13. In Section 2, we review the smooth case and highlight where the properties mentioned above are used. Our approach was inspired by [22]. Atiyah's algorithm to construct indecomposable vector bundles of any slope can be understood as an application of a sequence of twist functors with spherical objects. From this point of view, Atiyah shows that any indecomposable object of D b coh (E) is the image of an indecomposable torsion sheaf under an exact auto-equivalence of D b coh (E). In the case of a singular Weierstraß curve E, as our main technical tool we use Harder-Narasimhan filtrations in D b coh (E), which were introduced by Bridgeland [8]. Their general properties are studied in Section 3. The key result of Section 4 is the preservation of stability under Seidel-Thomas twists [34] with spherical objects. This allows us to show that, like in the smooth case, any category of semi-stable sheaves with fixed slope is equivalent to the category of coherent torsion sheaves on E. In the case of slope zero, this was shown in our previous work [12]. For the nodal case, an explicit description of semi-stable sheaves of degree zero viaétale coverings was given there as well. A combinatorial description of semi-stable sheaves of arbitrary slope over a nodal cubic curve was found by Mozgovoy [25]. On the other hand, a classification of all indecomposable objects of D b coh (E) was presented in [10]. A description of the Harder-Narasimhan filtrations in terms of this classification is a task for future work. However, if the singular point of E is a cusp, the description of all indecomposable coherent torsion sheaves is a wild problem in the sense of representation theory, see for example [15]. Nevertheless, stable vector bundles on a cuspidal cubic have been classified by Bodnarchuk and Drozd [6]. It turns out that semi-stable sheaves of infinite homological dimension are particularly important, because only such sheaves appear as Harder-Narasimhan factors of indecomposable objects in D b coh (E) which are not semi-stable (Proposition 4.6). The main result (Proposition 4.13) of Section 4 is the answer to a question of Polishchuk, who asked in [30], Section 1.4, for a description of all spherical objects on E. We also prove that the group of exact auto-equivalences of D b coh (E) acts transitively on the set of spherical objects. In Section 5 we study t-structures on D b coh (E) and stability conditions in the sense of [8]. We completely classify all t-structures on this category (Theorem 5.6). This allows us to deduce a description of the group of exact auto-equivalences of D b coh (E) (Corollary 5.8). As a second application, we calculate Bridgeland's moduli space of stability conditions on E (Proposition 5.12). The hearts D(θ, θ + 1) of the t-structures constructed in Section 5 are finite-dimensional non-Noetherian Abelian categories of infinite global dimension. In the case of a smooth elliptic curve, this category is equivalent to the category of holomorphic vector bundles on a non-commutative torus in the sense of Polishchuk and Schwarz [32], Proposition 3.9. It is an interesting problem to find such a differentialgeometric interpretation of these Abelian categories in the case of singular Weierstraß curves. Using the technique of Harder-Narasimhan filtrations, we gain new insight into the classification of indecomposable complexes, which was obtained in [10]. It seems plausible that similar methods can be applied to study the derived category of representations of certain derived tame associative algebras, such as gentle algebras, skew-gentle algebras or degenerated tubular algebras, see for example [11]. The study of Harder-Narasimhan filtrations in conjunction with the action of the group of exact auto-equivalences of the derived category should provide new insight into the combinatorics of indecomposable objects in these derived categories. Notation. We fix an algebraically closed field k of characteristic zero. By E we always denote a Weierstraß curve. This is a reduced irreducible curve of arithmetic genus one, isomorphic to a cubic curve in the projective plane. If not smooth, it has precisely one singular point s ∈ E, which can be a node or a cusp. If x ∈ E is arbitrary, we denote by k(x) the residue field of x and consider it as a sky-scraper sheaf supported at x. By D b coh (E) we denote the derived category of complexes of O E -modules whose cohomology sheaves are coherent and which are non-zero only in finitely many degrees. Background: the smooth case The purpose of this section is to recall well-known results about the structure of the bounded derived category of coherent sheaves over a smooth elliptic curve. Proofs of most of these results can be found in [2], [27], [22] and [35]. The focus of our presentation is on the features and techniques which are essential in the singular case as well. At the end of this section we highlight the main differences between the smooth and the singular case. It becomes clear that the failure of Serre duality is the main reason why the proofs and even the formulation of some of the main results do not carry over to the singular case. The aim of the subsequent sections will then be to overcome these difficulties, to find correct formulations which generalise to the singular case and to highlight the common features of the bounded derived category in the smooth and singular case. With the exception of subsection 2.7, throughout this section E denotes a smooth elliptic curve over k. Homological dimension. For any two coherent sheaves F , G on E, Serre duality provides an isomorphism This follows from the usual formulation of Serre duality and the fact that any coherent sheaf has a finite locally free resolution. As a consequence, Ext ν (F , G) = 0 for any ν ≥ 2, which means that Coh E has homological dimension one. This implies that any object X ∈ D b coh (E) splits into the direct sum of appropriate shifts of its cohomology sheaves. To see this, start with a complex X = (F −1 f −→ F 0 ) and consider the distinguished triangle in D b coh (E) ker(f ) [1] → X → coker(f ) ξ → ker(f ) [2]. 2.3. Jordan-Hölder factors. The full sub-category of Coh E whose objects are the semi-stable sheaves of a fixed slope is an Abelian category in which any object has a Jordan-Hölder filtration with stable factors. If F and G are non-isomorphic stable sheaves which have the same slope, we have Hom(F , G) = 0. Based on this fact, in the same way as before, we can deduce that an indecomposable semi-stable sheaf has all its Jordan-Hölder factors isomorphic to each other. 2.4. Simple is stable. It is well-known that any stable sheaf F is simple, i.e. Hom(F , F ) ∼ = k. On a smooth elliptic curve, the converse is true as well, which equips us with a useful homological characterisation of stability. To see that simple implies stable, we suppose for a contradiction that F is simple but not stable. This implies the existence of an epimorphism F → G with G stable and µ(F ) ≥ µ(G). Serre dual- But this produces a non-zero composition F → G → F which is not an isomorphism, in contradiction to the assumption that F was simple. 2.5. Classification. Atiyah [2] gave a description of all stable sheaves with a fixed slope in the form E(r, d) ⊗ L, where L is a line bundle of degree zero and E(r, d) is a particular stable bundle of the fixed slope. The bundle E(r, d) depends on the choice of a base point p 0 ∈ E and its construction reflects the Euclidean algorithm on the pair (rk, deg). We look at this description from a slightly different perspective. We use the twist functors T O and T k(p 0 ) , which were constructed by Seidel and Thomas [34] (see also [24]). They act as equivalences on D b coh (E) and, hence, preserve stability. A stable sheaf of rank r and degree d is sent by T O to one with (rk, deg) equal to (r − d, d). If r < d this is a shift of a stable sheaf. The functor T k(p 0 ) sends the pair (r, d) to (r, r + d) and its inverse sends it to (r, d − r). Therefore, if we follow the Euclidean algorithm, we find a composition of such functors which provides an equivalence between the category of stable sheaves with slope d/r and the category of simple torsion sheaves. Such sheaves are precisely the structure sheaves of closed points k(x), x ∈ E. They are considered to be stable with slope ∞. More generally, this procedure provides an equivalence between the category of semi-stable sheaves of rank r and degree d with the category of torsion sheaves of length equal to gcd(r, d). This shows, in particular, that the Abelian category of semi-stable sheaves with fixed slope is equivalent to the category of coherent torsion sheaves. 2.6. Auto-equivalences. By Aut(D b coh (E)) we denote the group of all exact auto-equivalences of the triangulated category D b coh (E). This group acts on the Grothendieck group K(E) ∼ = K(D b coh (E)). As the kernel of the Chern character is the radical of the Euler-form X, Y = dim(Hom(X, Y )) − dim(Hom(X, Y [1]) which is invariant under this action, it induces an action on the even cohomology H 2 * (E, Z) ∼ = Z 2 . Because dim(Hom(F , G)) > 0 if and only if F , G > 0, provided F ∼ = G are stable sheaves, the induced action on Z 2 is orientation preserving. So, we obtain a homomorphism of groups ϕ : Aut(D b coh (E)) → SL(2, Z), which is surjective because T O and T k(p 0 ) are mapped to a pair of generators of SL(2, Z). Explicitly, if G is an auto-equivalence, ϕ(G) describes its action on the pair (rk, deg). To understand ker(ϕ), we observe that ϕ(G) = 1 implies that G sends a simple torsion sheaf k(x) to some k(y)[2k], because indecomposability is retained. By the same reason, G(O) is a shifted line bundle of degree zero. However, Hom(L, k(y)[l]) = 0, if L is a line bundle and l = 0. Hence, after composing G with a shift, it sends all simple torsion sheaves to simple torsion sheaves, without a shift. Because E is smooth, we can apply a result of Orlov [28] which says that any auto-equivalence G is a Fourier-Mukai transform [26]. However, any such functor, which sends the sheaves k(x) to torsion sheaves of length one is of the form G(X) = f * (L ⊗ X), where f : E → E is an automorphism and L ∈ Pic(E) a line bundle. Hence, ker(ϕ) is generated by Aut(E), Pic 0 (E) and even shifts. This gives a complete description of the group Aut(D b coh (E)). A similar approach was used by Lenzing and Meltzer to describe the group of exact auto-equivalences of tubular weighted projective lines [23]. 2.7. Difficulties in the singular case. Let now E be an irreducible but singular curve of arithmetic genus one. The technical cornerstones of the theory as described in this section fail to be true in this case. More precisely: • the category of coherent sheaves Coh E has infinite homological dimension; • there exist indecomposable complexes in D b coh (E) which are not just shifted sheaves, see [10], section 3; • Serre duality fails to be true in general; • not all indecomposable vector bundles are semi-stable; • there exist indecomposable coherent sheaves which are neither torsion sheaves nor torsion free sheaves, see [10]. Most of the trouble is caused by the failure of Serre duality. The basic example is the following. Suppose, s ∈ E is a node, then Hom(k(s), k(s)) ∼ = k and Ext 1 (k(s), k(s)) ∼ = k 2 . Serre duality is available, only if at least one of the two sheaves involved has finite homological dimension. This might suggest that replacing D b coh (E) by the sub-category of perfect complexes would solve most of the problems. But see Remark 4.10. In the subsequent sections we overcome these difficulties and point out the similarities between the smooth and the singular case. Harder-Narasimhan filtrations Throughout this section, E denotes an irreducible reduced projective curve over k of arithmetic genus one. The notion of stability of coherent torsion free sheaves on an irreducible curve is usually defined with the aid of the slope function µ( · ) = deg( · )/ rk( · ). To use the phase function instead is equivalent, but better adapted for the generalisation to derived categories described below. By definition, the phase ϕ(F ) of a non-zero coherent sheaf F is the unique number which satisfies 0 < ϕ(F ) ≤ 1 and m(F ) exp(πiϕ(F )) = − deg(F ) + i rk(F ), where m(F ) is a positive real number, called the mass of the sheaf F . In particular, ϕ(O) = 1/2 and all non-zero torsion sheaves have phase one. A torsion free coherent sheaf F is called semi-stable if for any exact sequence of torsion free coherent sheaves 0 → E → F → G → 0 the inequality ϕ(E) ≤ ϕ(F ), or equivalently, ϕ(F ) ≤ ϕ(G), holds. It is well-known [33] that any torsion free coherent sheaf F on a projective variety has a Harder-Narasimhan filtration which is uniquely characterised by the property that all factors A i = F i /F i+1 are semi-stable and satisfy Originally, this concept of stability was introduced in the 1960s in order to construct moduli spaces using geometric invariant theory. It could also be seen as a method to understand the structure of the category of coherent sheaves on a projective variety. By Simpson, the notion of stability was extended to coherent sheaves of pure dimension. A very general approach was taken by Rudakov [33], who introduced the notion of stability on Abelian categories. Under some finiteness assumptions on the category, he shows the existence and uniqueness of a Harder-Narasimhan filtration for any object of the category in question. As an application of his work, the usual slope stability extends to the whole category Coh E of coherent sheaves on E. In particular, any non-zero coherent sheaf has a Harder-Narasimhan filtration and any non-zero coherent torsion sheaf on the curve E is semi-stable. Inspired by work of Douglas on Π-Stability for D-branes, see for example [14], it was shown by Bridgeland [8] how to extend the concept of stability and Harder-Narasimhan filtration to the derived category of coherent sheaves, or more generally, to a triangulated category. These new ideas were merged with the ideas from [33] in the paper [19]. We shall follow here the approach of Bridgeland [8]. In Section 5 we give a description of Bridgeland's moduli space of stability conditions on the derived category of irreducible singular curves of arithmetic genus one. However, throughout the present chapter we stick to the classical notion of stability on the category of coherent sheaves and the stability structure it induces on the triangulated category. In order to generalise the concept of a Harder-Narasimhan filtration to the category D b coh (E), Bridgeland [8] extends the definition of the phase of a sheaf to shifts of coherent sheaves by: where F = 0 is a coherent sheaf on E and n ∈ Z. A complex which is non-zero at position m only has, according to this definition, phase in the interval (−m, −m + 1]. If F and F ′ are non-zero coherent sheaves and a, b integers, we have the implication: For any ϕ ∈ R we denote by P(ϕ) the Abelian category of shifted semi-stable sheaves with phase ϕ. Of course, 0 ∈ P(ϕ) for all ϕ. If ϕ ∈ (0, 1], this is a full Abelian subcategory of Coh E . For any ϕ ∈ R we have P(ϕ+n) = P(ϕ) [n]. A non-zero object of D b coh (E) will be called semi-stable, if it is an element of one of the categories P(ϕ), ϕ ∈ R. Bridgeland's stability conditions [8] involve so-called central charges. In order to define the central charge of the standard stability condition, we need a definition of degree and rank for arbitrary objects in D b coh (E). Let K = O E,η be the field of rational functions on the irreducible curve E with generic point η ∈ E. The base change η : Spec(K) → E is flat, so that η * (F ), taken in the non-derived sense, is correctly defined for any F ∈ D b coh (E). We define rk(F ) := χ(η * (F )), which is the alternating sum of the dimensions of the cohomology spaces of the complex η * (F ) which are vector spaces over K. In order to define the degree, we use the functor , and set deg(F ) := χ(R Hom(O E , F )). Here, we denoted by D b coh (k) the bounded derived category of finite dimensional vector spaces over k. For coherent sheaves, these definitions coincide with the usual definitions of rank and degree. In particular, a torsion sheaf of length m which is supported at a single point of E has rank 0 and degree m. These definitions imply that rank and degree are additive on distinguished triangles in D b coh (E). Hence, they induce homomorphisms on the Grothendieck group K(D b coh (E)) of the triangulated category D b coh (E), which is by definition the quotient of the free Abelian group generated by the objects of D b coh (E) modulo expressions coming from distinguished triangles. Recall that K 0 (Coh(E)) ∼ = K(D b coh (E)), see [20]. We denote this group by K(E) Proof. Recall that the Grothendieck-Riemann-Roch Theorem, see [4] or [18], provides a homomorphism which depends functorially on E with respect to proper direct images. Moreover, (τ E ) Q : K(E) ⊗ Q → A * (E) ⊗ Q is an isomorphism, see [18], Cor. 18.3.2. If E is an irreducible singular projective curve of arithmetic genus one, we easily see that the Chow group A * (E) is isomorphic to Z 2 . The two generators are [x] ∈ A 0 (E) with x ∈ E and [E] ∈ A 1 (E). Note that [x] = [y] ∈ A 0 (E) for any two closed points x, y ∈ E, because the normalisation of E is P 1 . Using [18], Thm. 18.3 (5), we obtain τ E (k(x)) = [x] ∈ A 0 (E) for any x ∈ E. On the other hand, from [18], Expl. 18.3.4 (a), we obtain τ E (O E ) = [E] ∈ A 1 (E). Therefore, the classes of k(x) and O E define a basis of K(E) ⊗ Q. However, these two classes generate the group K(E), so that it must be a free Abelian group. The central charge of the standard stability structure on D b coh (E) is the homomorphism of Abelian groups which is given by If F is a non-zero coherent sheaf, Z(F ) is a point on the ray from the origin through exp(πiϕ(F )) in C. Its distance from the origin was called the mass of F . Although the phase ϕ(F ) is defined for sheaves and their shifts only, we are able to define the slope µ(F ) for any object in D b coh (E) which is not equal to zero in the Grothendieck group. Namely, the usual definition µ(F ) := deg(F )/ rk(F ) gives us now a mapping which extends the usual definition of the slope of a sheaf. Because Z(O E ) = i and Z(k(x)) = −1, Lemma 3.1 implies that Z is injective. Therefore, µ is defined for any non-zero element of the Grothendieck group. For arbitrary objects X ∈ D b coh (E) we have Z(X[1]) = −Z(X), hence µ(X[1]) = µ(X) when defined. In case of shifted sheaves, in contrast to the slope µ, the phase ϕ keeps track of the position of this sheaf in the complex. As an illustration, we include an example of an indecomposable object in D b coh (E) which has a zero image in the Grothendieck group. Example 3.2. Let s ∈ E be the singular point and denote, as usual, by k(s) the torsion sheaf of length one which is supported at s. This sheaf does not have finite homological dimension. To see this, we observe first that Ext k (k(s), k(s)) ∼ = H 0 (Ext k (k(s), k(s))). Moreover, as an O E,s -module, k(s) has an infinite periodic locally free resolution of the form For example, if s is a node, so that E is locally given by considered modulo f . More generally, any singular Weierstraß cubic f can be written as y·y−R·S with y, R, S all vanishing at the singular point. The off-diagonal elements of A and B are then formed by ±R, ±S. Therefore, all entries of the matrices A and B are elements of the maximal ideal of the local ring O E,s . Hence, the application of Hom( · , k(s)) produces a complex with zero differential, which implies that Ext k (k(s), k(s)) is two-dimensional for all k ≥ 1. In particular, Ext 2 (k(s), k(s)) ∼ = k 2 , and we can pick a non-zero element w ∈ Hom(k(s), k(s) [2]). There exists a complex X ∈ D b coh (E) which sits in a distinguished triangle Because the shift by one corresponds to multiplication by −1 in the Grothendieck group, this object X is equal to zero in K(E). On the other hand, X is indecomposable. Indeed, if X would split, it must be X ∼ = k(s) ⊕ k(s) [1], because the only non-zero cohomology of X is If all ingredients of a HNF are shifted by one, we obtain a HNF of X [1]. The shifted sheaves A j are called the semi-stable HN-factors of X and we define ϕ + (X) := ϕ n and ϕ − (X) := ϕ 0 . Later, Theorem 3.10, we show that the HNF of an object X is unique up to isomorphism. This justifies this notation. For the moment, we keep in mind that ϕ + (X) and ϕ − (X) might depend on the HNF and not only on the object X. Before we proceed, we include a few remarks about the notation we use. Distinguished triangles in a triangulated category are either displayed in the form where the arrow which is marked with + is in fact a morphism Z → X [1]. We shall use the octahedron axiom, the axiom (TR4) in Verdier's list, in the following convenient form: if two morphisms X f −→ Y g −→ Z are given, for any three distinguished triangles with bases f, g and g • f there exists a fourth distinguished triangle which is indicated below by dashed arrows, such that we obtain the following commutative diagram: S S S S S S Proof. If we apply the octahedron axiom to the composition A → V → U [1] we obtain the following commutative diagram, which gives the claim. Proof. The first statement is clear, because we can cut off the HNF where the first arrow is the composition of the morphisms in the HNF of V . Using the octahedron axiom, we obtain for any i ≤ k a commutative diagram which implies the second claim. Remark 3.6. The statement of Lemma 3.5 is true with identical proof if we relax the assumption of being a HNF by allowing ϕ(A k ) = ϕ(A k−1 ) for the chosen value of k. is a sheaf, then H k (X) = 0. In particular, the following implication is true: coh (E) has a HNF. Proof. The existence of a HNF for objects of Coh E is classically known, see [21,33]. Therefore, we can proceed by induction on the number of non-zero cohomology sheaves of X ∈ D b coh (E). If n is the largest integer with H n (X) = 0, we have a distinguished triangle We prove now for any distinguished triangle is a coherent sheaf that the existence of a HNF for U with ϕ − (U) > ϕ + (V ) implies the existence of a HNF of X. Because V [n] is a sheaf, V has a HNF and we proceed by induction on the number of HN-factors of V . Let A be the leftmost object in a HNF of V , i.e. A ∈ P(ϕ + (V )). By Lemma 3.4 applied to the distinguished triangles (2) and is a coherent sheaf with a smaller number of HN-factors as V : , the left triangle can be concatenated to the given HNF of U in order to provide a HNF for W . The start of the induction is covered as well: it is the case V ′ = 0. Proof. If X, Y are semi-stable sheaves, this is well-known and follows easily from the definition of semi-stability. Because Hom(X, Y [k]) = 0, if X, Y are sheaves and k < 0, the claim follows if X ∈ P(ϕ) and Theorem 3.10 ( [8,19]). The HNF of any non-zero object are HNFs of X, we have to show that there exist unique isomorphisms of distinguished triangles for any k ≥ 0 Without loss of generality, we may assume ϕ ≥ ψ. This implies [5], Proposition 1.1.9, we obtain the existence and uniqueness of the morphisms f ′ , g. It remains to show that they are isomorphisms. If g were zero, the second morphism in the triangle . Hence, g = 0 and Lemma 3.9 implies ϕ(A) ≤ ϕ(B), i.e. ϕ = ψ. So, the same reasoning as before gives a unique morphism of distinguished triangles in the other direction. The composition of both are the respective identities which follows again from the uniqueness part of [5], Proposition 1.1.9. This proves the claim. We need the following useful lemma. Lemma 3.11. ([29], Lemma 2.5) Let D be a triangulated category and be a distinguished triangle in D. Then G ∼ = H 1 ⊕G ′ splits and the given triangle is isomorphic to The results in this section are true for more general triangulated categories than D b coh (E). Without changes, the proofs apply if we replace D b coh (E) by the bounded derived category of an Abelian category which is equipped with the notion of stability in the sense of [33]. In particular, these results hold for polynomial stability on the triangulated categories D b coh (X) where X is a projective variety over k. The structure of the bounded derived category of coherent sheaves on a singular Weierstaß curve In this section, we prove the main results on which our understanding of D b coh (E) is based. Again, E denotes a Weierstraß curve. Our main focus is on the singular case, however all the results remain true in the smooth case as well. A speciality of this category is the nonvanishing result Proposition 4.12. Unlike the smooth case, there exist indecomposable objects in D b coh (E), which are not semi-stable. Their Harder-Narasimhan factors are characterised in Proposition 4.6. We propose to visualise indecomposable objects by their "shadows". As an application of our results, we give a complete characterisation of all spherical objects in D b coh (E). As a consequence, we show that the group of exact auto-equivalences acts transitively on the set of spherical objects. This answers a question which was posed by Polishchuk [30]. Let us set up some notation. For any ϕ ∈ (0, 1] we denote by P(ϕ) s ⊂ P(ϕ) the full subcategory of stable sheaves with phase ϕ. We extend this definition to all ϕ ∈ R by requiring P(ϕ + n) s = P(ϕ) s [n] for all n ∈ Z and all ϕ ∈ R. We already know the structure of P(1) s . Because P(1) is the category of coherent torsion sheaves on E, the objects of P(1) s are precisely the structure sheaves k(x) of closed points x ∈ E. In order to understand the structure of all the other categories P(ϕ) s , we use Fourier-Mukai transforms. Our main technical tool will be the transform F which was studied in [12]. It depends on the choice of a regular point p 0 ∈ E. Let us briefly recall its definition and main properties. It was defined with the aid of Seidel-Thomas twists [34], which are functors coh (E) depending on a spherical object E ∈ D b coh (E). On objects, these functors are characterised by the existence of a distinguished . In [34] is was shown that twist functors can be described as integral transforms and that F is isomorphic to the functor FM P , which is given by . This is a shift of a coherent sheaf on E × E, on which we denote the ideal of the diagonal by I ∆ ⊂ O E×E and the two projections by π 1 , π 2 . In order to understand the effect of F on rank and degree, we look at the distinguished triangle The additivity of rank and degree implies rk ] as a basis of K(E), which means that we use coordinates (rk, − deg), then the action of T O , T k(p 0 ) and of F on K(E) is given by the matrices In particular, for any object F ∈ D b coh (E) which has a slope, we have µ(T k(p 0 ) (F )) = µ(F ) + 1 and µ(F(F )) = − 1 µ(F ) using the usual conventions in dealing with ∞. If F is a sheaf or a twist thereof, we defined the phase ϕ(F ). In order to understand the effect of F on phases, it is not sufficient to know its effect on the slope. This is because the slope determines the phase modulo 2Z only. However, if F is a coherent sheaf, the description of F as FM P shows that F(F ) can have non-vanishing cohomology in degrees −1 and 0 only. If, in addition, F(F ) is a shifted sheaf, this implies ϕ(F(F )) ∈ (0, 2]. From the formula for the slope it is now clear that ϕ(F(F )) = ϕ(F ) + 1 2 for any shifted coherent sheaf F . The following result was first shown in [3]. We give an independent proof here, which was inspired by [3], Lemma 3.1. Proof. Note that, by definition, a semi-stable sheaf of positive rank is automatically torsion free. The only sheaf with degree and rank equal to zero is the zero sheaf. Throughout this proof, we let F be a semistable sheaf on E. If deg(F ) = 0 this sheaf is torsion free and the claim was shown in [12], Thm. 2.21, see also [17]. For the sake of clarity we would like to stress here the fact that [12], Section 2, deals with nodal as well as cuspidal Weierstraß curves. Next, suppose deg(F ) > 0. If rk(F ) = 0, F is a coherent torsion sheaf. Again, the claim follows from [12], Thm. 2.21 and Thm. 2.18, where it was shown that F • F = i * [1], for any Weierstraß curve. Here, i : E → E is the involution which fixes the singularity and which corresponds to taking the inverse on the smooth part of E with its group structure in which p 0 is the neutral element. Therefore, we may suppose F is torsion free. As observed before, the complex F(F ) ∈ D b coh (E) can have non-vanishing cohomology in degrees −1 and 0 only. We are going to show that F(F )[−1] is a sheaf, which is equivalent to the vanishing of the cohomology object H 0 (F(F )) ∈ Coh E . Recall from [12], Lemma 2.13, that for any smooth point where x ∈ E is an arbitrary smooth point. These vector spaces vanish as F was assumed to be semi-stable and of positive degree. Because cohomology of the complex F(F ) vanishes in positive degree, there is a canonical morphism F(F ) → H 0 (F(F )) in D b coh (E), which induces an injection of functors Hom(H 0 (F(F )), · ) ֒→ Hom(F(F ), · ). Therefore, the vanishing which was obtained above, shows Hom(H 0 (F(F )), k(y)) = 0 for any point y ∈ E. This implies the vanishing of the sheaf H 0 (F(F )). Hence, F := F(F )[−1] is a coherent sheaf and the definition of F implies that there is an exact sequence of coherent sheaves This sequence implies, in particular, that F is torsion free. Before we proceed to show that F is semi-stable, we apply duality to prove that F(F ) is a sheaf if deg(F ) < 0. Let us denote the dualising functor by D := RHom( · , O E ). This functor satisfies DD ∼ = 1. In [13], Cor. 3.4, we have shown that there exists an isomorphism Because F is a torsion free sheaf on a curve, it is Cohen-Macaulay and since E is Gorenstein, this implies Ext i (F , O) = 0 for any i > 0. Therefore, we have D(F ) ∼ = F ∨ and this is a semi-stable coherent sheaf of positive degree. Thus, [−1] • F sends F ∨ to a torsion free sheaf, on which D is just the usual dual. Now, we see that F(F ) is a torsion free sheaf if F was semi-stable and of negative degree. It was shown in [12] that we obtain an action of the group SL(2, Z) on D b coh (E) by sending generators of this group to T O , T k(p 0 ) and the translation functor [1] respectively. Let us denote Q := {ϕ ∈ R \| P(ϕ) contains a non-zero object}. The action of a group G on Q is called monotone, if ϕ ≤ ψ implies g · ϕ ≤ g · ψ for every g ∈ G and ϕ, ψ ∈ Q. coh (E) induces a monotone and transitive action on the set Q. All isotropy groups of this action are isomorphic to Z. Proof. As seen above, for any ψ ∈ Q and 0 = A ∈ P(ψ), we have ϕ(F(A)) = ϕ(A) + 1 2 and µ(T k(p 0 ) (A)) = µ(A) + 1. Therefore, by Theorem 4.1 it is clear that we obtain an induced monotone action of SL(2, Z) on Q. The group SL(2, Z) acts transitively on the set of all pairs of co-prime integers which we interpret as primitive vectors of the lattice Z ⊕ iZ ⊂ C. Hence, the action of SL(2, Z) on Q is transitive as well. So, all isotropy groups are isomorphic. Finally, it is easy to see that the isotropy group of 1 ∈ Q is generated by T k(p 0 ) . As an important consequence we obtain the following clear structure result for the slices P(ϕ). Corollary 4.3. The category P(ϕ) of semi-stable objects of phase ϕ ∈ Q is equivalent to the category P(1) of torsion sheaves. Any such equivalence restricts to an equivalence between P(ϕ) s and P(1) s . Under such an equivalence, stable vector bundles correspond to structure sheaves of smooth points. Moreover, if ϕ ∈ (0, 1)∩Q, P(ϕ) s contains a unique torsion free sheaf, which is not locally free. It correspond to the structure sheaf k(s) ∈ P(1) s of the singular point. Recall that an object E ∈ D b coh (E) is called perfect, if it is isomorphic in the derived category to a bounded complex of locally free sheaves of finite rank. Thus, a sheaf or shift thereof is called perfect, if it is perfect as an object in D b coh (E). If E is smooth, any object in D b coh (E) is perfect. However, if s ∈ E is a singular point, the torsion sheaf k(s) is not perfect. If E is singular with one singularity s ∈ E, the category P(1) s contains precisely one object which is not perfect, the object k(s). Hence, by Proposition 4.2, for any ϕ ∈ Q there is precisely one element in P(ϕ) s which is not perfect. We shall refer to it as the extreme stable element with phase ϕ. So, the sheaf k(s) is the extreme stable element with phase 1. The extreme stable element is never locally free. A stable object is either perfect or extreme. We shall need the following version of Serre duality, which can be deduced easily from standard versions: If E, F ∈ D b coh (E) and at least one of them is perfect, then there is a bi-functorial isomorphism If neither of the objects is perfect, this is no longer true. For example, Hom(k(s), k(s)) ∼ = k, but Hom(k(s), k(s) [1]) ∼ = Ext 1 (k(s), k(s)) ∼ = k 2 . Any object X in the Abelian category P(ϕ) has a Jordan-Hölder filtration (JHF) Observe that for any two objects J ∼ = J ′ ∈ P(ϕ) s we can apply Serre duality because at most one of them is non-perfect. Proof. (1) This follows from Proposition 4.2 because the shift functor corresponds to an element in the centre of SL(2, Z) and therefore Φ(P(ϕ)) = P(1) implies Φ(P(ϕ − 1)) = P(0). (2) The statement is clear in case ϕ = 1 and follows from (1) in the general case. (3) Using (1) we can assume ψ = 1, which means that Y is a coherent torsion sheaf. By Proposition 4.2 this implies ϕ ∈ (0, 1) and X is a torsion free coherent sheaf. If Y ∈ P(1) s the statement is clear, because any torsion free sheaf has a non-zero morphism to any Y = k(x), x ∈ E. If Y ∈ P(1) is arbitrary, there exists a point x ∈ E and a nonzero morphism k(x) → Y . The claim follows now from left-exactness of the functor Hom(X, · ). (4) If J ′ ∈ P(ϕ) s is a JH-factor of X, we have J ∼ = J ′ . From (2) and Serre duality together with Lemma 3.9 we obtain Hom(J, J ′ [i]) = 0 for any i ∈ Z. Using the JHF of X, the claim now follows. (5) It is easy to prove by induction that any X ∈ P(ϕ) can be split as a finite direct sum X ∼ = ⊕X k , where each X k has all JH-factors isomorphic to a single element J k ∈ P(ϕ) s . This implies, the claim. (6) By (1) we may assume ϕ(X) = ϕ(Y ) = 1. This means, both objects are indecomposable torsion sheaves with support at the singular point s ∈ E. Such sheaves always have an epimorphism to and a monomorphism from the extreme object k(s), hence the claim. It is interesting and important to note that an indecomposable semistable object can be perfect even though all its JH-factors are extreme. This is made explicit in [12], Section 4, in the case of the category P(1) of coherent torsion sheaves. If E is nodal, there are two kinds of indecomposable torsion sheaves with support at the node s ∈ E: the so-called bands and strings. The bands are perfect, whereas the strings are not perfect. Using the action of SL(2, Z) this carries over to all other categories P(ϕ) with ϕ ∈ Q. An object X ∈ P(ϕ) will be called extreme if it does not have a direct summand which is perfect. This implies that, but is not equivalent to the property that all its JH-factors are extreme. An example can be found below, see Ex. 4.9. From the above we deduce that any X ∈ P(ϕ) can be split as a direct sum X ∼ = X e ⊕ X p with X e extreme and X p perfect. All direct summands of the extreme part have the unique extreme stable element with phase ϕ as its JH-factors. On the other hand, all the direct summands of X p are perfect and they can have any object of P(ϕ) s as JH-factor. Proof. The assumption implies that F is indecomposable. If F were not even semi-stable, it would have at least two HN-factors. Using Corollary 4.4, we may assume that ϕ + (F ) = 1. Thus, F is a coherent sheaf which is neither torsion nor torsion free. This implies that there is a non-invertible endomorphism F → k(s) → tors(F ) → F , in contradiction to the assumption. Hence, F ∈ P(ϕ) is semi-stable. Let J ∈ P(ϕ) be its JH-factor. From Corollary 4.4 (6) we obtain a nonzero endomorphism F → J → F , which can only be an isomorphism, if F ∼ = J , so F is indeed stable. The following method can be used to visualise the structure of the category D b coh (E): the vertical slices in Figure 1 are thought to correspond to the categories P(t) s of stable objects. They are non-empty if Any indecomposable object 0 = X ∈ D b coh (E) has a shadow in such a picture: it is the set of all stable objects which occur as JH-factors in the HN-factors of X. If this set consists of more than one point, the shadow is obtained by connecting these points by line segments. The following proposition shows that the shadow of an indecomposable object which consists of more than one point is completely contained in the extreme line. shadows Figure 2 shows the shadows of five different indecomposable objects: the shift of an indecomposable semi-stable locally free sheaf, • X 3 a genuine complex with three extreme HN-factors, one in Coh E [2] and the other two in Coh E [1], • X 4 an indecomposable torsion free sheaf which is not semistable, • X 5 ∈ Coh E an indecomposable and semi-stable torsion free sheaf which could be perfect or not (a band or a string in the language of representation theory). The shadow of an indecomposable object is a single point if and only if this object is semi-stable. Proposition 4.6. Let X ∈ D b coh (E) be an indecomposable object which is not semi-stable. Then, all HN-factors of X are extreme. be a HNF of X. If the HN-factor A i were not extreme, it could be split into a direct sum Lemma 3.9 and Serre duality imply Hence, we can apply Lemma 3.11 to the distinguished triangle We proceed by descending induction on j ≤ i to show that there exist decompositions F j X ∼ = F ′ j X ⊕ A ′ i . This is obtained from Lemma 3.11 applied to the distinguished triangle and using Lemma 3.9, Serre duality and ϕ(A ′ i ) > ϕ(A j−1 ) to get Because X was assumed to be indecomposable, we should have X ∼ = A ′ i , but this was excluded by assumption. This contradiction shows that all HN-factors A i are necessarily extreme. (1) semi-stable with perfect JH-factor; (2) semi-stable, perfect but its JH-factor extreme; ( 3) semi-stable and extreme; (4) not semi-stable, with all its HN-factors extreme. A similar statement is true for D b coh (E). In this case, the objects of types (1), (2) and (3) are shifts of coherent sheaves, whereas genuine complexes are possible for objects of type (4). Types (2), (3) and (4) were not available in the smooth case. Examples of type (1) are simple vector bundles and structure sheaves k(x) of smooth points x ∈ E. All indecomposable objects with a shadow not on the extreme line fall into type (1). Under the equivalences of Corollary 4.3, indecomposable semi-stable locally free sheaves with extreme JH-factor correspond, in the nodal case, precisely to those torsion sheaves with support at the node s, which are called bands, see [12]. Examples of type (3) are the stable coherent sheaves which are not locally free and the structure sheaf k(s) of the singular point s ∈ E. Moreover, in the nodal case, the torsion sheaves with support at s, which are called strings in [12], are of type (3) as well. Examples of objects of type (4) are given below. Recall from [16] that any indecomposable torsion free sheaf which is not locally free, is isomorphic to a sheaf S(d) = p n * L(d). We use here the notation of [12], Section 3.5, so that p n : I n → E denotes a certain morphism from the chain I n of n smooth rational curves to the nodal curve E. If d = (d 1 , . . . , d n ) ∈ Z n , we denote by L(d) the line bundle on I n which has degree d ν on the ν-th component of I n . We know rk(S(d)) = n and deg(S(d)) = 1 + d ν . We obtain, in particular, that for any ϕ ∈ Q ∩ (0, 1) there exist n ∈ Z and d(ϕ) ∈ Z n such that S(d(ϕ)) is the unique extreme element in P(ϕ) s . On the other hand, if d ′ ∈ Z n ′ , d ′′ ∈ Z n ′′ and d = (d ′ + , d ′′ ) ∈ Z n ′ +n ′′ , where d ′ + is obtained from d ′ by adding 1 to the last component, we have an exact sequence 0 → S(d ′ ) → S(d) → S(d ′′ ) → 0 see for example [25]. Hence, if we start with a sequence 0 < ϕ 0 < ϕ 1 < . . . < ϕ m < 1 where ϕ ν ∈ Q and define we obtain an indecomposable torsion free sheaf S(d (0) ) whose HNfactors are the extreme stable sheaves S(d(ϕ ν )) ∈ P(ϕ ν ), 0 ≤ ν ≤ m. The HNF of this sheaf is given by The sheaf S(d (0) ) is of type (4) and not perfect. Example 4.9. Suppose E is nodal and let π : C 2 → E be anétale morphism of degree two, where C 2 denotes a reducible curve which has two components, both isomorphic to P 1 and which intersect transversally at two distinct points. By i ν : P 1 → E, ν = 1, 2 we denote the morphisms which are induced by the embeddings of the two components of C 2 . There is a k × -family of line bundles on C 2 , whose restriction to one component is O P 1 (−2) and to the other is O P 1 (2). The element in k × corresponds to a gluing parameter over one of the two singularities of C 2 . If L denotes one such line bundle, E := π * L is an indecomposable vector bundle of rank two and degree zero on E. Let us fix notation so that i * There is an exact sequence of coherent sheaves on E Because the torsion free sheaves i 2 * O P 1 and i 1 * O P 1 (−2) have rank one and E is irreducible, they are stable. Because ϕ(i 2 * O P 1 ) = 3/4 and ϕ(i 1 * O P 1 (−2)) = 1/4, Theorem 3.10 implies that the HNF of E is given by the exact sequence (4). The HN-factors are the two torsion free sheaves of rank one i 2 * O P 1 and i 1 * O P 1 (−2), which are not locally free. These are the extreme stable elements with phases 3/4 and 1/4 respectively. Therefore, the indecomposable vector bundle E is a perfect object of type (4) which satisfies ϕ − (E) = 1/4 and ϕ + (E) = 3/4. Remark 4.10. This example shows that the full sub-category of perfect complexes in the category D b coh (E) is not closed under taking Harder-Narasimhan factors. We interpret this to be an indication that the derived category of perfect complexes is not an appropriate object for homological mirror symmetry on singular Calabi-Yau varieties. Remark 4.11. It seems plausible that methods similar to those of this section could be applied to study the derived category of representations of certain derived tame associative algebras. Such may include gentle algebras, skew-gentle algebras and degenerated tubular algebras. The study of Harder-Narasimhan filtrations in conjunction with the action of the group of exact auto-equivalences of the derived category may provide new insight into the combinatorics of indecomposable objects in these derived categories. Proof. If X and Y are semi-stable objects, the claim (1) was proved in Corollary 4.4 (3). Similarly, (2) for two semi-stable objects follows from Corollary 4.4 (6), because there is only one non-perfect object in P(ϕ) s . For the rest of the proof we treat both cases, (1) and (2) simultaneously. For the proof of (2) we keep in mind that Proposition 4.6 implies that no HN-factor has a perfect summand, if the object is indecomposable but not semi-stable. If X ∈ P(ϕ) is semi-stable but Y ∈ D b coh (E) is arbitrary, we let Hence, by Lemma 3.9, Hom(X, B i [−1]) = 0 and the cohomology se- Finally, in the general case, we let gives us now an inclusion 0 = Hom(A 0 , Y ) ⊂ Hom(X, Y ) and so the claim. In [30], Polishchuk asked for the classification of all spherical objects in the bounded derived category of a singular projective curve of arithmetic genus one. Below, we shall solve this problem for irreducible curves. Let E be an irreducible projective curve of arithmetic genus one over our base field k. Recall that in this case an object Proposition 4.13. Let E be an irreducible projective curve of arithmetic genus one and X ∈ D b coh (E). Then the following are equivalent: (1) X is spherical; (3) X is perfect and stable; (4) there exists n ∈ Z such that X[n] is isomorphic to a simple vector bundle or to a torsion sheaf of length one which is supported at a smooth point of E. In particular, the group of exact auto-equivalences of D b coh (E) acts transitively on the set of all spherical objects. We observed earlier that all the JH-factors of an indecomposable semi-stable object are isomorphic to each other. Therefore, any indecomposable semi-stable object which is not stable has a space of endomorphisms of dimension at least two. So, we conclude X ∈ P(ϕ) s for some ϕ ∈ R. Because Hom(k(s), k(s) [2]) ∼ = Ext 2 (k(s), k(s)) = 0, the transitivity of the action of SL(2, Z) on the set Q implies that none of the extreme stable objects satisfies the condition (2). Hence, X is perfect and stable. To prove (3)⇒(1), we observe that the group of automorphisms of the curve E acts transitively on the regular locus E \ {s}. Hence, by Proposition 4.2, the group of auto-equivalences of D b coh (E) acts transitively on the set of all perfect stable objects. Because, for example, the structure sheaf O E is spherical, it is now clear that all perfect stable objects are indeed spherical and that the group of exact auto-equivalences of D b coh (E) acts transitively on the set of all spherical objects. To show the equivalence with (4), it remains to recall that any perfect coherent torsion free sheaf on E is locally free. This follows easily from the Auslander-Buchsbaum formula because we are working in dimension one. Description of t-structures in the case of a singular Weierstraß curve The main result of this section is a description of all t-structures on the derived category of a singular Weierstraß curve E. This generalises results of [19] and [31], where the smooth case was studied. As an application, we obtain a description of the group Aut(D b coh (E)) of all exact auto-equivalences of D b coh (E). A second application is a description of Bridgeland's space of stability conditions on E. Recall that a t-structure on a triangulated category D is a pair of full subcategories (D ≤0 , D ≥0 ) such that, with the notation D ≥n := D ≥0 [−n] and D ≤n := D ≤0 [−n] for any n ∈ Z, the following holds: (1) D ≤0 ⊂ D ≤1 and D ≥1 ⊂ D ≥0 ; (2) Hom(D ≤0 , D ≥1 ) = 0; (3) for any object X ∈ D there exists a distinguished triangle ) is a t-structure then A = D ≤0 ∩D ≥0 has a structure of an Abelian category. It is called the heart of the t-structure. In this way, t-structures on the derived category D b coh (E) lead to interesting Abelian categories embedded into it. The natural t-structure on D b coh (E) has D ≤n equal to the full subcategory formed by all complexes with nonzero cohomology in degree less or equal to n only. Similarly, the full subcategory D ≥n consists of all complexes X with H i (X) = 0 for all i < n. The heart of the natural t-structure is the Abelian category Coh E . In addition to the natural t-structure we also have many interesting t-structures on D b coh (E). In order to describe them, we introduce the following notation. We continue to work with the notion of stability and the notation introduced in the previous section. If P ⊂ P(θ) s is a subset, we denote by D[P, ∞) the full subcategory of D b coh (E) which is defined as follows: if and only if X = 0 or all its HN-factors, which have at least one JH-factor which is not in P, have phase ϕ > θ. Similarly, D(−∞, P] denotes the category which is generated by P and all P(ϕ) with ϕ < θ. Proposition 5.1. Let θ ∈ R and P(θ) − ⊂ P(θ) s be arbitrary. Denote by P(θ) + = P(θ) s \ P(θ) − the complement of P(θ) − . Then, The heart A(θ, P(θ) − ) of it is the category D[P(θ) − , P(θ) + [1]], which consists of those objects X ∈ D b coh (E) whose HN-factors either have phase ϕ ∈ (θ, θ + 1) or have all its JH-factors in P(θ) − or P(θ) + [1]. Proof. The only non-trivial property which deserves a proof is (3) in the definition of t-structure. Given X ∈ D b coh (E), we have to show that there exists a distinguished triangle A → X → B + → with A ∈ D ≤0 and B ∈ D ≥1 . In order to construct it, let In particular, F k X ∈ D ≤0 . In this case, we define A := F k X and let A = F k X → X be the composition of the morphisms in the HNF. If, however, ϕ(A k ) = θ, there is a splitting A k ∼ = A − k ⊕ A + k such that all JH-factors of A − k (resp. A + k ) are in P(θ) − (resp. P(θ) + ). Now, we apply Lemma 3.4 to the distinguished triangles given by the splitting of A k , to obtain a factorisation F k+1 X → A → F k X of f and two distinguished triangles Part of the given HNF of X together with the left one of these two triangles form a HNF of A, whence A ∈ D ≤0 . Again, we let A → X be obtained by composition with the morphisms in the HNF of X. In any case, we choose a distinguished triangle A → X → B + →, where A → X is the morphism chosen before. From Lemma 3.5 or Remark 3.6 we obtain B ∈ D ≥1 . This proves the proposition. We shall also need the following standard result. and B ∈ D ≥1 , which exists due to the definition of a tstructure. If X ∈ D ≤0 , we necessarily have g = 0 and B = 0. Because Hom(D ≤0 , D ≥1 ) = 0, the composition X ⊕ Y p −→ X g −→ B, in which p denotes the natural projection, must be zero. If i : X → X ⊕Y denotes the canonical morphism, we obtain g = g • p • i = 0, a contradiction. In the same way it follows that Y ∈ D ≤0 . Recall that an Abelian category is called Noetherian, if any sequence of epimorphisms stabilises, this means that for any sequence of epimorphisms f k : A k → A k+1 there exists an integer k 0 such that f k is an isomorphism for all k ≥ k 0 . Proof. If P(θ) = {0} then A(θ, P(θ) − ) = D(θ, θ + 1). This category is not Noetherian. To prove this, we follow the proof of Polishchuk in the smooth case [31], Proposition 3.1. We are going to show for any nonzero locally free shifted sheaf E ∈ D(θ, θ + 1), the existence of a locally free shifted sheaf F and an epimorphism E ։ F in D(θ, θ + 1), which is not an isomorphism. This will be sufficient to show that D(θ, θ + 1) is not Noetherian. By applying an appropriate shift, we may assume 0 < θ < 1. Under this assumption, for every stable coherent sheaf G we have For any two objects X, Y ∈ D b coh (E) we define the Euler form to be which is the imaginary part of Z(X)Z(Y ). If X and Y are coherent sheaves and one of them is perfect, we have This remains true, if we apply arbitrary shifts to the sheaves X, Y , where we understand Let E ∈ D(θ, θ+1) be an arbitrary non-zero locally free shifted sheaf. We look at the strip in the plane between the lines L(0) := R exp(iπθ) and L(E) := L(0) + Z(E). This strip must contain lattice points in its interior. Therefore, there exists a lattice point Z F in this strip which (1) the only lattice points on the closed triangle whose vertices are 0, Z(E), Z F , are its vertices; (2) ϕ F > ϕ(E). By ϕ F we denote here the unique number which satisfies θ < ϕ F < θ+1 and Z F ∈ R exp(iπϕ F ). Because SL(2, Z) acts transitively on Q, there exists a stable non-zero locally free shifted sheaf F ∈ D(θ, θ + 1) with Z(F ) = Z F and ϕ(F ) = ϕ F . The assumption P(θ) = {0} implies R exp(iπθ) ∩ Z 2 = {0}, hence, Z(E) is the only lattice point on the line L(E). This implies that Z(F ) is not on the boundary of the stripe between L(0) and L(E). In particular, Z(E) − Z(F ) is contained in the same half-plane of L(0) as Z(E) and Z(F ), see Figure 3. Condition (1) implies E, F = 1. Because E is locally free, condition (2) implies Hence, Hom(E, F ) ∼ = k. The evaluation map gives, therefore, a distinguished triangle . Because E is a stable non-zero shifted locally free sheaf, it is spherical by Proposition 4.13 and so T E is an equivalence. This implies that T E (F ) is spherical and, by Proposition 4.13 again, C is a stable non-zero shifted locally free sheaf. All morphisms in the distinguished triangle (5) are non-zero because C, E, F are indecomposable, see Lemma 3.11. Using Lemma 3.9, this implies θ − 1 < ϕ(C) < θ + 1. However, we have seen in which half-plane Z(C) is contained, so that we must have θ < ϕ(C) < θ + 1, which implies C ∈ D(θ, θ +1). The distinguished triangle (5) and the definition of the structure of Abelian category on the heart D(θ, θ + 1) imply now that the morphism E → F in (5) is an epimorphism in D(θ, θ + 1). This gives an infinite chain of epimorphisms which are not isomorphisms, so that the category D(θ, θ + 1) is indeed not Noetherian. In order to show that A(θ, P(θ) − ) is not Noetherian for P(θ) − = ∅ we may assume θ = 0. If there exists a stable element k(x) ∈ P(0) − [1] ⊂ P(1), where x ∈ E is a smooth point, we have exact sequences in Coh E . Because B is a semi-stable torsion sheaf with support at s, all its JH-factors are isomorphic to k(s) and we conclude as above. and Y ∈ D ≥1 . Suppose X = 0 and Y = 0 in D b coh (E). We decompose both objects into indecomposables X = X i and Y = Y j . By Lemma 5.2 we have X i ∈ D ≤0 and Y j ∈ D ≥1 . If one of the components of the morphisms were zero, by Lemma 3.11 we would obtain a direct summand X i or Y j in B. Because B was assumed to be indecomposable, this implies the claim of the proposition. For the rest of the proof we suppose that all components of these two morphisms are non-zero. This implies that X i and Y j are non-perfect for all i, j. Indeed, if X i were perfect, we could apply Serre duality (3) to obtain Hom(Y, X i [1]) = Hom(X i , Y ) * , which is zero because X i ∈ D ≤0 and Y ∈ D ≥1 . The case with perfect Y j can be dealt with similarly. Using Lemma 3.11 again, it follows that none of the components of f : X i → B or g : B → Y j is zero, because none of the X i could be a direct summand of Y [−1] and none of the Y j could be a summand of X [1]. Using Lemma 3.9, this implies ϕ − (X i ) ≤ ϕ(B) ≤ ϕ + (Y j ) for all i, j. If there exist i, j such that ϕ − (X i ) − ϕ + (Y j ) ∈ Z, there exists an integer k ≥ 0 such that ϕ − (X i [k]) < ϕ + (Y j ) < ϕ − (X i [k]) + 1. Using Proposition 4.12 (1) this implies Hom(X i [k], Y j ) = 0. But, for , Y j ) = 0, which follows from Proposition 4.12 (2) because X i and Y j are not perfect. The conclusion is now that we must have X = 0 or Y = 0, which implies the claim. Proof. Suppose ϕ − (F ) < ϕ + (G). It is sufficient to derive a contradiction for indecomposable objects F and G. Because, for any k ≥ 0, F [k] ∈ D ≤0 , we may replace F by F [k] and can assume 0 < ϕ + (G) − ϕ − (F ) ≤ 1. Now, there exists a stable vector bundle B on E and an integer r such that Remark 5.7. In the case of a smooth elliptic curve Theorem 5.6 was proved in [19]. If θ ∈ Q the heart D(θ, θ + 1) of the corresponding t-structure is a finite-dimensional non-Noetherian Abelian category of infinite global dimension. In the smooth case, they correspond to the category of holomorphic vector bundles on a non-commutative torus in the sense of Polishchuk and Schwarz [32]. It is an interesting problem to find a similar interpretation of these Abelian categories in the case of a singular Weierstraß curve E. To complete this section we give two applications of Theorem 5.6. The first is a description of the group of exact auto-equivalences of the triangulated category D b coh (E). The second application is a description of Bridgeland's space of all stability structures on D b coh (E). In both cases, E is an irreducible curve of arithmetic genus one over k. Corollary 5.8. There exists an exact sequence of groups coh (E)) is generated by tensor products with line bundles of degree zero, automorphisms of the curve and the shift by 2. Proof. The homomorphism Aut(D b coh (E)) → SL(2, Z) is defined by describing the action of an auto-equivalence on K(E) in terms of the coordinate functions (deg, rk). That this is indeed in SL(2, Z) follows, for example, because Aut(D b coh (E)) preserves stability and the Euler-form for stable and perfect sheaves F , G. Clearly, tensor products with line bundles of degree zero, automorphisms of the curve and the shift by 2 are contained in the kernel of this homomorphism. In order to show that the kernel coincides with Aut 0 (D b coh (E)), we let G be an arbitrary exact auto-equivalence of D b coh (E). Then, G(Coh E ) is still Noetherian and it is the heart of the t-structure (G(D ≤0 ), G(D ≥0 )). From Theorem 5.6 and Lemma 5.3 we know all Noetherian hearts of t-structures. We obtain G(Coh E ) = D(θ, θ + 1] with P(θ) = {0}. Now, by Corollary 4.4 there exists Φ ∈ SL(2, Z) which maps D(θ, θ+1] to D(0, 1] = Coh E . This implies that the auto-equivalence Φ • G induces an auto-equivalence of the category Coh E . It is well-known that such an auto-equivalence has the form f * (L ⊗ · ), where f : E → E is an isomorphism and L is a line bundle. Note that f * (L ⊗ · ) is sent to the identity in SL(2, Z), if and only if L is of degree zero. The composition of Φ • G with the inverse of f * (L ⊗ · ) satisfies the assumptions of [7], Prop. A.3, hence is isomorphic to the identity. Because the kernel of the homomorphism SL(2, Z) → SL(2, Z), which is induced by the action of SL(2, Z) on D b coh (E) and the above homomorphism Aut(D b coh (E)) −→ SL(2, Z), is generated by the element of SL(2, Z) which acts as the shift by 2, the claim now follows. For our second application, we recall Bridgeland's definition of stability condition on a triangulated category [8]. Recall that we set K(E) = K 0 (Coh(E)) ∼ = K 0 (D b coh (E)). Following Bridgeland [8], we call a pair (W, R) a stability condition on D b coh (E), if W : K(E) → C is a group homomorphism and R is a compatible slicing of D b coh (E). A slicing R consists of a collection of full additive subcategories R(t) ⊂ D b coh (E), t ∈ R, such that (1) ∀t ∈ R R(t + 1) = R(t) [1]; (2) If t 1 > t 2 and A ν ∈ R(t ν ), then Hom(A 1 , A 2 ) = 0; (3) each non-zero object X ∈ D b coh (E) has a HNF Compatibility means for all non-zero A ∈ R(t) By ϕ R we denote the phase function on R-semi-stable objects. Similarly, we denote by ϕ R + (X) and ϕ R − (X) the largest, respectively smallest, phase of an R-HN factor of X. The standard stability condition, which was studied in the previous section, will always be denoted by (Z, P). This stability condition has a slicing which is locally finite, see [8], Def. 5.7. A slicing R is called locally finite, iff there exists η > 0 such that for any t ∈ R the quasi-Abelian category D R (t − η, t + η) is of finite length, i.e. Artinian and Noetherian. This category consists of those objects X ∈ D b coh (E) which satisfy t − η < ϕ R − (X) ≤ ϕ R + (X) < t + η. In order to obtain a good moduli space of stability conditions, Bridgeland [8] Proof. As seen above, there exists a not necessarily invertible matrix A such that W = A • Z. If A were not invertible, there would exist a number t 0 ∈ R such that W (K(E)) ⊂ R exp(iπt 0 ). This implies that there may exist a non-zero object in The assumption that the slicing R is locally finite implies now that R(t) is of finite length for any t ∈ R. On the other hand, the heart of the t-structure, which is defined by (W, R) is R(t 0 ) up to a shift. However, in Lemma 5.3 we determined all Noetherian hearts of t-structures on D b coh (E) and none of them is Artinian. This contradiction shows that A is invertible. It was shown in [8] that the group GL + (2, R) acts naturally on the moduli space of stability conditions Stab(E). This group is the universal cover of GL + (2, R) and has the following description: where compatibility means that f is strictly increasing, satisfies f (t + 1) = f (t) + 1 and induces the same map on S 1 ∼ = R/2Z as A does on S 1 ∼ = R 2 \{0}/R * . The action is simply (A, f )·(W, Q) = (A −1 •W, Q•f ). So, this action basically is a relabelling of the slices. The following result generalises [8], Thm. 9.1, to the singular case. Here, G ∈ SL(2, Z) is the image of G ∈ Aut(D b coh (E)) under the homomorphism of Corollary 5.8 and G(R)(t) := G(R(t)). Because automorphisms of E and twists by line bundles act trivially on Stab(E), we obtain Stab(E)/ Aut(D b coh (E)) ∼ = GL + (2, R)/ SL(2, Z), which is a C × -bundle over the coarse moduli space of elliptic curves. This result coincides with Bridgeland's result in the smooth case. The main reason for this coincidence seems to be the irreducibility of the curve. Example 5.14 below shows that the situation is significantly more difficult in the case of reducible degenerations of elliptic curves. Remark 5.13. Our results show that singular and smooth Weierstraß curves E share the following properties: (1) A coherent sheaf F is stable if and only if End(F ) ∼ = k. (2) Any spherical object is a shift of a stable vector bundle or of a structure sheaf k(x) of a smooth point x ∈ E. coh (E)) ∼ = GL + (2, R)/ SL(2, Z). Example 5.14. Let C 2 denote a reducible curve which has two components, both isomorphic to P 1 and which intersect transversally at two distinct points. This curve has arithmetic genus one and appears as a degeneration of a smooth elliptic curve. On this curve, there exists a line bundle L which fails to be stable with respect to some stability conditions. To construct an explicit example, denote by π : C 2 → C 2 the normalisation, so that C 2 is the disjoint union of two copies of P 1 . There is a k × -family of line bundles whose pull-back to C 2 is O P 1 on one component and O P 1 (2) on the other. The element in k × corresponds to a gluing parameter over one of the two singularities. Let L denote one such line bundle. If i ν : P 1 → C 2 , ν = 1, 2 denote the embeddings of the two components, we fix notation so that i * 1 L ∼ = O P 1 and i * 2 L ∼ = O P 1 (2). There is an exact sequence of coherent sheaves on C 2 (7) 0 → i 2 * O P 1 → L → i 1 * O P 1 → 0. Moreover, the only non-trivial quotients of L are L ։ i 1 * O P 1 and L ։ i 2 * O P 1 (2). Using the exact sequence (7), we obtain W a,b (L) = −2 + i(a + b). Using results of [33], it is easy to see that W a,b has the Harder-Narasimhan property in the sense of [8]. Hence, by [8], Prop. 5.3, W a,b defines a stability condition on D b coh (C 2 ). With respect to this stability condition, the line bundle L is stable precisely when 2/(a + b) < 1/a, which is equivalent to a < b. It is semi-stable, but not stable, if b = a. If a > b, L is not even semi-stable. This example illustrates an interesting effect, which was not available on an irreducible curve of arithmetic genus one. It is an interesting problem to describe the subset in Stab(E) for which a given line bundle L is semi-stable. This is a closed subset, see [8]. Physicists call the boundary of this set the line of marginal stability, see e.g. [1]. The example above describes the intersection of this set with a two-parameter family of stability conditions in Stab(E). Remark 5.15. In the case of an irreducible curve of arithmetic genus one, we have shown in Proposition 4.13 that Aut(D b coh (E)) acts transitively on the set of all spherical objects on E. Polishchuk [30] conjectured that this is likewise true in the case of reducible curves with trivial dualising sheaf. However, on the curve C 2 there exists a spherical complex which has non-zero cohomology in two different degrees, see [9]. This indicates that the reducible case is more difficult and involves new features.	2014-10-01T00:00:00.000Z	2005-03-23T00:00:00.000	{ "year": 2005, "sha1": "18dc67e93e432307e08894e32e5ad51912a64183", "oa_license": null, "oa_url": "https://www.cambridge.org/core/services/aop-cambridge-core/content/view/90973D9BA03382C84BD14A1C55CCF08D/S0010437X06002090a.pdf/div-class-title-derived-categories-of-irreducible-projective-curves-of-arithmetic-genus-one-div.pdf", "oa_status": "BRONZE", "pdf_src": "Arxiv", "pdf_hash": "18dc67e93e432307e08894e32e5ad51912a64183", "s2fieldsofstudy": [], "extfieldsofstudy": [ "Mathematics" ] }
250394025	pes2o/s2orc	v3-fos-license	Preservation of forensic traces by Nursing in emergency services: a scoping review Abstract Objective: to map the scientific production on the preservation of forensic traces by Nursing professionals working in emergency services. Method: a scoping review, with searches for studies carried out in six databases, in the gray literature available in Google Scholar and in the references of the studies selected. For analysis, the data reduction method was adopted. Results: 26 studies were included, organized into five categories: 1) Nursing professionals’ knowledge on the preservation of forensic traces; 2) Procedures performed by Nursing to preserve traces in the victim’s body; 3) Procedures performed by Nursing to preserve traces in the victim’s belongings/objects; 4) Procedures performed by Nursing to document traces; and 5) Actions to maintain the chain of custody performed by Nursing. Conclusion: the studies showed situations in which the emergency nurse may act in the preservation of forensic traces present in the victim’s body and in objects, as well as in the registration of traces, verifying the role of Nursing to ensure integrity of the chain of custody, especially in situations of aggression, firearm injury, sexual violence, child abuse and assistance to trauma victims. Introduction Emergency health services often act in the care of victims of crime situations; thus, they have in these environments a privileged opportunity to identify, collect and preserve forensic traces (1)(2) . These traces can include palm and plantar fingerprints; biological elements such as blood, semen, saliva, hair, bones, teeth, hair and vaginal secretions; and physicochemical traces such as chemicals, projectiles, melee weapons, firearms and sharp objects or instruments (3)(4)(5)(6) . Nursing professionals, who are on the front line in the care of patients in emergency services, in addition to having specific attributions to preserve life and reduce sequelae, must collaborate with the preservation of the traces present in the victim, in the possible aggressor, in the objects and at the crime scene (3) . Such traces, of high presence in Nursing care in emergency services, are essential elements for the success of the criminal investigation and for integrity of the chain of custody, as such chain consists of maintenance and documentation of traces, from their identification, collection, possession and handling until their disposal (1) . Collaboration of the Nursing professionals in forensic investigation can prevent the unnecessary loss or destruction of evidence; however, lack of knowledge in these professionals who work in the emergency services about proper preservation of traces exerts an impact on the work of the expert team (7) . Although the action towards victims of crimes occurs in the Nursing practice, most of the professionals do not have access to information on the theme (3) . Lack of training or moments of permanent education on the preservation of forensic traces, a content transversal to the Forensic Nursing specialty, results in the nonassociation of forensic care as inherent to the Nursing actions in emergency services (2) . In this context, knowing the scientific production about the preservation of forensic traces by Nursing professionals who work in emergency services is relevant, as it may enable nurses to access scientific information about the preservation of traces, given the growing reality of situations involving crime in emergency services. Thus, it is pointed out that this study will allow for the compilation and construction of new knowledge, which can be used in the training and qualification of Nursing professionals who work in the emergency services, in order to empower them with regard to the correct performance in situations in which forensic remains need to be preserved. In addition, although the review focus is on the Nursing team, it is noteworthy that this study is of potential interest to the multidisciplinary health team and managers whose professional performance permeates the context of emergency services. Thus, this study mapped the scientific production on preservation of forensic traces by Nursing professionals working in emergency services. Method Type of study This is a scoping review that followed the stages recommended by the Joanna Briggs Institute (JBI) (8) and the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist (9) . The review was developed in five stages: identification of the research question; survey of relevant studies; selection of the studies; data mapping; and presentation of the results (10) . Study setting This review was conducted in six databases: National Period The study was conducted between August and October 2021. Population The study population consisted of the 190 scientific articles found in the searches in the databases and in the gray literature available on Google Scholar. Selection criteria Articles with different types of research addressing the preservation of forensic traces by Nursing professionals in emergency services were included, without limitation regarding language or year of publication. For exclusion of the studies, we adopted the criteria of letters to the editor, abstracts of annals of events and not presenting information that contemplated the population, concept and context of interest of this study. Study variables The study variables were as follows: title of the article; year of publication; country; journal; language; objective; type of study; public studied; type of injury; location of the trace (crime scene, objects and Silva RX, Ferreira CAA, Sá GGM, Souto RQ, Barros LM, Galindo-Neto NM.. The results obtained in the databases were exported to the Rayyan reference manager, developed by the Qatar Computing Research Institute (QCRI) (12) , for removal of duplicates, selection and screening of THE studies by two researchers, independently and masked, and the divergences were resolved with the participation of a third examiner. After the search developed according to the strategy outlined above, the studies were selected. In addition, there was a search in the gray literature available on Google Scholar and a survey of potentially belongings, victim's body or others); and information on the preservation of forensic traces. After data extraction, the findings of both reviewers were compared, any and all discrepancies were resolved and the information was grouped into a single table. Instruments used to collect the information The diverse information extracted from the studies was recorded in a data collection instrument adapted from a form recommended by the JBI, organized in a Microsoft Excel 2001 spreadsheet (11) . Data collection In a previous search in the JBI database, no reviews were found to investigate the issue. Data treatment and analysis For analysis, the data reduction method was adopted, which aims at conceptually classifying the results after critical reading (13) . For the review report, the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) were followed (9) . Ethical aspects As the studies used were of public domain access, there was no need to submit the study to the Research Ethics Committee. Silva RX, Ferreira CAA, Sá GGM, Souto RQ, Barros LM, Galindo-Neto NM.. Rev. Latino-Am. Enfermagem 2022;30:e3540. Main results To address the role of Nursing within an emergency department and the final impact on forensic cases with the evidence preserved (14) . Emergency professionals are responsible for the evaluation, collection, documentation, and for properly following with the chain of custody. To present how evidence is collected in the emergency department in a situation of sexual assault (15) . It is necessary to evaluate, know the rudiments of collection, storage of physical evidence, document injuries by photographic means, avoid biological degradation and follow the chain of custody. To address obstacles faced for the implementation of the sexual assault response team (16) . The obstacles were funding, team composition and adequate training of professionals. To describe the functions of the forensic nurse in the emergency department (17) . Recognize a potential forensic situation, proceed with collection, preservation, complete documentation of objective and subjective signs, follow up in the chain of custody, notify closest relatives, see the link between patient, physician and investigator. To review the literature related to the recognition and collection of forensic materials in emergency departments by nurses (18) . Recognition and collection of evidence should be performed by nurses to forward to the authorities. These professionals must identify physical and nonphysical traces, document and maintain the chain of custody. To evaluate the clinical Nursing practice as a sexual assault examiner in an urgency care center for sexual attacks (19) . The practice consisted of physical evaluation, gathering evidence and documentation of assault and injuries, screening for Sexually Transmitted Infections and pregnancy, providing treatment and medication, and testifying in court. To evaluate the Nursing programs as sexual assault examiner (20) . Medical care was provided 24 hours a day for rape survivors. The minority of the victims received information to understand the medical services, such as pregnancy risk, contraception to prevent pregnancy and the risk of Sexually Transmitted Infections. To compare the collection of sexual assault nurse examiners (SANEs * ) in standardized evidence kits with other emergency departments without a SANE * (21) . The examiners trained in the SANE * program perform more complete collections and more compatible with the forensic evidence standards. To evaluate the education level of clinical practice trauma nurses in a trauma center (22) . Emergency room professionals received further instructions of forensic protocols on chain of custody maintenance, evidence gathering, and highquality documentation. To describe the perception of the importance of forensic role behaviors performed by nurses in emergency departments ((23) . In cases of child physical abuse, the forensic practice must be identified and performed. This is one of the main functions performed by forensic nurses. To assess the impact of introducing a sexual assault and domestic violence program in the emergency department with forensic evidence collection (24) . There was a significant increase in pelvic examinations, in the use of the forensic kit, and in the forensic chain. The rates of anogenital lesions, pregnancy and Sexually Transmitted Infections decreased. The flow in the emergency department was optimized. To develop an evidence-based set of guidelines for forensic evidence collection (25) . The guidelines address the use of equipment by the professionals; use of the evidence collection kit; general physical evaluation of the patient; adequate collection of fluids, belongings and materials; storage and proper documentation, use of photographs; collection of evidence left by the victim during evaluation. To describe the types of forensic evidence of firearm injuries and to describe the nurses' role in treating victims of gunshot wound (26) . The primary evidence on a gunshot victim is clothing, bullets, gunpowder, and primer. Nursing has the role of acting in the clinic, in communication and collaboration with the authorities. To describe and compare the forensic knowledge, practice and experiences of nurses and emergency physicians (27) . Physicians and nurses have the same level of knowledge, and they showed confidence in the collection and documentation of possible evidence. To compare the care quality indicators in a pediatric emergency department before and after the implementation of a program of pediatric sexual assault nurse examiners (28) . Implementation of the program maximized care, reduced the time of care, increased the collection of sexual violence kits, and the evaluation and documentation of pregnancy testing and of Sexually Transmitted Infections. To show the role of collecting and preserving evidence as a Nursing competence of the emergency department (29) . For collection and preservation, there should be informative evidence (stories, expressions, odors); tangible physical evidence (macroscopic and/or microscopic) through protocols and techniques, using forensic kits; to establish the chain of custody. To describe the nurses' views on the forensic care provided to victims of violence and their families in emergency departments (30) . Most of them knew about the care protocols, cooperated with the authorities and involved families in the victim's care process. To identify categories of forensic patients in the intensive care unit and emergency department (31) . The identification of 27 categories of forensic patients made it possible to properly recognize, document and collect evidence according to each lesion, before offering the necessary care for the patient's wounds. Nursing professionals' knowledge about the preservation of forensic traces • Safety of the professionals -Use Personal Protective Equipment. • Firearm -Remove the projectile from the patient's body. • Signs of child abuse -Identify physical and emotional abuse. -Collect for investigation purposes. Actions performed by Nursing to maintain the chain of custody (7,(16)(17)25) . The synthesis of the scientific production is presented in Figure 5. Objective Main results To evaluate professionals who worked in 112 emergency centers on how to act in forensic cases (6) . Most of the professionals were unable to recognize, preserve, protect and did not communicate adequately to the authorities. To describe the need for a collaborative relationship between the advanced practice of Forensic Nursing in the emergency department and intensive care environments (32) . Nurses are the link between the patient, the health system and the law. Their function is the recognition, collection, documentation, photodocumentation and completion of reports, so that the evidence is delivered to the authorities. To address incidents with active snipers in the United States in public places, including hospitals (33) . Spontaneous reaction, post-event actions and evidence collection procedures should be part of the continuous actions of emergency nurses. To define Nursing care in cases of trauma on the collection and preservation of forensic evidence in the emergency department (34) . Nurses should recognize the impact of trauma on the patients, early recognition of evidence, so that collection and preservation is part of the care plan, collaborate with other emergency professionals and guide the patients' privacy and rights. To investigate the knowledge and practice of emergency professionals on forensic processes during care for victims of violence (7) . Half of them were unaware of the collection, documentation and preservation procedures. There was no habit of collecting and documenting objects present in the victim's body, hospital materials and devices used during the procedures. To determine the levels of knowledge of emergency department health professionals in the treatment of frequently encountered forensic cases (35) . There was a low level of forensic training and the professionals, who are not prepared to carry out the practices, must make their own identification record for the authorities. To determine the Nursing professionals' knowledge level regarding the approach to cases and forensic tests (36) . The Nursing professionals did not have the knowledge and technical competence to perform forensic tasks; as they did not receive adequate training for this, they thought that the role would be the Police responsibility. To evaluate the relationship between nurses' knowledge and performance of forensic evidence procedures (37) . Most of the nurses knew the items for documentation and performed them in practice. Only 10.2% knew the procedures for preservation and evidence collection, but less than 1% practiced them. Rev. Latino-Am. Enfermagem 2022;30:e3540. Procedures performed by Nursing to preserve traces in the victim's body • Firearm -Wrap the shooter's hands in a paper bag for gunpowder collection and prime. -Take photographs of the wounds. -Do not puncture intravenous access in the hands of the possible aggressor. -Dry outdoors and store in paper bag the first dressing in gunshot wounds. • Violence in general -Collect blood before administration of crystalloids, medications or blood products. -Store the gastric contents in cans or empty bottles. -In cases of bites, collect the wound with moistened swabs and photograph the injury site. -Collect saliva with sterile and moist swab, on the tongue and cheek. -Collect and store mouthwash water. -Circle with a marker and take photographs of the wounds or injuries. -If nail scrapes occur, store aseptically. -Perform pelvic examination in a multidisciplinary approach with a physician. -Allow shower bath after collection and documentation of injuries. • Run overs -Store in paper bags the shards of glass on the victim or on the hospital gurney sheets. Procedures performed by Nursing to preserve traces in objects • Victim's clothes -Remove without cuts in the holes of firearm or melee arms. -Cut along the seams, and store in a paper bag. -Undress the patient with the use of paper sheets in order to find strands of hair and dirt. • Bed sheets -Allow to dry at room temperature. -Store hospital bed sheets in separate paper bags. • Shoes -Store each piece of shoe in a separate packaging. Procedures performed by Nursing to document traces -Record the patient's condition. Actions performed by Nursing to maintain the chain of custody • Care of the objects collected -Seal all containers with adhesive tapes, label with the collector's name, collection date and time. -Do not discard the projectiles or any item that constitutes evidence. -Give the guns to the police. • Care with recording the facts -Complete the chain of custody form, with all object transfer information. to permanent education (7) . In this sense, the procedures processes, which caused insecurity (18,(34)(35)(36)(37) . In Turkey, this scenario was also identified: 80% of the nurses who attended to forensic cases were able to differentiate types of evidence; however, they did not know how to collect, store and refer to the competent authorities (6) . In and recognition of traces in cases of trauma and sexual assault (7,18,(20)(21)(22)25,27,29,(34)(35)37) . It is inferred that Nursing has sufficient training to prioritize physiological, pharmacological and procedural aspects of such themes, which may even involve subjective issues such as humanization. However, they may be unaware of the judicial consequences of their assistance actions, so that they discard relevant evidence for criminal investigation. This fact was observed in a research study conducted in New Zealand, which investigated nurses in the emergency department and whose results showed limited knowledge Rev. Latino-Am. Enfermagem 2022;30:e3540. about criminal legislation, as well as that 84% reported considering the theme important for their professional practice (38) . These findings In addition to allowing understanding the processes in forensic cases, education for emergency nurses on how to preserve traces enables improvements in patient care. However, training on this theme for Nursing is scarce, including the gap of intersectoral articulation with public safety (7,18,(34)(35)(36) . In the United States, it is common to offer online training courses, which have already shown a 25% increase in the participants' knowledge (39) . Therefore and establishes the use of protocol and forensic evidence kits, necessary for the collection of traces (15,19,21,24,28) . The program also works with preventive measures for the victims, which include pregnancy tests, emergency gestational contraception and administration of prophylactic drugs for Sexually Transmitted Infections (STIs), as well as emergency contraceptives, clinical and psychological professional monitoring for 72 hours, in addition to storage of the evidence kit for up to six months, until the victim decides to use its content (40)(41) The studies of the sample pointed to the preservation of traces performed by emergency nurses, which occurred in shoes, bed sheets and other objects of the victim (17)(18)23,25,29,33,(35)(36) . These findings differ from those found in a study conducted in Brazil, whose results showed that, although the Nursing professionals recognized the need to preserve traces, it did not occur in such items, due to lack of routine and absence of documentation/ registration about the victim's objects and belongings (7,23) . Thus, it is perceived that to strengthen the practice in the Brazilian territory, it is necessary to devise and implement institutional protocols, in order to better guide the forensic practice by nurses working in emergency services. After the collection stage, documentation of the traces must be done in a thorough and attentive manner by Nursing because it is through this procedure that it will be possible to structure diverse information and prepare arguments to be analyzed in order to solve the crime. In this review, contents were identified that include actions ranging from the record about the patient's condition to the detailing of the record about the objects found (19,21,(24)(25)27,29,(32)(33)(34)(35) . Such actions are in line with a study carried out in Portugal, which highlights the relevance of detailing the documentation and record, which must be descriptive and accompanied by Silva RX, Ferreira CAA, Sá GGM, Souto RQ, Barros LM, Galindo-Neto NM.. a photographic record (43) . Despite being often associated with the bureaucratic and tiring routine, documentation/ recording stage, with a wealth of details not only contributes to justice occurring through the resolution of a crime, but also culminates in legal support for the professional Nursing practice and can be triangulated with the professional's report/testimony, if summoned to give testimony to the Police and/or judicial authorities. In the chain of custody, follow-up of the stages causes concern to nurses, as it does not consist only in storing the evidence in sealed and labeled containers, but also in the delivery of weapons and projectiles to the authors of the law and in the recording, by stamp and signature, of all the information therein contained (7,17,25) . A case report for the implementation of the SANE course in Brazil pointed out that the nurses' actions can contribute to the suitability of the chain of custody in the health services (44) . In the case of Saudi Arabia, nurses' concern with legal responsibilities, in the face of forensic cases, proved to be a barrier to maintaining the chain of custody (45) . Thus, it is noteworthy that clarification about the stages that make up the chain of custody, as well as the importance of the nurses' role for the success of involving firearms, sexual violence, child abuse, and in the assistance provided to trauma victims.	2022-07-10T15:05:16.761Z	2022-07-08T00:00:00.000	{ "year": 2022, "sha1": "49f9156042b8da1f7d040a816aa84b7f186b636a", "oa_license": "CCBY", "oa_url": "https://www.scielo.br/j/rlae/a/r9k3xVpVBZ5X9XRDKBxFssR/?format=pdf&lang=en", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "4ec5dc85aedc336c0d2adb2217a42a064db89efb", "s2fieldsofstudy": [], "extfieldsofstudy": [] }
214456365	pes2o/s2orc	v3-fos-license	Redox-Responsive Polycondensate Neoepitope for Enhanced Personalized Cancer Vaccine A versatile and highly effective platform remains a major challenge in the development of personalized cancer vaccines. Here, we devised a redox-responsive polycondensate neoepitope (PNE) through a reversible polycondensation reaction of peptide neoantigens and adjuvants together with a tracelessly responsive linker-monomer. Peptide-based neoantigens with diverse sequences and structures could be copolymerized with molecular adjuvants to form PNEs of high loading capacity for vaccine delivery without adding any carriers. The redox-responsive PNEs with controlled molecular weights and sizes efficiently targeted and accumulated in draining lymph nodes and greatly promoted the antigen capture and cross-presentation by professional antigen presenting cells. Mice immunized with PNEs showed markedly enhanced antigen-specific T cell response and the protective immunity against the tumor cell challenge. Instruments Nuclear magnetic resonance (NMR) spectra were acquired on a Bruker AVANCE NEO 400 MHz spectrometer (Billerica, Massachusetts, USA). Electrospray ionisation mass spectra (ESI-MS) was acquired on a LTQ Orbitrap ELITE ETD (Thermo Fisher Scientific). Matrix-assisted laser desorption/ionization time-of-flight mass spectra (MALDI-TOF-MS) was aquired on an Autoflex Speed (Bruker, Billerica, Massachusetts, USA). Polycondensate neoepitope (PNE) was characterized by PL-gel permeation chromatography (GPC) 50+ Integrated GPC/SEC System (Agilent, Santa Clara, California, USA) and UltiMate TM 3000 ultra high-performance liquid chromatography (UHPLC) system (Thermo Fisher Scientific) equipped with a Hypersil Gold TM C18 selectivity LC column or a BioBasic™ SEC 300 LC column, and detectors of diode array detector (UV, DIONEX UltiMate TM 3000) and charged aerosol detector (CAD, DIONEX Corona ultra RS). The fluorescent image of sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) were taken by Bio-Rad ChemiDoc MP imaging system (Hercules, California, USA). The size of particle was measured by dynamic light scattering (DLS) on Malvern NanoZS (Worcester, UK). The transmission electron microscopy (TEM) images of PNEs were acquired on FEI Tecnai Osiris TEM instrument (FEI, Oregon, USA). The atomic force microscopy (AFM) images of free LEQ and Pam were acquired on a Park NX-10 analytical system (Suwon, South Korea) with non-contact amplitude modulation (PPP-NCHR, Park system) in ambient condition. The fluorescence intensity of samples was measured with a Varioskan® Lux microplate reader (Thermo Fisher Scientific). All the flow cytometry data were acquired using an Attune NxT flow cytometer (Thermo Fisher Scientific). Confocal fluorescent microscope images were acquired with a LSM 700 with 40X or 63X oil objectives (Zeiss, Oberkochen, Germany). Mouse tissue imaging was conducted with an in vivo imaging system (IVIS, PerkinElmer, Waltham, Massachusetts, USA). Preparation of polycondensate neoepitopes (PNEs) Depending on the solubility of peptides and adjuvants, synthesis of PNEs can be done in anhydrous DMSO or aqueous solution. The detailed information of composition and conditions for the syntheses of each PNE has been summarized in Table S1. rpm. The resultant product was purified following the same procedure as described above, and characterized with DLS, GPC, HPLC, TEM, and SDS-PAGE. Gatan Orius CCD camera. Release kinetics and characterization of LEQ from PNE(LEQ-Pam) Freshly Fluorescent labelling of LEQ, Pam and PNE(LEQ-Pam) Solutions of LEQ, Pam, or the mixture of two were added with Alexa Fluor TM 647 NHS ester (10 mgmL -1 in anhydrous DMSO) with equal stoichiometry and then shaken with an Eppendorf ThermoMixer at 25 °C (600 rpm, 10 min). The "labeled" mixture was used for the next step without purification. Additional LEQ, Pam, or both (unlabeled) were then added the labeled mixture followed by the similar procedures for preparation and purification of PNEs as described above. For in vitro studies, 10% of LEQ or Pam was fluorescently labeled; for in vivo studies, 50% was labeled. In vitro BMDC internalization Immature BMDCs were prepared as described above. On Day 6, the BMDCs were plated in 24-well plates (5×10 5 cells per well) with RPMI 1640 medium (1 mL In vitro BMDC activation Immature BMDCs were prepared as described above. On Day 6, immature BMDCs were plated in 24-well plates (5 × 10 5 cells per well) with RPMI 1640 complete medium (1 mL Tumor challenging: On day 37 (9 days post the final immunization), the immunized mice (as described above) were challenged subcutaneously with YUMM1.7-OVA cells (5 × 10 5 ) on the right flank. Tumor growth was monitored every other day from the fifth day post tumor inoculation. Tumor area (product of two measured orthogonal diameters) and body weight were measured every 2 days. Mice were euthanized when the body weight loss was > 15% of the animal had become moribund. In vivo therapeutic study Female C57BL/6 mice were subcutaneously inoculated with YUMM1.7-OVA cells (5 × 10 5 ) on the right flank. The vaccine treatment containing LEQ (15 nmol) and Pam (5 nmol) in a PBS solution (50 μL) in forms of a simple mixture, an emulsion formulation in Montainide, or PNE(LEQ-Pam) were given on day 5, 12 and 19. Tumor growth were monitored every other day following a smilar procedure as described above. Statistical analysis. Statistical analysis was performed using GraphPad Prism 8 (GraphPad software, Inc., La Jolla, CA, USA). Unless otherwise noted, the data are presented as Mean ± SEM. Comparisons of two groups were performed by using two-tailed unpaired Student's t test. Comparisons of multiple groups at a single time point were performed by using one-way analysis of variance (ANOVA). Comparisons of survival curves were performed by using Log-rank analysis. Pvalues were presented as P < 0.05; * P < 0.01; * P < 0.001; ** P < 0.0001.	2020-02-06T09:04:42.271Z	2020-02-03T00:00:00.000	{ "year": 2020, "sha1": "305f15933bd44d23db65c61a44510178b2f4fd75", "oa_license": "acs-specific: authorchoice/editors choice usage agreement", "oa_url": "https://pubs.acs.org/doi/pdf/10.1021/acscentsci.9b01174", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "782164dd24c3f979e000a1c12a00de1d89f83a8c", "s2fieldsofstudy": [ "Medicine", "Chemistry" ], "extfieldsofstudy": [ "Chemistry", "Medicine" ] }
246308084	pes2o/s2orc	v3-fos-license	Slip versus Slop: A Head-to-Head Comparison of UV-Protective Clothing to Sunscreen Simple Summary Photoprotection reduces invasive melanoma incidence and mortality, but not all sun protection modalities are created equal. Dermatologists have long debated the pros and cons of photoprotective clothing and sunscreen, but few studies compare the effectiveness of these two modalities head-to-head. This study uses both in vitro and in vivo techniques to compare the ultraviolet radiation (UVR) protective capacity of four modern textiles and two commercially available, broad-spectrum sunscreens. Abstract Ultraviolet radiation (UVR) exposure is the most important modifiable risk factor for skin cancer development. Although sunscreen and sun-protective clothing are essential tools to minimize UVR exposure, few studies have compared the two modalities head-to-head. This study evaluates the UV-protective capacity of four modern, sun-protective textiles and two broad-spectrum, organic sunscreens (SPF 30 and 50). Sun Protection Factor (SPF), Ultraviolet Protection Factor (UPF), Critical Wavelength (CW), and % UVA- and % UVB-blocking were measured for each fabric. UPF, CW, % UVA- and % UVB-blocking were measured for each sunscreen at 2 mg/cm2 (recommended areal density) and 1 mg/cm2 (simulating real-world consumer application). The four textiles provided superior UVR protection when compared to the two sunscreens tested. All fabrics blocked erythemogenic UVR better than the sunscreens, as measured by SPF, UPF, and % UVB-blocking. Each fabric was superior to the sunscreens in blocking full-spectrum UVR, as measured by CW and % UVA-blocking. Our data demonstrate the limitations of sunscreen and UV-protective clothing labeling and suggest the combination of SPF or UPF with % UVA-blocking may provide more suitable measures for broad-spectrum protection. While sunscreen remains an important photoprotective modality (especially for sites where clothing is impractical), these data suggest that clothing should be considered the cornerstone of UV protection. UVB and UVA Cause Skin Cancer via Cyclobutane Pyrimidine Dimers (CPDs) Exposure to ultraviolet radiation (UVR) remains the most important and most modifiable risk factor for the development of skin cancer [1,2]. As skin cancer incidence rises worldwide, minimization of exposure to UVR is critical for reducing the morbidity, mortality, and cost associated with skin cancer [3]. To achieve this, the American Academy of Dermatology recommends sun avoidance, application of broad-spectrum sunscreen (SPF 30 or higher), and use of hats and protective clothing [4]. Since avoidance of UVR is not always possible, sunscreen and sun-protective clothing are essential features of a multi-pronged approach to skin cancer prevention. Recent data from Queensland, Australia underscore the fact that photoprotection is effective. Aitken et al. showed a decline in invasive melanoma incidence and mortality in individuals under age 40 years in Queensland [5]. The steepest decline was in those born after 1980, the decade in which photoprotection education initiatives such as the "Slip" (on a shirt), "Slop" (on sunscreen), and "Slap" (on a hat) campaign were implemented [5,6]. Both UVB (280-315 nm) and UVA (315-400 nm) radiation are carcinogenic. Of note, UVC (100-280 nm) is the most damaging type of UVR but is completely attenuated by ozone in the atmosphere before reaching the earth's surface. The UV portion of solar radiation that reaches the earth's surface comprises 5% UVB and 95% UVA radiation. UVB is a higher energy radiation but is less able to penetrate through the skin than the longer wavelengths of UVA. UVB causes erythema of skin and has historically been implicated as the main contributor to the development of skin cancer [7]. UVB is directly absorbed by DNA bases resulting in the formation of cyclobutane pyrimidine dimers (CPDs) and 6-4 photoproducts that can result in mutations if not repaired [7]. UVA causes pigment darkening or tanning of the skin and is associated with photoaging [8]. Recent work shows that UVA has a much greater role than previously understood in skin cancer development due to its ability to induce reactive oxygen and nitrogen species, and ultimately generate CPDs hours after UVA exposure [9][10][11][12]. UV-induced CPDs are more mutagenic than 6-4 photoproducts, likely due to their slower repair [13]. The most common UV-induced mutation in human skin cancer is the C>T transition. This mutation can be caused by the rapid deamination of cytosine (C) or 5-methylcytosine in CPDs, transforming them into uracil and thymidine, respectively. Error-free replication of the deaminated CPDs by DNA polymerase η then passes on C>T mutations to daughter cells [14]. The CPDs thereby cause skin cancer [15]. Sunscreens and Clothing Protect against UVR-Induced Mutagenesis Active ingredients of sunscreens fall into two categories: inorganic (also known as mineral or physical) and organic (also referred to as chemical). Inorganic ingredients include zinc oxide (ZnO) and titanium dioxide (TiO 2 ). New data suggest that these inorganic compounds primarily absorb UV radiation within the UVB and short UVA wavelengths and reflect radiation in the long UVA and visible wavelengths [16]. Organic UVR filters contain aromatic hydrocarbons that absorb photons in the UV spectrum and emit lower-energy, longer wavelength photons and/or heat that do not damage the skin [17]. Both inorganic and organic sunscreens prevent actinic keratoses (premalignant keratinocytic neoplasms) and squamous cell carcinoma [18][19][20]. Sunscreen is also effective in prevention of basal cell carcinoma and melanoma [18,21,22]. Furthermore, routine sunscreen application prevents photoaging [17]. Clothing provides protection by scattering and absorbing UVR. Mouse models have shown that sun-protective clothing can prevent skin cancer, but little human data exist in the literature [23]. However, since UVR is a causal agent in skin cancer, protection with clothing is likely to be dependent on the degree of protection provided from UVR. The degree of protection depends on the color, material, fiber, yarn and fabric structure [24]. Fabric structure is one of the most important factors, with the least porous material providing the greatest protection [25][26][27]. Synthetic fabrics have demonstrated the highest UV protection [28][29][30]. Dark colors absorb more UVR and thus provide higher protection than light colors [29,[31][32][33]. Other factors that impact penetration of UVR through the clothing include: stretch, wetness, wear (from use or washing), color loss (bleaching), UVR-absorbing additives, and yarn morphology [34][35][36][37]. Although all clothing blocks some degree of UVR, some studies suggest many commonly worn fabrics may provide insufficient UVR protection [29,38,39]. Measures of UVR Protection The sun protection factor (SPF) rating traditionally used in sunscreen labeling is measured by exposing small areas of skin of human subjects (Fitzpatrick skin types I-III) to simulated solar radiation for varying durations. The smallest dose of UVR that produces visible, well-circumscribed redness on tested skin is called the minimal erythemal dose (MED). SPF is the ratio of the MED with sunscreen uniformly applied at 2 mg/cm 2 (MED protected ) to that of skin without sunscreen (MED unprotected ): Simply stated, SPF reflects the factor by which one can spend more time in the sun without getting burned. For example, SPF 30 would theoretically allow someone who would normally burn in 10 min to be exposed for 300 min before burning, assuming constant incident solar UVR energy. The ultraviolet protection factor (UPF) rating traditionally used for clothing labeling requires a laboratory spectrophotometer to measure UVR transmittance (the fraction of UVR that is transmitted through the clothing). Although related, SPF and UPF are not equivalent [40]. In contrast to SPF measurements (that utilize UVB-induced erythema as the readout), UPF objectively measures all wavelengths of solar simulated light transmitted through the clothing and then applies weighting constants to mathematically recapitulate SPF testing. These constants (known as the erythemal effectiveness function and solar spectral irradiance) heavily weight UPF toward the UVB wavelengths ( Figure 1A) because these are the primary wavelengths that induce erythema in the skin. The UPF of a garment represents how much erythemally-weighted UVR is transmitted through the clothing. UPF values are the ratio of erythemally-weighted UVR detected with or without a specimen (clothing) between the light source and detector: E λ is the relative erythemal spectral effectiveness, a constant that adjusts for the ability of each wavelength (λ) to generate cutaneous erythema [41]. S λ is the solar spectral irradiance (Wm −2 nm −1 ), a constant that represents sun intensity at each wavelength at noon in Albuquerque, New Mexico [42]. T λ is the average measured spectral transmittance of the specimen and ∆λ is the measured wavelength interval (nm). Both of these measures represent the combined effect of incident dose and a biological response. This biological response, slight reddening, is related to discomfort but it is blistering sunburn that is related to skin cancer [43,44]. Neither SPF or UPF rating systems take into consideration the fact that non-erythema-inducing UVR wavelengths are carcinogenic. Nor do they take into consideration the fact that erythema and skin cancer have different dose-dependencies. One measure currently used to address this deficit in evaluating the performance of sunscreen is the critical wavelength. The critical wavelength (CW, λ c ) is calculated from the measured absorbance of a sunscreen across the entire UV spectrum and is intended to provide an objective quantifi- cation of how well a sunscreen reduces exposure to both UVB and UVA wavelengths [45]. Absorbance (A), is the negative of the (base ten) logarithm of transmittance (T): For example, at a given wavelength, a sunscreen or garment that blocks 90% of incident UV radiation has an absorbance of 1 [−log (0.1)], while one that blocks 99% of incident UV has an absorbance of 2 [−log (0.01)]. The transmittance of a sunscreen or textile is measured with a spectrophotometer equipped with an integrating sphere, which allows capture of all radiation that is diffusely scattered and transmitted through the sample. It is important to note that absorbance, as used in the quantitative expressions described here, represents more than just absorbed light; it also includes alternate means of radiation attenuation including the reflection and/or scattering of incident light. The critical wavelength is the wavelength below which 90% of the total absorbance (A) of a sunscreen in the atmosphere-penetrating UVR region is contained ( Figure 1B): In other words, the absorbance of a particular sunscreen is measured as a function of wavelength between 290-400 nm (UVB from 290-315 and UVA from 315-400). UVBspecific sunscreen ingredients are not effective at preventing transmission in the UVA range, producing absorbance curves that peak at the shorter end of the wavelength spectrum. In contrast, UVA filters shift the absorbance curves to the right at the longer UVA wavelengths. Sunscreens that block both UVA and UVB radiation have absorbance curves that extend across the majority of the spectral region ( Figure 1B). The area under the curve is summed (integrated) from 290 nm to 400 nm. The critical wavelength is the wavelength at which 90% of the total area under the absorbance curve is reached. Therefore, a higher critical wavelength indicates that the absorbance of a material is proportionally greater at longer wavelengths and offers more relative protection from UVA radiation. However, a UVAspecific agent with little absorbance in the UVB range could have a very high critical wavelength yet inadequate protection against UVB or sunburn. Therefore, the critical wavelength should never be interpreted in isolation and is only meaningful when SPF is also taken into consideration. The United States Food and Drug Administration (FDA) requires sunscreen to have a CW greater than 370 nm to be labeled as broad-spectrum [46]. % UV-Blocking. The critical-wavelength requirement does not exist for garments. Instead, some countries require sun-protective garments to have less than 5% transmittance in the UVA region (T(UVA), Equation (5)) [47]. These values are often reported as % UVA blocking = 100% − T(UVA), where T(UVA) is expressed as a percentage. Similarly, transmittance in the UVB region, T(UVB), is determined using an analogous equation but evaluating wavelengths from 280 to 315 nm, and % UVB blocking = 100% − T(UVB) where T(UVB) is expressed as a percentage. In this paper, we compare the SPF, UPF, CW, and % UV-Blocking of commercial sunscreens to those of modern sun-protective clothing. We follow this with a discussion of the consequences of exposure of human skin to UVA radiation and summarize new data on the mutagenic properties of visible light. We highlight the potential of sun-protective clothing to offer superior protection from the underappreciated risks posed by these wavelengths of solar radiation. UPF is a mathematical function designed to recapitulate sun protection factor (SPF) from a laboratory measurement of transmittance and is weighted toward the UVB portion of the spectrum. Left: Plots of the erythemal effectiveness function (Eλ) and the solar spectral irradiance (Sλ) over the UVR spectral range. Right: The product of Eλ and Sλ has a peak that lies predominantly (75%) within the UVB range (280-315 nm). Very little of the UPF function comes from wavelengths longer than 360 nm, where the UVR intensity is highest but the erythemal effectiveness is near zero [48]. (B) Visual representation of the CW as the wavelength below which 90% of the total absorbance (area under the curve) in the UVR region is contained. Left: Hypothetical sunscreen that primarily blocks UVB radiation has a critical wavelength below the 370 nm threshold required by the FDA to be labeled broad-spectrum. Right: Hypothetical sunscreen with improved UVA blocking performance meets the broad-spectrum criterion. A sunscreen could meet the CW criterion of 370 nm without comprehensively blocking UVA radiation. Fabrics Four fabric samples (Table 1, Figure 2) used in commercial sun-protective apparel (Columbia Sportswear, Portland, OR, USA) were chosen for the study because they are representative of modern sun-protective materials: one nylon woven and three polyester Figure 1. Ultraviolet protection factor (UPF) is primarily a measure of UVB protection whereas critical wavelength (CW) is a measure of the degree of broad-spectrum protection. (A) UPF is a mathematical function designed to recapitulate sun protection factor (SPF) from a laboratory measurement of transmittance and is weighted toward the UVB portion of the spectrum. Left: Plots of the erythemal effectiveness function (E λ ) and the solar spectral irradiance (S λ ) over the UVR spectral range. Right: The product of E λ and S λ has a peak that lies predominantly (75%) within the UVB range (280-315 nm). Very little of the UPF function comes from wavelengths longer than 360 nm, where the UVR intensity is highest but the erythemal effectiveness is near zero [48]. (B) Visual representation of the CW as the wavelength below which 90% of the total absorbance (area under the curve) in the UVR region is contained. Left: Hypothetical sunscreen that primarily blocks UVB radiation has a critical wavelength below the 370 nm threshold required by the FDA to be labeled broad-spectrum. Right: Hypothetical sunscreen with improved UVA blocking performance meets the broad-spectrum criterion. A sunscreen could meet the CW criterion of 370 nm without comprehensively blocking UVA radiation. Fabrics Four fabric samples (Table 1, Figure 2) used in commercial sun-protective apparel (Columbia Sportswear, Portland, OR, USA) were chosen for the study because they are representative of modern sun-protective materials: one nylon woven and three polyester knit fabrics. The three knit fabrics include two different common knit structures, a pique and an interlock. Of the two interlock knit fabrics, one includes a TiO 2 dot print covering about 30% of its surface. These white dots are present for heat mitigation, and work by reflecting more solar radiation and emitting more thermal radiation than the underlying polyester fabric [49]. These four fabrics are generally thinner and lighter than the ones for which UPF and SPF data have been previously reported in the literature [24]. All fabrics were dyed to be off-white. Sunscreens Two organic, broad-spectrum sunscreens (SPF 30 and SPF 50, Table 1) were selected for comparison with the four fabrics. The sunscreens were the same brand and contained identical active ingredients: avobenzone, homosalate, octisalate, and octocrylene. The SPF 50 sunscreen had a higher percentage of homosalate (10% vs. 8%) and octocrylene (8% vs. Sunscreens Two organic, broad-spectrum sunscreens (SPF 30 and SPF 50, Table 1) were selected for comparison with the four fabrics. The sunscreens were the same brand and contained identical active ingredients: avobenzone, homosalate, octisalate, and octocrylene. The SPF 50 sunscreen had a higher percentage of homosalate (10% vs. 8%) and octocrylene (8% vs. 6%). These active ingredients are four of the sixteen UV filters listed in the FDA Code of Federal Regulations [50] and are a common combination used in multiple brands of U.S. sunscreens. Tests were performed with newly opened, unexpired products. 2.3. In Vitro UPF Testing 2.3.1. Fabric UPF UPF measurements of the textile fabrics were conducted as described in the American Association of Textile Chemists and Colorists (AATCC) Test Method 183 [48]. Briefly, each dry, unstretched fabric was placed in a UV-2000S Ultraviolet Transmittance Analyzer (Labsphere Inc., North Sutton, NH, USA) to capture diffuse transmittance from 280 to 400 nm. Five unique measurements were taken at different sample orientations rotated by 45 • between each measurement. UPF was calculated for the transmission spectrum of each sample using the formula in Equation (2), and then an average UPF and standard deviation were calculated from the individual UPF values [48]. Sunscreen UPF UPF measurements of two commercial broad-spectrum sunscreens were conducted using AATCC Test Method 183 [48]. Of note, many studies refer to spectrophotometric testing of sunscreens as "in vitro SPF" and to erythema measurements of textiles on human skin as "in vivo UPF." To eliminate confusion, we will use the terms "sunscreen UPF" and "fabric SPF" throughout. The two sunscreens were applied in accordance with International Organization for Standardization (ISO) 2444 to UVR-transparent quartz slides at a density of 2 mg/cm 2 . Additional testing was performed at a density of 1 mg/cm 2 to simulate real-world consumer application [51]. Transmittance was measured using a Thermo Fisher Evolution Bio260 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). To account for non-uniformity in coating thickness, three slides were prepared for each sunscreen SPF level and coating density. Ten measurements were taken in random positions on each slide. UPF was calculated for each measurement, and the UPF is reported as the mean and standard deviation of these measurements. SPF of Fabrics The fabric SPF on human skin of the four textile samples were measured by a third party (AMA Laboratories, New City, NY, USA) using the standard procedure described by the International SPF Test Method of COLlPA, CTFA, JCIA and CTFA-SA [52]. The study was approved by the AMA Laboratories Internal Review Board. Briefly, the dose of UVR was supplied by either a 150 or 300 watt Xenon Arc Solar Simulator (Solar Light Co., Philadelphia, PA, USA), each with a continuous emission spectrum from 290 to 400 nm. Both were equipped with dichroic mirrors and a 1-mm Schott WG-320 filter to produce a simulated solar UVA-UVB emission spectrum. These were also equipped with a 1-mm UG 11 filter to remove reflected heat as well as visible and infrared radiation. A total of five healthy adult volunteers (ages 36 to 64 years) were selected by AMA laboratories. Three volunteers (one from each Fitzpatrick Skin Type I, II, and III to maximize the human variation studied) underwent SPF testing for each fabric [53] (Tables S1 and S2). All individuals provided written informed consent. The bilateral infrascapular area of each subject's back was used as the test site. Test sites were cleaned with a dry cotton pad, and rectangular areas of at least 30 cm 2 were demarcated. A minimum of five progressive UVR doses were administered within this site to determine a subject's minimal erythemal dose on unprotected skin (MED unprotected ). Each subject's MED unprotected was defined as the shortest time of exposure (or lowest UVR dose required) that produced minimally perceptible erythema at 16 to 24 h post exposure. A control sunscreen with standard SPF 15 or 16 was used as a verification technical control. Once the MED unprotected had been determined, subjects returned to the lab for SPF testing of the fabrics. The test fabric was secured closely to each subject's skin without stretching and using a thin layer of adhesive tape on the sample periphery to cover a minimum area of 30 cm 2 . Based upon each subject's previously determine MED unprotected , the test areas were irradiated with a series of progressively higher UVR doses (minimum of five). The subjects returned to the testing facility 16-24 h after UVR exposure for determination of the MED of protected skin (MED protected ) by a blinded evaluator. Each fabric was tested on three subjects and the mean SPF was determined for each individual fabric as the ratio of MED protected/ MED unprotected (Equation (1)). SPF of Sunscreens For the purposes of this study, the package label SPFs of the commercial sunscreens were used. SPFs of commercial sunscreens in the U.S. are determined in accordance with FDA regulations as described in the Code of Federal Regulations 21 [46]. In Vivo SPF Values Are Higher for Fabrics Than Sunscreens In vivo SPF measurements of the fabrics ranged from 60 to 80, with the nylon woven fabric having the lowest SPF and the polyester interlock knit having the highest SPF ( Table 2). All fabrics had a higher SPF than the SPFs on the sunscreen package labels. In Vitro UPF Values Are Higher for Fabrics Than Sunscreens In vitro UPF measurements for all four fabrics exceeded 200 (Table 2). UPF values of sunscreens demonstrated dose-and SPF-dependence, as expected. Sunscreen B (SPF 50) applied at an areal density of 2 mg/cm 2 showed the greatest UPF (31 ± 19), while Sunscreen A (SPF 30) applied at 1 mg/cm 2 had the lowest UPF (5.3 ± 3.2). All Fabrics Have CWs That Meet the Criteria for Broad-Spectrum Labeling All four fabrics had critical wavelengths (CWs) greater than or equal to 370 nm, the minimum value that the FDA requires for a sunscreen to be considered broad-spectrum (Table 2). Both sunscreens exhibited a broad-spectrum CW of at least 370 nm at an areal density of 2 mg/cm 2 , but not at 1 mg/cm 2 . % UVB-Blocking Is Greater in Fabrics Relative to Sunscreens All four textiles blocked > 99% of UVB (which equates to a transmittance < 1% in the UVB region) with little difference detected between the four fabric types ( Table 2). Both sunscreens blocked UVB in a dose-and SPF-dependent manner (range of sunscreen protection = 76-94%), with neither providing as much protection as the fabrics. The fabrics exhibited consistently higher and less variable UVB blocking compared to the two commercial sunscreens at both areal densities. The UVR spectra of the textiles and sunscreens (Figure 3) show that all textiles had lower transmittance than the sunscreens over the UVB wavelengths. % UVA-Blocking Is Greater in Fabrics Relative to Sunscreens UVA blocking by the fabrics ranged from 96% to 98%, with the polyester fabrics having slightly better performance than the nylon fabric (transmittance < 4% in the UVA region, Table 2). As with UVB, sunscreen UVA blocking increased with areal density and SPF, ranging from 54% for the SPF 30 formulation applied at 1 mg/cm 2 to 82% for the SPF 50 sunscreen applied at 2 mg/cm 2 . The best sunscreen UVA blocker, SPF 50 applied at 2 mg/cm 2 , did not block UVA as well as the lowest performing fabric (nylon, at 96%). All four textiles had much lower UVA transmittance than the sunscreens. The transmittance of all tested sunscreens increased rapidly in the UVA region from 380 to 400 nm ( Figure 3). Discussion The four textiles in our study provided superior UVR protection when compared to the two commercial sunscreens tested (Table 2, Figure 3). All four fabrics successfully blocked erythemogenic UVR at a level better than the sunscreens, as measured by UVB transmittance, SPF, and UPF metrics. All four fabrics were superior to the sunscreens with respect to blocking full spectrum UVR, as measured by UVA transmittance and CW metrics. There was substantial disparity between the SPF and UPF values determined for both fabric and sunscreen. Despite previous reports to the contrary [40], this is not surprising % UVA-Blocking Is Greater in Fabrics Relative to Sunscreens UVA blocking by the fabrics ranged from 96% to 98%, with the polyester fabrics having slightly better performance than the nylon fabric (transmittance < 4% in the UVA region, Table 2). As with UVB, sunscreen UVA blocking increased with areal density and SPF, ranging from 54% for the SPF 30 formulation applied at 1 mg/cm 2 to 82% for the SPF 50 sunscreen applied at 2 mg/cm 2 . The best sunscreen UVA blocker, SPF 50 applied at 2 mg/cm 2 , did not block UVA as well as the lowest performing fabric (nylon, at 96%). All four textiles had much lower UVA transmittance than the sunscreens. The transmittance of all tested sunscreens increased rapidly in the UVA region from 380 to 400 nm (Figure 3). Discussion The four textiles in our study provided superior UVR protection when compared to the two commercial sunscreens tested (Table 2, Figure 3). All four fabrics successfully blocked erythemogenic UVR at a level better than the sunscreens, as measured by UVB transmittance, SPF, and UPF metrics. All four fabrics were superior to the sunscreens with respect to blocking full spectrum UVR, as measured by UVA transmittance and CW metrics. There was substantial disparity between the SPF and UPF values determined for both fabric and sunscreen. Despite previous reports to the contrary [40], this is not surprising given the fact that they are measured in different ways, using transmittance of UVR to calculate UPF in contrast to induction of a downstream cutaneous reaction (erythema) to calculate SPF. UPF is indirectly calculated using a mathematic weighting to simulate the erythemogenic potential of the UVR, whereas SPF directly measures development of erythema in response to UVR. In addition to the absolute performance of fabrics and sunscreens, there was substantial variability of protection by sunscreens in our study, despite great efforts made to assure consistency ( Table 2). This variability is likely due to the difficulty in uniformly applying sunscreen and highlights an inherent weakness in sunscreens with respect to inconsistent application and the propensity of sunscreens to wear or wash off. Despite these challenges, sunscreens remain an important tool of our armamentarium against skin cancer when protective clothing is not an option. Tailored Use of Fabrics and Sunscreens In general, utilization of both photoprotective clothing and sunscreen offers improved UV protection. However, some activities and climates make wearing photoprotective clothing difficult or undesirable, and it is impractical to cover the entire body with clothing. Similarly, sunscreens can be messy and difficult to apply and re-apply in proper amounts, especially after vigorous exercise or water-related activities. Ideally, these two photoprotective methods can be combined and tailored for each person and for the outdoor activities in which they are engaged. However, there are additional factors that should be considered with respect to the use of clothing and sunscreen. Limitations in the Use of Sunscreen for Photoprotection We now know that radiation across the entire UVR spectrum damages DNA. A review of the published absorbance spectra of active organic and inorganic sunscreens shows that nearly all U.S. sunscreen ingredients provide little or limited protection at wavelengths longer than 380 nm [54][55][56][57][58][59] (Figure 4). The curves in Figure 4 reflect the transmission of sunscreen ingredients at the maximum concentration allowed by the FDA. When applied at high concentrations (25%), the inorganic compounds (ZnO and TiO 2 ) do provide superior protection across the UV spectrum compared to organic filters, but still show up to 20% transmittance beyond 380 nm. However, very few sunscreens contain inorganic filters at these high concentrations as they can be difficult to apply and have a chalky, cosmetically less acceptable appearance. Most inorganic sunscreens on the market contain 10-20% ZnO and 2-14% TiO2 [60]. It is also worth noting that eight broad-spectrum, organic UV filters commonly used in other countries but not yet approved by the FDA also show an increase in transmittance beyond 380 nm [61,62]. This gap in UVR protection is significant since long-wavelength UVA is capable of producing reactive oxygen species and CPDs that can lead to skin cancer [9][10][11]. Although the carcinogenic potential of UVB and UVC were first to be shown to produce mutagenic CPDs and 6-4 photoproducts in DNA, a robust body of literature now exists supporting the carcinogenic potential of UVA as well [9][10][11]63]. Recent studies also show that longer UVA wavelengths (UVA1, 340-400 nm) induce more solar UV-signature mutations than shorter UV wavelengths [64]. Lawrence and colleagues demonstrated that human skin irradiated at 385 nm generated CPDs, which increased for 2 h and persisted for 24 h without evidence of repair [65]. Additionally, irradiation of human skin with UVA1 and visible light produces biomarkers of DNA damage in the form of increased TP53 and BCL-2 expression, eliciting a DNA damage response without producing visible erythema [63]. Further, data from Runger et al. suggest that UVA-induced CPDs may be more mutagenic than those produced by UVB due to a lower and shorter-lived activation of protective cell cycle arrest pathways [66]. show up to 20% transmittance beyond 380 nm. However, very few sunscreens contain inorganic filters at these high concentrations as they can be difficult to apply and have a chalky, cosmetically less acceptable appearance. Most inorganic sunscreens on the market contain 10-20% ZnO and 2-14% TiO2 [60]. It is also worth noting that eight broad-spectrum, organic UV filters commonly used in other countries but not yet approved by the FDA also show an increase in transmittance beyond 380 nm [61,62]. . Transmittance of organic sunscreen filters (tested in this study) and inorganic sunscreen filters (ZnO and TiO2). Data were obtained from the BASF sunscreen simulator [59] where the maximum concentration allowable by the FDA [46] was used to generate each curve. Figure 4. Transmittance of organic sunscreen filters (tested in this study) and inorganic sunscreen filters (ZnO and TiO 2 ). Data were obtained from the BASF sunscreen simulator [59] where the maximum concentration allowable by the FDA [46] was used to generate each curve. The high absorbance of textiles in the near UVA region raises the question of whether there are biological effects of visible light against which textiles might also be protective. UVR and visible light have been shown to generate reactive oxygen species (ROS) that damage sensitive biomolecules in the skin. ROS produced by solar radiation in the UVB, UVA, and visible wavelengths cause damage to DNA that is potentially mutagenic. Photons in the UV region are the most efficient at producing ROS, but studies of the action spectrum of sunlight indicate that more than 50% of free radicals, including ROS, arise from visible light with wavelengths in the range of 400 to 700 nm [67]. The free radicals produced by visible light have the same carcinogenic and ageing effects as their counterparts produced by UV. The principal differences in the effects of various wavelengths of light on generation of ROS in the skin are the chromophores that mediate radical production. Light-absorbing chromophores in the skin include nucleic acids, aromatic amino acids, urocanic acid, NADH and NADPH, cytochromes, riboflavins, porphyrins, melanin and its precursors, and βcarotene [68]. These chromophores can act as photosensitizers that catalyze the generation of ROS and reactive nitrogen species that, if not quenched by cellular antioxidant defenses, can damage sensitive biomolecules such as DNA, RNA, proteins and lipids. Blue light (400-500 nm) is known to generate pigmentary changes in human skin, particularly visible in persons with darker skin (Fitzpatrick phototypes III and IV). Irradiation of the skin on the backs of healthy volunteers with 415 nm blue-violet light produced hyperpigmentation, independent of p53 activation, that persisted for months after exposure [69]. In subsequent mechanistic studies, the laboratory of Thierry Passeron showed evidence that this effect is mediated by a dedicated sensor, opsin-3 (OPN3) which is expressed in melanocytes [70]. Activation of melanogenesis downstream of OPN3 is calcium dependent, activates the protein kinase CAMKII, and leads to the phosphorylation of the transcription factor MITF and thus increased transcription of melanin synthesis enzymes tyrosinase and dopachrome tautomerase (DCT). Blue light also facilitates the formation of a protein complex that contains tyrosinase and DCT. This complex leads to sustained tyrosinase activity that likely mediates the long-lasting hyperpigmentation that is observed in skin type III and higher. Taken together, these data suggest that available sunscreens fail to protect against wavelengths of radiation, both UVB and UVA, that produce CPD and 6-4 photoproducts, which can result in mutations that drive carcinogenesis. In addition, there are hyperpigmentation effects of visible light, also not blocked by current sunscreens, that are of significant cosmetic concern to many individuals with darker skin. This makes protection from the full spectrum of UV and visible radiation an important goal, but one that may be difficult to achieve because of the strong cosmetic preference for transparent sunscreens. Sunscreens that protect in the 380-700 nm region are opaque to the human eye, so most sunscreens lack protection in the long UVA and visible spectrum. Importantly, photoprotective clothing may represent a partial solution to this problem since opaque clothing effectively blocks both UV and visible radiation. Limitations in the Use of Clothing for Photoprotection Few studies in the literature have directly compared the performance of UV-protective clothing to sunscreen. In 2018, Coyne et al. reported the broad-spectrum protection afforded by then-available commercial clothing [71]. Selected results from analysis of those data are shown in Figure 5 re-plotted as transmittance versus wavelength for comparison with the four fabrics that we tested. Because this figure only includes fabrics, the upper limit of the transmittance scale is 15%. In contrast, the transmittance scale in Figure 3, which includes both fabrics and sunscreens, must extend to 100% to display the increasing transmittance, and reduced protection, of the sunscreens beginning around 370 nm. sunscreens lack protection in the long UVA and visible spectrum. Importantly, photoprotective clothing may represent a partial solution to this problem since opaque clothing effectively blocks both UV and visible radiation. Limitations in the Use of Clothing for Photoprotection Few studies in the literature have directly compared the performance of UV-protective clothing to sunscreen. In 2018, Coyne et al. reported the broad-spectrum protection afforded by then-available commercial clothing [71]. Selected results from analysis of those data are shown in Figure 5 re-plotted as transmittance versus wavelength for comparison with the four fabrics that we tested. Because this figure only includes fabrics, the upper limit of the transmittance scale is 15%. In contrast, the transmittance scale in Figure 3, which includes both fabrics and sunscreens, must extend to 100% to display the increasing transmittance, and reduced protection, of the sunscreens beginning around 370 nm. The data from Coyne, et al. were digitized, converted to transmittance and represent the average of the 16 measurements as reported in the original study. White cotton and dark grey cotton were GAP, Inc. 100% cotton shirts. The denim was 69% cotton, 30% polyester, 1% spandex GAP jeans. Polyester refers to a 84% polyester, 16% spandex Coolibar rash guard [71]. The Coyne data plotted in Figure 5 are summarized in Table 3. The denim jeans, dark grey cotton shirt, and polyester/spandex photoprotective rash guard provided the most protection. With the exception of the white cotton shirt, all fabrics blocked > 99% of UVB, blocked > 96% of UVA, and met the minimum critical wavelength (370 nm) to be considered broad-spectrum protection. The fabrics in our study provide comparable broad-spectrum protection (Figure 3, Table 2) without requiring dark dyes. Since lighter colored fabrics (such as those tested in our study) generally provide the least protection for a given fabric structure, dyes and pigments added for deeper coloration would lend even more UVR protection than that indicated in Table 2. As shown in the Coyne data, a common white cotton shirt does not typically provide as much UVR protection (UPF = 9, Table 3) Figure 5. Transmittance spectra of the four fabrics designed for UV protection tested in this study (orange and purple lines) compared to "normal" clothing items tested by Coyne, et al. (gray lines). The data from Coyne, et al. were digitized, converted to transmittance and represent the average of the 16 measurements as reported in the original study. White cotton and dark grey cotton were GAP, Inc. 100% cotton shirts. The denim was 69% cotton, 30% polyester, 1% spandex GAP jeans. Polyester refers to a 84% polyester, 16% spandex Coolibar rash guard [71]. The Coyne data plotted in Figure 5 are summarized in Table 3. The denim jeans, dark grey cotton shirt, and polyester/spandex photoprotective rash guard provided the most protection. With the exception of the white cotton shirt, all fabrics blocked > 99% of UVB, blocked > 96% of UVA, and met the minimum critical wavelength (370 nm) to be considered broad-spectrum protection. The fabrics in our study provide comparable broad-spectrum protection ( Figure 3, Table 2) without requiring dark dyes. Since lighter colored fabrics (such as those tested in our study) generally provide the least protection for a given fabric structure, dyes and pigments added for deeper coloration would lend even more UVR protection than that indicated in Table 2. As shown in the Coyne data, a common white cotton shirt does not typically provide as much UVR protection (UPF = 9, Table 3) as a dark grey shirt made of the same material (UPF = 98, Table 3). On the other hand, the denim tested in the Coyne study provided even better UVR protection (UPF = 2000, Table 3). While fabric weights and thicknesses were not provided in the Coyne study, denim is typically heavier and darker than fabrics used for shirts. Thus, the Coyne data additionally demonstrate that sufficiently heavy and dark fabrics, such as denim, will effectively block UVR. The primary challenge for textile engineers is to provide outstanding UVR protection in garments using fabrics that are lightweight and air permeable (attributes that are linked to comfort in hot environments), thereby increasing the likelihood that a person would choose clothing for photoprotection. Limitations of Current Photoprotection Metrics Our data and that of Coyne, et al. also demonstrate that CW alone does not provide a useful characterization of UVR protection. The white cotton fabric exhibits a critical wavelength of 389 nm (Table 3), which technically meets the FDA definition of broadspectrum protection for a sunscreen. However, the same fabric has a UPF rating of 9 and the actual protection provided across wavelengths is the lowest of all of the fabrics in this comparison. The % UVA and UVB blocking values, 91.7% and 89.9%, respectively, are better indicators of the minimal level of broad-spectrum protection provided by this white cotton shirt. Additionally, the specific shape of the UV transmittance curve can generate misleading results for measurements of SPF, UPF and CW. In the case of materials that exhibit UV transmission that increases rapidly through the UVA region, it is possible for a sunscreen or a garment to provide poor UV protection despite having a seemingly adequate SPF, UPF and CW. We illustrate this concept in Figure 6 and Table 4. The solid curve represents the measured transmittance of Sunscreen B applied at 2 mg/cm 2 , which has a labeled SPF of 50 but a measured UPF of 31. The dashed curve is a hypothetical sunscreen with a similar transmittance shape profile to Sunscreen B, but modified to achieve a UPF of 50. One can conceptualize this shift as a person applying sunscreen at a higher areal density than the recommended 2 mg/cm 2 (which is much higher than that typically applied by consumers [51]). Although the hypothetical dashed curve now represents a UPF value of 50, and maintains a broad-spectrum CW of 374 nm, this "new" sunscreen still only blocks 87% of UVA, which is considered inadequate relative to the 95% UVA blocking criteria used by some countries for sun-protective clothing. As this example demonstrates, the combination of UPF and UVA blocking may provide more suitable measures for broadspectrum protection. Numerous jurisdictions, including the European Union, require this combination as an indication of broad-spectrum protection for garments [47]. Figure 6. Illustration of the limitations of current measures of sun protection. The solid curve shows the measured transmittance of Sunscreen B applied at 2 mg/cm 2 , which has a SPF rating of 50 but had a measured UPF of only 31. The dashed curve was generated using a shifting function to simulate a transmittance curve with an average UPF of 50. For both the measured and simulated curve, the rapid increase in transmittance in the UVA region represents reduced protection from UVA radiation. Limitations of this Study Our small sample size and use of only one brand of commercial organic sunscreen may limit generalizability, although most brands select from the same ingredient list. Our study did not directly test inorganic UV filters, but the transmission curves generated using the BASF sunscreen simulator [59] (Figure 4) show that ZnO and TiO2 demonstrate up to 20% transmittance beyond 380 nm. The high concentrations of ZnO and TiO2 required to achieve adequate UVA protection may result in a sunscreen that can be difficult to apply and have a chalky, cosmetically less acceptable appearance. The method we used to determine the UPF of sunscreens had limitations including variability in sunscreen application on the slide and the slide's differing properties and structure compared to human skin. Challenges posed by in vitro UPF testing of sunscreens have been demonstrated in prior studies, and to date, no in vitro UPF test for sunscreens has been approved due to difficulty in generating reproducible and reliable results from lab to lab [72,73]. Figure 6. Illustration of the limitations of current measures of sun protection. The solid curve shows the measured transmittance of Sunscreen B applied at 2 mg/cm 2 , which has a SPF rating of 50 but had a measured UPF of only 31. The dashed curve was generated using a shifting function to simulate a transmittance curve with an average UPF of 50. For both the measured and simulated curve, the rapid increase in transmittance in the UVA region represents reduced protection from UVA radiation. Limitations of This Study Our small sample size and use of only one brand of commercial organic sunscreen may limit generalizability, although most brands select from the same ingredient list. Our study did not directly test inorganic UV filters, but the transmission curves generated using the BASF sunscreen simulator [59] (Figure 4) show that ZnO and TiO 2 demonstrate up to 20% transmittance beyond 380 nm. The high concentrations of ZnO and TiO 2 required to achieve adequate UVA protection may result in a sunscreen that can be difficult to apply and have a chalky, cosmetically less acceptable appearance. The method we used to determine the UPF of sunscreens had limitations including variability in sunscreen application on the slide and the slide's differing properties and structure compared to human skin. Challenges posed by in vitro UPF testing of sunscreens have been demonstrated in prior studies, and to date, no in vitro UPF test for sunscreens has been approved due to difficulty in generating reproducible and reliable results from lab to lab [72,73]. The SPF and UPF measurements were made under controlled laboratory conditions. These do not account for the real-world photodegradation of sunscreen UV filters, or the changes in performance due to the known decrease in areal density of the sunscreens as a consequence of friction, sweating, or water exposure [74,75]. In contrast, real-world performance of textiles is likely superior in most situations. In our study, fabrics were taped directly to the skin instead of suspended above the skin. On-skin testing has been reported to drastically reduce SPF in fabrics due to the incident UVR passing directly through the open parts of the fabric structure [76,77]. Although the SPF of the fabrics tested may represent a "worst-case" scenario, all SPFs exceeded 60. Conclusions In our study, we demonstrate that the four tested fabrics provided superior UVB, UVA, and overall broad-spectrum protection when compared to two commercial sunscreens. These data, coupled with mounting concerns regarding sunscreen ingredients and excipients (including biological impacts [78], potential environmental harms [79], and difficulty to apply correctly [51]), underscore that clothing should be considered the cornerstone of UV protection. Nevertheless, sunscreen remains an important modality for UV protection, especially on areas of the body such as the face and hands were where clothing may be impractical. In the future, the most effective and widely adopted strategies will likely incorporate both photoprotective clothing and sunscreens with absorbances extending into the visible light range. Photoprotection modalities, and their regulatory labeling, are imperfect and must evolve with our understanding of the mutagenic potential of solar radiation beyond that of UVB: erythema is not the only biologically relevant endpoint and may not be the endpoint most closely associated with carcinogenesis. To provide comprehensive protection from skin cancer, photo-ageing, and hyperpigmentation, strategies that impede the entire spectrum of potentially damaging solar light are necessary. A better future metric of photoprotection might simultaneously assess total transmittance across the entire spectrum of radiation and then mathematically adjust for the known rate of mutations generated at each wavelength. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cancers14030542/s1, Table S1: Inclusion and exclusion criteria of subjects undergoing in vivo SPF testing of fabrics. Table S2 Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The original data presented in this study are available in the text and supplementary figures. Figure 5 was generated using data available from Coyne et al. [71]. Figure 6 was generated using the publicly available BASF sunscreen simulator: https://sunscreensimulator. basf.com/Sunscreen_Simulator [59]. Conflicts of Interest: E.G.B. has no relevant conflicts to declare; J.B. (Joshua Bezecny) has no relevant conflicts to declare; M.A. works at Exponent Inc., a consulting firm contracted by Columbia Sportswear Company to assist with the measurements and data analysis; T.P.S. works at Exponent Inc., a consulting firm contracted by Columbia Sportswear Company to assist with the measurements and the data analysis; D.M.A. works at Exponent Inc., a consulting firm contracted by Columbia Sportswear Company to assist with the measurements and the data analysis; H.W.B. is an employee of Columbia Sportswear Company; R.A.D. was employed by Columbia Sportswear Company at the time of this project; T.C. is an employee of Columbia Sportswear Company; J.B. (Jennifer Beem) is an employee of Columbia Sportswear Company; D.B. has no relevant conflicts to declare; R.K. has no relevant conflicts to declare; P.B.C. has no relevant conflicts to declare; S.A.L. has no relevant conflicts to declare.	2022-01-28T17:07:36.724Z	2022-01-21T00:00:00.000	{ "year": 2022, "sha1": "58b113a4c716b39428aaa8a720fd89357c9f3554", "oa_license": "CCBY", "oa_url": null, "oa_status": null, "pdf_src": "PubMedCentral", "pdf_hash": "9fe70a4c8165115663451872779e883a291226bd", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
257028420	pes2o/s2orc	v3-fos-license	Comparison of Biaxial Biomechanical Properties of Post-menopausal Human Prolapsed and Non-prolapsed Uterosacral Ligament Uterosacral ligaments (USLs) provide structural support to the female pelvic floor, and a loss of USL structural integrity or biomechanical function may induce pelvic organ prolapse (POP). Alterations in extracellular matrix composition and organization dictate USL mechanical function. Changes in USL microstructure and corresponding mechanical properties, however, are not fully understood, nor is it understood how microstructure and mechanics change with onset and progression of POP. This is due, in part, as USL properties are primarily characterized along a single direction (uniaxial test), whereas the USL is loaded in multiple directions simultaneously within the body. Biaxial testing permits the acquisition of biomechanical data from two axes simultaneously, and thus simulates a more physiologic assessment compared to the traditional uniaxial testing. Therefore, the objective of this study was to quantify the biaxial biomechanical properties and histological composition of the USL in post-menopausal women with and without POP at various stages. Potential correlations between tissue microstructural composition and mechanical function were also examined. Tangential modulus was lower and peak stretch higher in POP III/IV compared to non-POP and POP I/II in the main in vivo loading direction; however, no significant differences in mechanical properties were observed in the perpendicular loading direction. Collagen content positively correlated to tangential modulus in the main in vivo loading direction (r = 0.5, p = 0.02) and negatively correlated with the peak stretch in both the main in vivo (r = −0.5, p = 0.02) and perpendicular loading directions (r = −0.3, p = 0.05). However, no statistically significant differences in USL composition were observed, which may be due to the small sample size and high variability of small sections of human tissues. These results provide first step towards understanding what microstructural and mechanical changes may occur in the USL with POP onset and progression. Such information may provide important future insights into the development of new surgical reconstruction techniques and graft materials for POP treatment. Methods Samples were obtained from the Ochsner Biorepository Unit with signed pre-operative informed consent from women undergoing hysterectomy. All methods were performed in accordance with relevant local guidelines and regulations. Experimental protocols were approved by the relevant local Institutional Review Board (Ochsner Health System IRB approved: 2017.016A). Prior to surgical intervention, two Fellowship-trained Urogynecologists at Ochsner Medical Center in New Orleans, LA determined the POP-Q stage. Menopausal status, the permanent termination of menstruation for more than one year, was collected from the patient's clinical chart/medical record. During surgery, the aforementioned Fellowship-trained Urogynecologists excised bilateral USL samples located at the insertion into the uterine body from post-menopausal women (Fig. 1A). Immediately following excision, the surgeons marked the USL samples with a suture to denote the perpendicular loading direction to ensure sample orientation was consistent during mechanical and histological assessment. Next, samples were snap-frozen and stored at −80 °C until the day of biomechanical assessment. All patient records were de-identified by Ochsner Biorepository Unit personnel. Research personnel were blinded to patient case status and demographics during mechanical and histological analyses to prevent bias. For patient recruitment, the inclusion criteria for non-POP controls (n = 10) were female, menopause, no symptoms or evidence of prolapse on exam, undergoing hysterectomy for benign indications. The exclusion criteria were male, non-menopausal, cancer diagnosis, connective tissue disorder, previous surgery for pelvic floor disorder, symptomatic prolapse, and hormone replacement therapy ( Table 1). The inclusion criteria for the POP cases (n = 14) were female, symptomatic prolapse undergoing hysterectomy for benign indications, and the exclusion criteria (male, cancer diagnosis, connective tissue disorder, previous surgery for pelvic floor disorder, and hormone replacement therapy) ( Table 1). Samples from patients characterized with POP-Q prolapse stages of I and/or II in any compartment were classified as POP I/II (n = 8). POP III/IV (n = 6) were samples from patients with POP-Q stages III and/or IV in any compartment 29,30 (Table 1). Biaxial testing. Prior to the biomechanical test, samples were thawed at room temperature, and cut into squares to ensure consistent sample dimensions and to remove any tissue serosa, charred sections from cauterization, or inhomogeneous portions due to surgical intervention along the main in vivo loading direction (X) and the direction perpendicular to that (Y). Sample orientation was denoted by the surgeon's suture, which was replaced with India ink to denote the perpendicular loading direction. This ensured consistent and repeatable orientation during mechanical testing and subsequent histological analysis (Fig. 1B). Using a non-contacting laser micrometer, tissue thickness was measured across multiple locations and averaged (3.4 ± 0.2 mm). The unloaded width along both directions were also measured with a caliper (7.4 ± 0.2 mm). The minimum length and thickness used were 3.5 mm and 1.3 mm, respectively. The square samples were then speckle coated with alcohol ink (Ranger Industries, Tinton Falls, NJ, USA) to enable optical strain tracking during biomechanical assessment. Four fish hooks were used to mount each sample side (Fig. 1C) into a custom planar biaxial test device equipped with load cells (22 N) in both axes, and a camera 25,27 . The device permits independent control of displacement in the two orthogonal axes. The samples were fully submerged in Hank's balance saline solution (HBSS) at room temperature throughout the experiment. To prevent movement of HBSS around the mounted sample, an acrylic glass table was placed over the bath, making a complete contact with the HBSS solution as described previously 25 . A tare load of 0.1 N was applied in the main in vivo and perpendicular loading directions, followed by 10 equibiaxial preconditioning cycles to a target stress of 0.1 MPa at 0.2%/sec, after which the samples were allowed to equilibrate for 10 minutes. Next, samples were loaded to a maximum stress level of 0.1 MPa in the main in vivo and perpendicular loading directions to different X:Y loading ratios 36-38 of 1:0.5, 1:0.75, 1:1, 0.75:1 and 0.5:1 at 0.2%/sec, for 6 cycles. The maximum stress was informed by prior studies in the USL that indicated no damage to the underlying tissue microstructure during mechanical assessment 26,39 . A loading ratio of 1:1, denotes an equibiaxial loading to the maximum stress of 0.1 MPa, whiles 1:0.5 denotes a maximum stress of 0.1 MPa in the main in vivo loading direction and 0.05 MPa in the perpendicular loading direction. Each loading ratio lasted ~15-20 minutes. Therefore, samples were subjected to mechanical testing for approximately 100 minutes for the 5 loading ratios herein, as well as 5 additional protocols for pilot studies -total ~200 minutes. The experiments were terminated following the completion of the 6 th cycle (once the sample was returned to the original tare load). Data from the 6th unloading cycle was used for the analyses. Using a custom-written Matlab script, the images were exported to GOM Correlate, a commercially available software (GOM GmbH, Braunschweig, Germany, v2.0.1). This software enabled tracking of selected speckle points for non-contact strain measurement. Four points in the central portion of the sample (when connected form a quadrilateral), away from the tissue edges were selected for tracking (Fig. 1C). The displacement components in the main in vivo and perpendicular loading directions of the four selected locations were obtained and . Custom-built planar biaxial testing device equipped with load cells in both axes and a camera (B). USL samples were speckle-coated and mounted in a planar biaxial testing device via fishhooks. Strain was tracked optically in the center of the specimen (4 points shown in blue) (C). Histological sections were obtained along the main in vivo and perpendicular loading directions for the compositional and structural analyses (D). saved as CSV files. Similar to other soft biological tissues, the USL's mechanical response was nonlinear (Fig. 2). Thus, a phenomenological Fung constitutive model was used to describe the experimental data due to low stiffness at small strains, followed by high stiffness at higher strains 36,40 . A custom-written Matlab script was then used to calculate the deformation gradient tensor F ( ) from the displacements of the selected speckle points 41 . The first Piola-Kirchhoff stress tensor P (in both main in vivo and perpendicular loading directions) was calculated from the measured axial load in one direction divided by the undeformed cross-sectional area of the sample. Using Nanson's formula, and with the assumption of incompressibility ( = = J F det 1 ) 37,41 , Cauchy stress, σ was then determined from the relation: It was assumed that the USL maintains its volume during deformation (incompressibile). The Fung-type exponential pseudostrain-energy function is given by: www.nature.com/scientificreports www.nature.com/scientificreports/ deformation. Model parameters c 1 and c 2 indicate the biomechanical contributions from the collagen fibres aligned in the main in vivo and perpendicular directions, respectively. Model parameter c 3 represents the potential interaction between collagen fibres aligned in the main in vivo and perpendicular loading directions -thus may include physical phenomena such as crosslinks or connections between the collagen fibres to each other or to the ground matrix 42 . E xx , E yy are Green strains in the main in vivo and perpendicular loading directions, respectively 42 . The in-plane Green-Lagrange strain tensor E ( ) for each direction was then calculated using is the right Cauchy-Green deformation tensor, and I is the second-order identity tensor. With the assumption of incompressibility, the theoretical Cauchy stress tensor was determined by 43 : where p is a Lagrange multiplier that enforces incompressibility ( = F det 1) Using a custom-written Matlab code and a built-in minimization algorithm (lsqnonlin), the Fung-type constitutive model was used to describe the experimental data (combined ratios) of the control non-POP and POP cases to determine the direction-dependent nonlinear tissue behaviour by minimizing the objective function where σ xx and σ yy denote the Cauchy stress in the main in vivo and perpendicular loading directions respectively and, superscripts th and exp denote theoretically computed and experimentally measured values respectively. Material anisotropy index, φ was calculated based on obtained model parameters of Eq. (2) 36,37 using equation: The material anisotropy index, φ, is a direction-dependent numerical value between 0 and 1 that provides information on the contributions from the main in vivo and perpendicular loading directions to the USL stiffness. Specifically, φ = 1 is an indication that the USL is perfectly isotropic, and smaller values of φ are associated with increasing anisotropy 36,37 . The tangential modulus was calculated as the slope of the linear portion of stress (force/ area)-stretch curve 44 . This is a measure of the resistance to deformation of the USL normalized by cross-sectional area (material stiffness). The tangential modulus determined herein serves as a first approximation to a potential target mechanical property for future graft development, in addition to the aforementioned nonlinear properties quantified herein. The peak stretch was determined from the stretch value that corresponded to the maximum stress. This parameter quantified USL extension under the largest applied load. Histology. Following mechanical testing, samples were cut in half, and each was randomly allocated for histological analyses along either the main in vivo or perpendicular loading direction (Fig. 1D). Each loading direction was marked with a different colour, fixed in a 10% formalin solution for 24 hours, and then embedded in paraffin. Subsequently, 4 µm sections were cut along the main in vivo and perpendicular loading directions (Figurer 1D). Thereafter, sections were stained with Masson's trichrome (MTC), Picrosirius red (PSR), and Hart's elastin stains (Fig. 3). MTC was used to quantify collagen and SMC, PSR was used to quantify large-diameter and small-diameter collagen fibres, as well as collagen structural information (orientation, alignment and straightness ratio). Hart's elastin was also used to quantify elastic fibres. Brightfield (MTC, Hart's elastin) and darkfield (PSR) images were obtained with an Olympus BX51 microscope, an Olympus DP27 digital camera, and cellSense TM software (Olympus Corporation, Center Valley, PA, U.S.A). For the PSR images, a quarter wavelength retardation plate was used, which converts linearly polarized light input to a circularly polarized output wave front by introducing a relative retardation of exactly one-quarter wavelength or 90 degrees between the ordinary and extraordinary wave front. Throughout PSR imaging, the microscope stage was not rotated. All images were obtained at 4× magnification, and additional PSR images were obtained at 40× magnification (Figs. 3E,F) for the quantification of collagen orientation, alignment, and straightness ratio 17,45 . To quantify the SMC and collagen area fractions, MTC images were analysed with ImageJ's (National Institutes of Health, Bethesda, MD, USA) open source colour deconvolution plug-in 45,46 and an open source GNU image manipulation program (GIMP) 45,[47][48][49] . The previously established colour deconvolution used herein is capable of separating the colours in their red, blue and green absorption characteristics [45][46][47] . The red and blue colours represent the area fractions of SMC and collagen respectively. To quantify the large-and small-diameter collagen, PSR images (4×) were utilized, leveraging a custom-written MATLAB code that calculates the area fractions of red and orange pixels (large-diameter collagen), and yellow and green pixels (small-diameter collagen) 45,50,51 . For the quantification of collagen structure, PSR images (40×) were used. An open source MATLAB software framework that includes two separate but linked packages (CurveAlign and CT-FIRE) was employed 52 . The CurveAlign package was used to quantify the collagen orientation and alignment, Non-parametric Kruskal-Wallis test was used for the USL group comparisons. There were no statistically significant differences in both the main in vivo and perpendicular loading directions which may be due to the small sample size, lack of specific immunohistochemical or western blot analyses, or analysis of small sections of tissue that may be represent the entire tissue. LD Col=large-diameter collagen, SD Col=small-diameter collagen. www.nature.com/scientificreports www.nature.com/scientificreports/ while the CT-FIRE was leveraged to quantify straightness ratio. To quantify the elastic fibre area fraction, Hart's stained images were used, employing GIMP's select-by-colour tool to isolate elastic fibres 17,45 . Statistical analysis. First, the clinical demographics of the patient populations between groups (age, BMI, parity, gravidity) were compared to determine if a correction factor was necessary. Then, comparisons between the USL groups (non-POP, POP I/II and POP III/IV) were performed for mechanics (peak stretch and tangential modulus), Fung model parameters (K c c c , , , 3 ), anisotropic index, composition (collagen, large-and small-diameter collagen, elastic fibre, SMC) and collagen microstructure (orientation, alignment, straightness ratio). First, the data was checked for normality with three assumptions, (1) test of normality by Shapiro-Wilk test, (2) homogeneity of variances by Levene's test, and (3) detection of outliers. Values more than 3 times the interquartile range (IQR) were considered as outliers using box and whisker plots 53 . When all aforementioned normality assumptions were satisfied, ANOVA was implemented, followed by post-hoc Tukey's HSD, or an independent samples t-test was used when comparing two groups (risk factor and composition). When the data was not normally distributed, a non-parametric Kruskal-Wallis test (at least 3 groups) or Mann-Whitney U test (2 groups) was implemented. To identify potential correlations between extracellular matrix microstructural composition (collagen, elastic fibre, SMC, orientation, alignment, straightness ratio), and biomechanical function (peak stretch and tangential modulus) Pearson's or Kendall's tau-b correlations were performed if the error was normally or not normally distributed, respectively (Table 2). Further, we investigated potential relationships between POP clinical demographics (age, BMI, parity, gravidity) and histological composition. Biaxial biomechanical properties. All USL samples exhibited a nonlinear behaviour in both the main in vivo and perpendicular loading directions (Fig. 2). POP III/IV USLs were the most extensible when compared to the USLs from non-POP or POP I/II patients in the main in vivo loading direction regardless of loading ratio ( Fig. 2A-D). Contrastingly, no significant differences were observed in the perpendicular direction (Fig. 2E-H). Peak stretch was normally distributed; hence, a 3-way ANOVA was used to determine the potential effects of loading direction, POP group, and loading ratio. Significant main effects (p < 0.001, MSE = 0.001) of loading direction (F = 23.2) and USL POP group (F = 11.5) were identified (Fig. 4A,B). There was also a significant interaction between loading direction and USL POP group (F = 7.6), as well as between loading direction and loading ratios (F = 11.6). Post hoc analyses using Tukey's HSD indicated that POP III/IV was significantly different (p = 0.001) from both non-POP and POP I/II (Fig. 4A,B). Tangential modulus was normally distributed. The 3-way ANOVA (loading direction, POP group, loading ratio) showed a significant main effect (p = 0.02, MSE = 13.1) of USL POP group (F = 4.0) (Fig. 4C,D). Also, there was a significant interaction between loading direction and USL POP group (F = 4.4, p = 0.01), as well as between loading direction and loading ratios (F = 7.7, p = 0.001). Post hoc analyses using Tukey's HSD indicated that POP III/IV was significantly different from both non-POP (0.03) and POP I/II (p = 0.04). No other significant differences were observed for the peak stretch and tangential modulus. fung constitutive model parameters. All the Fung model parameters and the anisotropic index values were not normally distributed; therefore, a non-parametric Kruskal-Wallis test was used to evaluate differences between POP groups (Table 3). No statistically significant differences were identified (p > 0.05) between the groups for model parameters or the anisotropic index. In general, USL samples from POP I/II women exhibited the highest values for all model parameters, including the anisotropy index, except for the parameter K -which describes the global tissue resistance to deformation. www.nature.com/scientificreports www.nature.com/scientificreports/ constituents of USL. The composition data was not normally distributed; therefore, a non-parametric Kruskal-Wallis test was used to evaluate differences between POP groups. No statistically significant differences (p > 0.05) were exhibited between the groups (Fig. 3I,J). For the structural data, collagen orientation and alignment were normally distributed, therefore, a two-way ANOVA (loading direction, POP) was used. The straightness ratio was not normally distributed. Therefore, a non-parametric Kruskal-Wallis test was used. For the collagen orientation, there were no interaction between loading direction and POP group (Table 4). Collagen orientation was significantly different in POP III/IV compared to POP I/II (p = 0.04, Table 4). No statistically significant differences (p > 0.05) were observed in alignment and straightness ratio (Table 4). correlations between composition and mechanics. In the main in vivo loading direction, collagen content significantly correlated with the tangential modulus (r = 0.5, p = 0.02) and negatively with the peak stretch in both the main in vivo (r = -0.5, p = 0.02) and perpendicular loading directions (r = −0.3, p = 0.05) ( Table 2). No other correlations were statistically significant. correlations between composition and clinical demographics. No statistically significant correlations were observed between the USL composition and clinical demographics (age, BMI, parity and gravidity). Discussion This was the first study, to the authors' knowledge, to investigate potential relationships between biaxial biomechanical properties and microstructure in USL tissues from pos-tmenopausal women with and without POP. USLs demonstrated non-linear stress-strain relationships and direction-dependent mechanical function. USL peak stretch increased and tangent modulus decreased in stage III/IV POP compared to both non-POP and POP I/II groups in the main in vivo loading direction, but no significant differences were observed in the perpendicular direction. Further, a significant positive correlation was identified between collagen content and tangential modulus in the main in vivo loading direction. A significant negative correlation was identified between collagen There were significant main effects of loading direction (p < 0.001, peak stretch) and USL group (p < 0.001, peak stretch; p = 0.028, tangential modulus). Also, there were significant interactions (p < 0.01, peak stretch and tangential modulus) between the loading direction and the USL group, as well as, between the loading direction and the loading ratios. Tukey's HSD post hoc analyses indicated that POP III/IV was significantly different from non-POP (p = 0.005, peak stretch) and POP I/II (p = 0.001, peak stretch; p = 0.037, tangential modulus) USLs. (2020) 10:7386 \| https://doi.org/10.1038/s41598-020-64192-0 www.nature.com/scientificreports www.nature.com/scientificreports/ content and peak stretch in both loading directions. However, contrary to our hypothesis, no statistically significant differences in collagen content were observed between POP groups and no significant correlations were identified between USL microstructure and POP risk factors such as BMI and parity. The results herein contribute to the understanding of USL direction-dependent mechanical properties with and without POP. These properties are important to develop new graft materials for POP repair and surgical reconstruction techniques. Results also suggest that it may be important to consider direction-dependent changes in USL microstructure and mechanical content function over time and with POP progression. The increased peak stretch and decreased tangent modulus observed in the POP III/IV group in the main in vivo loading direction suggest that USL mechanical function changes with POP progression which may have implications for pelvic organ support. While the etiology of POP is multifactorial and not fully understood, the observed alterations in the USL suggest that changes in pelvic support tissue mechanical properties may contribute to POP progression. Further, the increased extensibility observed in the POP III/IV group may suggest changes in USL length occur with POP progression due to decreased ability to resist in vivo loads, which may have implications for reconstructive surgical planning 25 . While the correlations identified in this study may suggest that increased extensibility is due to decreased collagen content, no significant differences in collagen content were identified between POP groups. This may be due to the small sample size, as prior studies show decreased collagen content in women with POP or stress incontinence 54,55 . Additionally, Gabriel et al. (2005) reported lower collagen content than that observed in this study. This may be a limitation of the use of PicroSirius Red to quantify collagen content versus specified immunohistochemistry. However, collagen and smooth muscle content measured in this study were similar to those previously quantified using similar methods 25 . These discrepancies may be due, in part, to limitations in histology and immunohistochemistry which analyse small, arbitrary sections of the bulk tissue. In addition to collagen content, increased collagen type III may be connected with pelvic floor dysfunction and advanced utero-vaginal prolapse 56,57 as it corresponds to increased extensibility in various soft biological tissues 19 . While no significant differences were identified in this study, large-diameter collagen fibres (potentially indicating collagen type I) 58 and small-diameter fibres (potentially comparable to collagen III) 58 appear to decrease and increase, respectively, with POP progression consistent with previous studies in other pelvic tissues 19,56,57,[59][60][61] . Future work leveraging immunohistochemistry or protein quantification of type I and III collagen instead of PicroSirius Red would be highly valuable to further delineate relationships between USL microstructure and mechanical function. This study demonstrated that USL mechanical behaviour was anisotropic, similar to prior studies 26 , but no significant differences in collagen content were identified between the two loading directions. However, a significant change in collagen orientation was observed between the POP III/IV and POP I/II groups suggesting remodeling of collagen fibres towards the perpendicular direction. This observation was supported by the smaller, but non-significant value of the anisotropy ratio in the POP III/IV group which may suggest decreased mechanical properties in the main in vivo direction and sustained properties in the perpendicular. It is possible that collagen fibres in the perpendicular direction are maintained throughout POP progression while collagen fibres in the main in vivo direction are potentially damaged due to loading or undergoing remodeling by resident cells. This is potentially supported by the lack of statistically significant differences in mechanical and histological properties in the perpendicular direction. These results may have important implications for designing synthetic or biomaterial grafts to bolster pelvic support, as prior work demonstrates that minimizing differences in graft and adjacent native tissue microstructure and mechanical function is critical to reduce graft complications 62 . Table 4. Structural information (median (min -max)) of USL from women with and without POP obtained from PSR images at 40X magnification (Fig. 3E,F). An orientation of 90̊ indicates that the collagen fibres are predominantly along the direction of loading. www.nature.com/scientificreports www.nature.com/scientificreports/ While no significant differences in peak stretch or tangent moduli were observed between the non-POP and POP I/II groups, the nonlinear model parameters corresponding to direction-dependent properties did increase compared to the non-POP. This may suggest extracellular matrix remodeling occurs with POP onset, but that those changes primarily correspond to decreases in extensibility within the low-strain regime that were not captured by the linearized tangent moduli values that describe the resistance to load in the high-strain regime. Interestingly, the average POP I/II stress-strain curves were shifted to the left, indicating the lowest extensibility compared to both the non-POP and POP III/IV groups. This may suggest that the USL remodels throughout POP progression and that different graft designs may be optimal for interventions at different POP stages. Further, the lack of significant correlation between collagen content and tangential modulus in the perpendicular direction suggests that other extracellular matrix components such as glycosaminoglycans or proteoglycans may also contribute to mechanical function. Contrary to our hypothesis, elastin content was not significantly altered nor significantly correlated to USL function, consistent with previous studies in the USL 57 and contrasting those in the vaginal wall 63 . This may indicate that USL remodelling is driven by changes in loading following levator ani injury and/or collagen remodelling instead of elastic fibres. This may suggest that the structural processes during POP differ for each pelvic soft tissue, even though they are composed of similar proteins. Additionally, there is a considerable amount of smooth muscle cells in the USL 19 , however, smooth muscle cells minimally contribute to passive biomechanical properties 64 measured in this study. Future work is needed to characterize the contractile response of the USL with and without POP to better understand the contribution of smooth muscle cells. To accomplish this, samples must be immediately subjected to mechanical testing following surgery in media containing calcium, as opposed to flash frozen and tested in calcium-free media as performed in this study. Recent studies quantified the active properties of other pelvic tissues, such as the vagina, utilizing media with and without calcium, electric field stimulation, as well as various chemicals to induce either smooth muscle contraction or relaxation 62,65,66 . Smooth muscle cell contractility may play an important role in USL function, therefore future work is needed to understand how active mechanical properties may change in addition to the passive properties quantified in this study. In addition to quantifying relationships between USL composition and mechanical function, potential correlations between established POP risk factors and USL composition were also examined. No significant correlations were identified, potentially due to the small sample size available herein. However, it is also worth noting that the non-POP control group did not show significant differences in BMI and parity compared to those with POP. This was surprising given the strong evidence that high BMI and multiparty are POP risk factors. This may be explained, in part, by the fact that 64% of adults in New Orleans were overweight or obese, with a higher obesity rate (31.4%) than the national rate (27.5%) in 2010 67 and the obesity rate between the ages of 45-64 years was 42.9% in 2018 in Louisiana 68 . Hence, while our study was blind to BMI at the time of patient recruitment, the overall obesity rate in the study's geographical location may have biased the control group. This study was not without limitations. One limitation was the lack of racial diversity in the POP patient population. POP incidence in the African-American/Black population, however, is much lower than other racioethnic groups (Table 1). For example, Latina and white women have 4-5 times higher risk of symptomatic prolapse compared to African-American women 69 . Another limitation is that only a single portion of the USL was examined, yet the USL is heterogeneous 16,[70][71][72][73] . To consistently and repeatably test a similar section of the USL across women, only the insertion sites of the USL were used in this study. Although this site is also the most easily accessible, the USL is a heterogeneous tissue and the biomechanical properties and composition/structure of the insertion site might differ from the main ligament 16,[70][71][72][73] . The insertion site of USL by itself also exhibits greater anatomic variation, which may affect the extensibility and stiffness of the USL 71 . Further, the clinical significance of a small distal portion of the USL is uncertain, as it may not represent how the entire USL remodels and the mid-portion of the USL is used for POP repairs. Regardless, the information obtained herein provides a first step towards understanding how the USL remodels with POP progression in multiple directions and provides potential insights for direction-and stage-dependent treatments for POP. Another limitation was the use of histology to quantify USL composition instead of immunohistochemistry or assays to measure protein content. Further, while histology provided information on collagen orientation in the same region as the mechanical analysis, immunohistochemistry or protein quantification would have provided more specific information on multiple USL components and is needed in future studies. Lastly, while designed to detect statistically significant differences in mechanical properties, the study was underpowered due to the difficulty in obtaining samples from human patients and variability in the small sections of tissue -as a power analysis demonstrates that 54 samples would be required to detect statistical differences in USL composition. Despite these limitations, this comparative study highlighted the importance of the biomechanical function of the USL in post-menopausal women at different stages of POP. In summary, the biomechanical properties in the main in vivo loading direction demonstrated that POP III/IV was more extensible than both non-POP controls and POP I/II. Further, the anisotropy of the USL changes with POP stage, which may be important for POP intervention design. Prior work demonstrated that incorporating biaxial characterization of pelvic floor muscle mechanical function improved the ability of finite element models to test structural and mechanical hypotheses of why and how failures of pelvic floor integrity occur 24 . In this study, we provide the first biaxial mechanical properties of the human USL with and without POP. Such information can be directly incorporated into existing finite element models to better understand how changes in USL function with POP stage may influence POP progression and surgical interventions. Data availability The datasets generated and the codes will be available to the public. Datasets will be stored in Dryad and the codes in GitHub, after manuscript has been accepted.	2023-02-20T14:47:57.928Z	2020-04-30T00:00:00.000	{ "year": 2020, "sha1": "21293988bf5efcea537df3a044024e8a4e3b69c9", "oa_license": "CCBY", "oa_url": "https://www.nature.com/articles/s41598-020-64192-0.pdf", "oa_status": "GOLD", "pdf_src": "SpringerNature", "pdf_hash": "21293988bf5efcea537df3a044024e8a4e3b69c9", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [] }
254854564	pes2o/s2orc	v3-fos-license	Block-wise quantum grayscale image representation and compression scheme using state connection Quantum computing draws huge attention due to its faster computational capability compared to classical computing to represent and compress the classical image data into the quantum domain. The main idea of quantum domain representation is to convert pixel intensities and their coordinates i.e. state label preparation using quantum bits i.e. Qubits. For a bigger size image, the state label preparation takes more Qubits. To address more Qubits issues, a novel SCMNEQR (State Connection Modification Novel Enhanced Quantum Representation) approach has been proposed that uses fewer qubits to map the arbitrary size of the grayscale image using block-wise state label preparation. The proposed SCMNEQR approach introduces the state connection using a reset gate rather than repeating the use of the Toffoli gate used in the existing approach. The experimental results show that the proposed approach outperforms the existing methods in terms of compression. Introduction In the quantum computer discipline, physics, and computer science plays an important role in concrete the quantum information processing (QIP) field [1]. Superposition, Entanglement, and Parallelism are the main properties in quantum computing that make it unique [2,3,4,5,6]. On the other hand, the classical computer is unable to solve the NP (non-deterministic polynomial) hard problems rapidly. Meanwhile, its computing power has not increased significantly in the past decade [7]. Therefore, Feynman et al. demonstrated another way to increase computing power by introducing the quantum computer [8]. An algorithm for factorial calculation of integer was proposed in [9]. After that, following factorial calculation, a database search algorithm was found in [10]. Many other applications, such as flight schedules, and chess playing, it's still suffering from limited algorithmic support like today's classical computer. The ideas from the quantum computer should be on an atomic scale like classical computers [5,6,11,12]. Applications of a quantum computer in terms of image computing are compression, segmentation, security, watermarking, quantum machine learning, and remote sensing [5,13]. The complexity of the classical computer requires O(n * 2 n ) whereas the quantum computer is O(n). In a quantum computer, the number of required operational gates that determine the complexity of the system [14]. The preparation of the grayscale image requires different strategies compared to the color counterpart. In this work, a novel SCMNEQR approach has been proposed for JPEG grayscale image representation and compression arXiv:2212.09222v1 [quant-ph] 19 Dec 2022 to make the system more practical to use. The DCT is used as a preparation step. The main contributions of this proposed work are outlined as follows: • Representing and compressing the grayscale image in a quantum block-wise system • Minimizing the bit rate tremendously compared to the previous approach • Measuring the corresponding rate-distortion curve in terms of the quantum domain • Introducing a novel reset gate in replace of the Toffoli gate The remaining paper is organized as follows. Section 2 outlines the related works. Section 3 discusses the proposed approach. Section 4 represents the computational result and its summary. Section 5 concludes the proposed work. 2 Related Works Figure 1 shows the frequency of publications and citations on quantum image representation published between 1991 and 2022. Interest in this field is gradually increasing and the publication records were found higher in 2022 compared to other years. Figure 2 shows the publications and citations trend for more than three decades for image representation and compression. The graph shows that the initial work of quantum representation and its compression was seen in 1998. After that, there no significant work was done until 2011. The publication report increased exponentially till 2019. Between 2019 and 2021, the number of publications changes randomly. The solid line in Figure 2 shows the citation report for image representation and compression. The citation report started after 2011, and thereafter it grew tremendously. The citation report indicates that quantum computing is a growing field in the research community. However, still there have a lot of areas of development issues that are the main motivation of this research. Qubit lattice is the first quantum approach that demonstrates how to represent and restore a classical image into a quantum system [15]. A Real-ket-based image representation algorithm was proposed by Latorre et al. in [16]. An FRQI (Flexible Representation of Quantum Image) algorithm was developed in [17] that converts four random pixels into its equivalent quantum system using a single ancillary qubit as an angle. The Entanglement image representation was proposed in 2010 [18]. JPEG-based image compression was proposed by Jiang et al. in 2017 [19] that uses GQIR and DCT approaches together. An equivalent bit pixel image is investigated by Laurel et al. in [20]. A NEQR (Novel Enhanced Quantum Representation), was proposed to represent a square grayscale image and resolve the FRQI issue [21]. An INEQR (Improved Novel Enhanced Quantum Representation) approach was proposed [22] that represents tiny rectangular grayscale images rather than a color image. A NASS (normal arbitrary superposition state) was proposed in [23]. An EFRQI algorithm is proposed in 2021 to minimize the state preparation bits of the NEQR approach [24]. Grayscale image-based quantum image compression is found in [25]. Figure 3 shows an example of EFRQI approach for pixel values of 125(X=0,Y=0), 1(X=1,Y=0), 1(X=4,Y=0), 4(X=0,Y=1), and 16(Y=3,X=0) respectively. For each pixel state preparation, the EFRQI approach uses the same Toffoli gate twice (shown in the red circle) for connecting the state values qubits to the pixel values qubits. To complete the full connection, the Toffoli gate generates a higher amount of bits is the main drawback of the EFRQI approach. is considered to represent and compress an image in a block-wise quantum image. Where m and n are the rows and columns of each image. The steps involved in the SCMNEQR approach are. Step 2: Pre-process the pixel or coefficient which requires q + 2n + 1 qubits and initially set all qubit's values to \|0 . Where q is the number of required qubits to map non-zero pixel or coefficient values. On the other hand, n p = log 2 (S x ) and m p = log 2 (S y ) are the position representing qubits for representing the X and Y position of non-zero pixels or coefficients. Where S x and S y are the block size of the X and Y-position images. An auxiliary qubit is used to make the connection between the pixel or coefficient and the position representing qubits. The initial state of qubits can be expressed by the below equation [24]. (1) Then, (q + 1) identity gates and 2n Hadamard gates are used for pixels or coefficient preparation and its state preparation respectively, and shown below. In this step, the whole quantum step can be expressed as follows: The operator U transforms Ψ 0 from the initial state to the intermediate state, ψ 1 . The final preparation step is done using the U 2 quantum operator: where \|C Y X and Y X are pixels or coefficients and the position of the grayscale image. The quantum transform operator is U 2 is given below. The connection of the Toffoli and reset gate is given below. The required bit rate (BR) is calculated using the following equation. Where S state is the state preparation bit. q ones is the frequent number of ones from pixels or coefficients. S bit is the sign bit that represents the sign of the non-zero pixel or coefficient values. A N tcn is the total number of non-zero coefficient or pixel elements. An A bit is the number of bits that come from the auxiliary qubits. B e is the bit rate used to locate block position errors. Step 3: After applying 8X8 DCT, extract the non-zero quantized pixel or coefficient values in the 16 × 16 quantum block system. In the meantime, calculate the bit rate (BR) considering with position error of each block. Moreover, the sign bit (SB) is also considered to account for the sign of each non-zero pixel or coefficient value. Step 6: Measure the quality of the recovery image. Result and discussion In this section, the experimental results are analyzed for Deer(1024×1024), Baboons(512×512), Scenery (512×512), and Peppers(512 × 512) images for verification purposes of proposed approach [26,27,28]. Two 'experiments have been conducted to demonstrate the proposed method's performance. Experiments I and II analyze the proposed scheme's computational result directly and indirectly. Experiment I-result analysis of direct approach This section analyzes the experimental results related to the SCMNEQR direct approach. Figure 5 exhibits the required bit rate of the proposed SCMNEQR scheme compared with NCQI, INCQI, and EFRQI approaches for deer image. The comparison results show that the proposed SCMNEQR approach requires an approximate (21MB) bit rate which is less compared to NCQI(30MB), INCQI(39MB), and EFRQI(29MB) respectively. On the other hand, Figure 6 shows the required bit rate of the proposed SCMNEQR approach for Baboon's image Figure 5: SCMNEQR Bit rate comparison for Deer image compared to the other considered approaches. The comparison result reveals that the proposed SCMNEQR approach can represent Baboon's image efficiently because it requires lower bit rates than others. That means, it represents the Baboons image with low complexity circuit. Therefore, the number of required operational gates is lower in the case of the SCMNEQR approach compared to NCQI, INCQI, and EFRQI approaches respectively. Figure 7 depicts the required number of bit rates for scenery images and compares the result with NCQI, INCQI, and EFRQI approaches. The comparison result reveals that the SCMNEQR approach requires lower bit rates compared to others. On the Figure 6: SCMNEQR Bit rate comparison for Baboons image contrary, the NCQI and EFRQI require moderate bit rates compared to the SCMNEQR approach. Figure 8 presents the required number of bit rates for the scenery image and compares the result with the existing Figure 7: SCMNEQR Bit rate comparison for Scenery image approaches. Comparison results depict that the SCMNEQR approach represents the image inside the quantum processor more efficiently compared to INCQI, NCQI, and EFRQI approaches respectively. The result is expected due to the efficient architecture of the proposed SCMNEQR approach. this scenario, It is concluded that the proposed SCMNEQR scheme performs better than represents Figure 10 presents the comparative results of RDC of the proposed SCMNEQR approach and compares the result with DCT-INCQI, DCT-NCQI, and DCT-EFRQI approaches for Baboon's image. The comparison result shows that the SCMNEQR approach represents and compresses the gray channel of Baboon's image efficiently compared to all other considered approaches. The SCMNEQR approach requires lower operational gates than others means that it has more compression ability. Only, DCT-NCQI exhibits an adjacent bit rate whereas DCT-INCQI stays away from the SCMNEQR approach. Experiment II-result analysis of indirect approach In between DCT-EFRQI and DCT-INCQI approaches, at the initial quantization factor, QF=8, the DCT-INCQI shows better performance compared to DCT-EFRQI. For QF=16, the DCT-EFRQI requires higher operational gates or bit rates than DCT-INCQI. For this reason, a crossover happened between 8 and 16 quantization factors. Figure 11 shows the computational results of the SCMNEQR scheme for scenery images. From a comparative point of view, it has been seen that the SCMNEQR approach has a better ability to compress the scenery image over considered quantization factors compared to all other methods. In between DCT-INCQI and DCT-EFRQI, before QF=16, the DCT-EFRQI approach displays the better result, and thereafter DCT-INCQI performs better over the rest of the other quantization factors. On the contrary, DCT-NCQI requires less bit compared to DCT-INCQI and DCT-EFRQI approaches, but a higher bit rate compared to the SCMNEQR approach. Figure 12 shows the computational result of the SCMNEQR approach alongside DCT-INCQI, DCT-NCQI, and DCT-EFRQI approaches. Results show that the proposed SCMNEQR approach performs better compared to all other considered approaches. The DCT-NCQI approach provides closer RDC compared to the SCMNEQR approach. Conversely, both DCT-INCQI and DCT-EFRQI approaches demonstrate higher RDC values than the SCMNEQR approach. At the quantization factor 8, DCT-INCQI shows better performance compared to the DCT-INCQI approach. After quantization factor 16, the DCT-EFRQI approach requires a higher bit rate than the DCT-INCQI approach. For this reason, a crossover happened between 8 and 16 quantization factors. For the rest of the quantization factor, the DCT-INCQI performs better results compared to the DCT-EFRQI approach. conclusion In this article, we have presented a novel quantum circuit for grayscale quantum image representation and compression. The improvement is done in the state connection circuit for efficient representation and compression. One of the major advantages is that it uses a 16 × 16 quantum block. Besides, the required number of qubits for state preparation is 8. Any size of the grayscale image could be presented using the proposed SCMNEQR approach. In addition, it uses the universal quantum Toffoli gate, reset gate, and auxiliary qubit. Another advantage of the proposed scheme is that it does not require a look-up table to perform the operation. It is concluded that the performance of the proposed scheme is much better than the existing approaches.	2022-12-20T06:42:44.851Z	2022-12-19T00:00:00.000	{ "year": 2022, "sha1": "ce6fc3322ea8ddd4fb10f7f381b9562191dad669", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "ce6fc3322ea8ddd4fb10f7f381b9562191dad669", "s2fieldsofstudy": [ "Computer Science" ], "extfieldsofstudy": [ "Physics" ] }
227154212	pes2o/s2orc	v3-fos-license	Identification of Epithelial–Mesenchymal Transition-Related lncRNA With Prognosis and Molecular Subtypes in Clear Cell Renal Cell Carcinoma Epithelial–mesenchymal transition (EMT), a reversible cellular program, is critically important in tumor progression and is regulated by a family of transcription factors, induction factors, and an array of signaling pathway genes. The prognostic role and biological functions of EMT-related lncRNAs in ccRCC are largely unknown. In the present study, we analyzed the gene expression data and clinical information retrieved from The Cancer Genome Atlas (TCGA) database (N=512) and International Cancer Genome Consortium (ICGC) database (N=90) which served as training and external validation dataset, respectively. Then, we constructed an EMT-related lncRNA risk signature based on the comprehensive analysis of the EMT-related lncRNA expression data and clinical information. The Kaplan-Meier curve analysis revealed that patients in the low-risk and high-risk groups exhibited significant divergence in the overall survival (OS) and disease-free survival (DFS) of ccRCC, as was confirmed in the validation dataset. The Cox regression analysis of the clinical factors and risk signature in the OS and DFS demonstrated that the risk signature can be utilized as an independent prognostic predictor. Moreover, we developed an individualized prognosis prediction model relying on the nomogram and receive operator curve (ROC) analysis based on the independent factors. The Gene Set Enrichment Analysis (GSEA) indicated that patients in the low-risk group were associated with adherens junction, focal adhesion, MAPK signaling pathway, pathways in cancer, and renal cell carcinoma pathway. In addition, we identified three robust subtypes (named C1, C2 and C3) of ccRCC with distinct clinical characteristics and prognostic role in the TCGA dataset and ICGC dataset. Among them, C1 was associated with a better survival outcome, whereas C2 and C3 was associated with a worse survival outcome and have more advanced-stage patients. Moreover, C2 was more likely to respond to immunotherapy and was sensitive to chemo drugs, this may provide insights to clinicians to develop an individualized treatment. Collectively, this work developed a reliable EMT-related lncRNA risk signature that can independently predict the OS and DFS of ccRCC. Besides, we identified three stable molecular subtypes based on the EMT-related lncRNA expression, which may comprehensively be vital in elucidating the underlying molecular mechanism of ccRCC. INTRODUCTION Clear cell renal cell carcinoma (ccRCC) is the most common type of renal cell carcinoma, accounting for 70-80% of RCC cases (1). Surgical resection is the major treatment for localized ccRCC, however, it is associated with a poor prognosis, and approximately 30-40% of ccRCC patients develop to metastatic recurrence during the follow-up (2). Currently, immunotherapies such as programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) or cytotoxic T-lymphocyte-associated protein 4 (CTLA- 4) inhibitors have been approved for managing ccRCC (3). Also, such therapies have proved to inhibit immune checkpoints, thus show efficacy and achieve some progress. However, a percentage of patients remain response poorly, and developed to resistance or progression (4). Consequently, there is an urgent need to identify new biomarkers or therapeutic targets to predict ccRCC progression, prognosis, and response to treatment. Epithelial-mesenchymal transition (EMT) refers to a biological process in which epithelial cells dedifferentiate to mesenchymal cells (5,6). During the EMT process, the epithelial cells lose their polarity, cytoskeletal structure, and cell-cell adhesion but retrieved the migratory properties typical from mesenchymal cells (7). Moreover, a report showed that EMT linked to fibrosis exhibited a prominent feature in ccRCC, which is closely related to a worse survival outcome (8). Thus, the EMT-related genes may serve as a promising target for future therapeutic interventions. However, the expression profile of EMT-related lncRNA, the diverse pathological features of ccRCC and their prognostic value have not been systematically explored. EMT involves substantial molecular reprogramming of cells which endows several clinical features to cancer cells and significantly impact in tumor cell interactions within the tumor microenvironment (9). In addition, EMT is also known to play a crucial role in drug resistance, whereas massive EMT-associated pathways contribute to the drug resistance in tumor cells (10). The cells which are subjected to the EMT process exhibit similar functions to cancer stem cells (CSC), for example, increased drug efflux pumps and anti-apoptotic effects (10). Thus, targeting EMT is regarded as a potential treatment strategy to overcome drug resistance. Herein, we analyzed the RNAseq data and corresponding clinical information retrieved from TCGA (N=512) and ICGC (N=90 ) database to comprehensively explore the prognostic role of EMTrelated lncRNA in ccRCC. Moreover, we assessed the potential molecular subtypes of EMT-related lncRNA in ccRCC patients and analyzed the association between subtypes and immunotherapy, as well as drug sensitivity and tumor microenvironment. Data Acquisition The RNAseq reads count and clinical information were obtained from the TCGA database (https://portal.gdc.cancer.gov/) and ICGC database (https://dcc.icgc.org/), respectively. To ensure high-quality analyses, we retained the samples with a survival time of ≥30 days. Subsequently, 512 ccRCC patients from the TCGA and 90 ccRCC patients from the ICGC database were included in downstream analysis. Moreover, information on the disease-free survival was retrieved from the cbioportal database (https://www.cbioportal.org/). Moreover, the RNAseq transcriptome data were transformed in transcript per million (TPM) value. Correlation Analysis Here, we downloaded 200 EMT-related genes from the Molecular Signature Database v7.1 (MSigDB) (http://www. broad.mit.edu/gsea/msigdb/). To identify the EMT-related lncRNAs, we firstly extracted all lncRNA expression data in the TCGA database based on the GENCODE project (http:// www.gencodegenes.org). Then, Pearson correlation analysis was performed between EMT-related genes and all lncRNA expression data in samples to identify the EMT-related lncRNA on the basis of the correlation coefficient and p values (\|Cor pearson \| > 0.3 and p value < 0.01). Constructing a Risk Signature A univariate Cox regression analysis was performed to select significant EMT-related lncRNAs associated with the survival of ccRCC (p < 0.05). Then, we selected the lncRNAs with significant clinical variables and conducted on feature selection using the randomForestSRC software package (https://cran.r-project.org/ package=randomForestSRC). The randomSurvivalForest algorithm was employed to rank the prognostic genes (ntree=1000) based on their importance. The lncRNAs with a relative importance > 0.4 were subjected to a multivariate Cox regression analysis, after which we constructed a risk signature using the Akaike information criterion for stepwise backward/forward model selection. The risk score for each patient was calculated using the risk formula: where Exp i represent the expression of each prognostic lncRNA and b i represent the coefficient of each prognostic lncRNA. GSEA Enrichment Analysis To identify the potential KEGG pathway that involved in the lncRNA risk signature, Gene Set Enrichment Analysis (GSEA) was performed to find the significant enriched term in the highrisk and low-risk group The pathways with p < 0.05 and FDR < 0.05 were considered as statistically significantly. Independence of the EMT-Related lncRNA Signature The lncRNA signature and corresponding clinical information of the DFS and OS were exploited to identify the independence using the univariate Cox regression analysis and multivariate cox regression analysis. p < 0.05 was considered as statistically significant. Nomogram Construction and Validation The nomogram was established based on the all independent prognostic factors using the rms R package (https://cran.rproject.org/web/packages/rms/index.html). The calibration plot curve analysis was applied to evaluate the discrimination and the calibration of the nomogram. Non-Negative Matrix Factorization (NMF) Clustering To explore the potential molecular subtype, a non-negative matrix factorization (NMF) clustering algorithm was applied to cluster the ccRCC sample via the "NMF" R package (11). The number of cluster k was set from 2 to 7. Then, we selected the optimal k value based on the cophenetic coefficient. The gene mutation for each subtype was calculated, from which we selected the top 20 genes to visualized using the maftools R package. Tumor Microenvironment Analysis To assess the tumor microenvironment in ccRCC, we determined the infiltration levels of 22 immune cells using the CIBERSORT algorithm based on the all gene expression levels. We uploaded the gene expression data the to the CIBERSORTx web portal. Then, the algorithm was run using the LM22 signature for 1000 permutations (12). The ccRCC samples with an output P-value < 0.05 were selected for further analysis. Moreover, the immune core, and the stromal score were calculated using the "estmate" R package (http://r-forge.rproject.org). Predicting Chemotherapeutic Response We predicted the chemotherapeutic response for each ccRCC patient based on information retrieved from the Genomics of Drug Sensitivity in Cancer (GDSC) database (13). Two common chemo drugs, sorafenib and sunitinib, which have been approved for treating of metastatic RCC cases were selected to predict the chemotherapeutic response (14). The prediction procedure was conducted using the R package "pRRophetic" where the halfmaximal inhibitory concentration (IC50) of the samples was predicted by ridge regression. The accuracy was calculated through 10-fold cross-validation based on the GDSC training dataset (15). Statistical Analysis All statistical data were analyzed in the R environment (R version: 3.6.2). We applied the Wilcoxon test (Mann-Whitney test) to analyze continuous variables, whereas the Fisher's exact test or chi-square test was used to analyze the categorical data. The survival difference was calculated using the K-M analysis methods and the log-rank test. For all statistica analyses, a P-value less than 0.05 was regarded as statistically significant. EMT-Related lncRNA Identification The EMT-related genes were retrieved from the MSigDB database with the hallmark gene sets name: HALLMARK_ EPITHELIAL_MESENCHYMAL_TRANSITION. As a result, a total of 200 EMT-related genes were collected (Table S1). We characterized the EMT-related lncRNAs through correlation analysis. According to the screening criteria, a total of 2019 EMT-related lncRNAs were identified with the absolutely Pearson coefficient > 0.3 and P-value < 0.01 ( Table S2). The clinical information and data on lncRNA expression were merged for the downstream analysis. Constructing the EMT-Related lncRNA Signature Here, 512 ccRCC patients with survival time ≥ 30 days, and 2019 EMT-related lncRNAs were included in the TCGA-ccRCC cohort to identify the prognostic risk model. Through univariate Cox regression analysis, we obtained 491 lncRNAs with a significant prognostic difference. Then, the randomSurvivalForest algorithm was adopted to make a feature selection. The selected genes with a relative importance > 0.4 were further applied for multivariate stepwise Cox regression analysis. The relationship between the error rate and the number of tree is shown in Figure 1A, while the relative importance of genes based on the criteria is shown in Figure 1B. We then established an 11-lncRNA signature model by through a multivariate stepwise Cox regression analysis. The risk score for each patient in the TCGA cohort and ICGC cohort was calculated on the basis of the risk formula: Prognostic Value of the EMT-Related lncRNA in ccRCC We determined the potential of using the EMT-related lncRNA signature to predict the overall survival (OS) and disease-free survival (DFS) to predict the prognosis of ccRCC patients. Patients were categorized into high-risk group and low-risk group based on the median risk score. Then. the Kaplan-Meier curve analysis was conducted to assess the OS and DFS outcomes between two risk groups. As depicted in Figures 2A, B, the OS and DFS rate of patients in the low-risk group were significantly higher than those in the high-risk group (P < 0.001). Similarly, patients in the ICGC dataset exhibited a prolonged survival time in the low-risk group when compared to the high-risk group ( Figure 2C). Additionally, it was found the OS and DFS patients in the high-risk group were corresponded to more death cases, and high expression of APCDD1L−DT, LINC01559, AC063948.1, THUMPD3−AS1, and CD27−AS1, whereas, more alive cases and high expression level of LINC00957, LINC01507, LINC02532, AL357140.2, DOCK9−DT and AC002070.1 were reported in the low-risk group ( Figure S1). Based on the ROC analysis results generated by the risk model, the AUC value in the TCGA cohort and ICGC cohort was reached 0.761 and 0.742, indicating a good 5-year prediction prognostic accuracy ( Figures 3A, B). In addition, we performed a univariate cox regression and multivariate cox regression analyses to determine whether the EMT-related lncRNA signature can serve as an independent prognostic predictor for OS and DFS in ccRCC patients. As showed in Figures 4A-D, we observed that the lncRNA signature, stage and grade were listed as an independent predictor for OS and DFS. To further explore the prognostic value of the EMT-lncRNA signature in ccRCC patients stratified by clinical variables, we assigned the patients into different groups based on age, gender, stage and grade. Considering the different stratified analysis results, ccRCC patients in the low-risk group were characterized by a significantly prolonged OS time than patients in the high-risk group (P < 0.05) ( Figure 5). These findings demonstrated that EMT-related lncRNA signature for OS can predict the prognosis of ccRCC without considering the impact of clinical factors. Besides, we evaluated the ability of the EMT-related lncRNAs to promote the progression of ccRCC. Of note, we found that the risk score presented a significantly increasing trend in the stage, grade (Kruskal-Wallis P < 0.05). Furthermore, the risk score in males was significant higher than in females, whereas no significant difference was observed with age ( Figure 6). These findings suggested a higher the risk score, and a higher malignancy of the ccRCC. Therefore, the EMT-lncRNA signature could accurately predict the progression of ccRCC. Construction and Validation of Nomogram in the TCGA Cohort The nomogram was established based on the independent factors, including EMT-related lncRNA signature, stage and grade ( Figure 7A). The calibration plots present high performance in predicting 1-year, 3-year and 5-year OS in ccRCC ( Figures 7B-D). Furthermore, the prediction accuracy of the nomogram was evaluated via ROC analysis. Results showed that the AUC value for the 1-year, 3-year and 5-year in ccRCC was 0.846, 0.807, and 0.751, respectively ( Figure 7E), suggesting that our nomogram was highly accurate. GSEA Enrichment Analysis Result of the EMT-Related lncRNA Signature The Gene Set Enrichment Analysis (GSEA) analysis was applied to identify the significant pathway associated with the high-risk group and low-risk group in the TCGA cohort and ICGC cohort. Patients in the low-risk group were mainly enriched in focal adhesion, MAPK signaling pathway, pathways in cancer and renal cell carcinoma pathway (Figures 8A, B). However, no significantly pathway was enriched in the high-risk group. Three Molecular Subtypes of ccRCC Identified by NMF Clustering To explore the potential molecular subtypes of ccRCC based on the EMT-related lncRNAs, the lncRNAs with significant survival differences were selected following results from the univariate Cox regression analysis result. After screening, a total of 491 lncRNA with 512 patients in the TCGA cohort were included in the NMF consensus clustering analysis. The cophenetic correlation coefficients were calculated to determine the optimal k value, and k=3 was selected as the optimal k value after a comprehensively consideration (named C1, C2 and C3) ( Figure 9A). The consensus heatmap showed a sharp and crisp boundary for each subtype, suggesting that the robustness and reliability of the subtype ( Figure S2). Moreover, the principal component analysis (PCA) results indicated that there is a significant difference between C1, C2, and C3 ( Figure 9B). Further, the K-M curve analysis results suggested that subtype C1 had a better overall survival than subtype C2 and C3 (P < 0.001) ( Figure 9C). To further validate the stability of molecular subtype, we conducted an NMF clustering analysis in the ICGC cohort. Notably, similar results were obtained, except for the survival analysis results which could be attributed to the small sample size ( Figure S3). Landscape of Genomic Profiling and Immune Infiltration Level in Subtypes To explore the relationship between subtype of ccRCC and its clinical factors in the TCGA dataset, we generated a heatmap to depict the association between the lncRNA expression, molecular subtype and clinical factors ( Figure 10). It was found that the lncRNA expression is highly specific in subtypes, indicating that the different subtypes exhibit different functions. Moreover, we revealed that C1 was highly associated with more stage I, stage II, grade 1, and grade 2 patients, while C2 highly corresponded to more stage III, stage IV, grade 3, and grade 4 patients, demonstrating that the C2 subtype is related to the progression of ccRCC. Similarly, findings were reported in the ICGC result ( Figure S4). Also, the VHL and PBRM1 were found to be the most common mutation genes among the subtype, which is consistent with the previously reported ( Figure S5). Previously studies found that the EMTrelated genes are associated with the tumor microenvironment, whereas they were not discovered in the ccRCC. This prompted us to further explore the relationship between molecular subtype, immune checkpoint, and immune cell infiltration level. We found that the immune cell infiltration level including CD8 T cell, follicular helper T cells, CD4 memory activated T cells and gamma delta T cells were higher in C2 when compared to C1 and C3, whereas high levels of resting Dendritic cells, CD4 memory resting T cells and M1 Macrophages were found in subtype C1 ( Figure 11). In addition, the overall expression levels of known immune checkpoints including IL6, CXCR4, CD276, TGFB1, November 2020 \| Volume 10 \| Article 591254 CCL2, CTLA4, LAG3, CD274, and CD 4 were relatively higher in subtype C2 compared to subtype C1 and subtype C3 (Figure 11). More Sensitivity to Immuno/ Chemotherapies for C2 Subtype To assess the likelihood of how the three subtypes respond to immunotherapy, we used the TIDE algorithm. Results demonstrated that subtype C2 (30.3%, 44/145) was more likely to respond to immunotherapy than C3 (28.0%, 52/186) and C1 (16.0%, 29/181) (Kruskal-Wallis P < 0.001). In addition, sorafenib and sunitinib chemo drugs were applied to the treatment of metastatic RCC cases. Therefore, we further evaluated the response of three subtypes to two chemo drugs using the GDSC cell line data set. As shown in Figures 12A, B, it was evident that subtype C2 was more sensitive to sorafenib chemo drug compared to subtype C1 and C3. However, no significant difference was observed in the sunitinib drugs among the subtypes. Notably, subtype C2 was associated with more advanced patients and was susceptible to local recurrence or distant metastasis. In a nutshell, we concluded that the C2 subtype can benefit from the sorafenib therapy. DISCUSSION ccRCC is the most frequently subtype of renal cancer, which is highly associated with poor prognosis and distant metastasis (16). Exploring potential biomarkers is vital to the management and prognosis of ccRCC. Accumulated evidence shows that EMT is associated with the progression and metastasis of cancer (17). However, a majority of the studies focused on the impact of the EMT in tumor development and treatment, whereas fewer studies addressed the prognostic value of EMT-related genes or lncRNAs in cancer, especially in ccRCC. In the present study, we constructed a novel and efficient EMTrelated lncRNA signature based on the TCGA dataset and validated its efficiency in the ICGC dataset. The ROC analysis result in the TAGC dataset and ICGC dataset confirmed the high prognostic value of our signature. Moreover, the signature showed a significant correlation with clinical factors, further supporting the robustness of its prognostic value. We also demonstrated that the EMT-related lncRNAs can potentially be utilized as an independent predictor for the OS and DFS in the TCGA dataset. The nomogram constructed by the stage, grade and our signature showed a high performance in the 1-, 3-and 5-year, which may contribute to promote the individualized treatment of ccRCC patients. Collectively, these findings demonstrated that a great prognostic value of our EMT-related lncRNA signature, thus may provide a theoretical basis for EMT-related targeted therapies. In addition, our GSEA results indicated that numerous pathways were significantly enriched in the low-risk group, suggesting that EMT exerts more regulatory roles in the low-risk group compared to the high-risk group. Furthermore, most of the lncRNAs in our signature had been reported in the cancer types. For instance, LINC00957 was proved to be a potential biomarker and therapeutic target for colorectal cancer patients (18); LINC02532 was demonstrated to act as an oncogene in gastric cancer (GC) and promoted the GC cell proliferation, migration, and invasion (19); DOCK9-DT was revealed to play a protective role in the prognosis of thyroid carcinoma (20); THUMPD3-AS1 regulated non-small cell lung cancer cell self-renewal via the expression of miR-543 and ONECUT2, and THUMPD3-AS1 can serve as a potential biomarker or therapeutic target in non-small cell lung cancer (21). Additionally, APCDD1L-DT exhibits a significant prognostic value in lung squamous cell carcinoma (LUSC) and can independently predict the overall survival in LUSC patients (22). LINC01559, which can promote the pancreatic cancer (PC) cell proliferation and migration via the YAP-mediated pathway, providing a novel target therapy for clinical treatment of patients with PC (23). Moreover, the LINC01559 can act as an potential oncogene, thereby accelerating the resistance of hepatocellular carcinoma to oxaliplatin by sponging miR-6783-3p (24). Currently, the molecular mechanism or prognostic value of other lncRNAs including LINC01507, AL357140.2, AC063948.1, CD27-AS1, and AC002070.1 are elusive. Despite important prognostic value of lncRNAs, future experiments are necessary to elucidate their role in ccRCC. Although numerous ccRCC molecular subtypes based on gene expression have been proposed in recent years, those associated with the EMT-related lncRNAs are yet to be fully explored. Herein, we assessed the ccRCC subtypes associated with EMT, using the NMF algorithm to perform a consensus clustering to the ccRCC samples based on EMT-related lncRNA. Of note, three molecular subtypes (C1, C2 and C3) were identified and validated in the TCGA and ICGC datasets. The PCA findings confirmed the robustness and reliability of the subtypes identified in the present study. Furthermore, the K-M curve result suggested that subtype C1 was was associated with better survival outcomes and in a majority of early-stage patients, whereas subtype C2 was associated with worse survival outcomes and in more advanced-stage patients. Previous investigations reported that tumor microenvironment is associated with EMT. Thus, the relationship between tumor microenvironment and subtype were further explored. In the present work, we found that the CD8 + T cell, follicular helper T cells, CD4 memory activated T cells and gamma delta T cells were presented high expression level in subtype C2, interestingly, the overall expression level of immune checkpoints also showed a high expression level in subtype C2. These results can be interpreted that the anti-tumor effect of high level of T cell infiltration is offset by the strong immunosuppressive pathway activated by over-expressed immune checkpoint proteins, and might provide the likelihood of immunotherapy for C2 patients (25,26). We further evaluated the likelihood of response to immunotherapy in the three subtypes using the TIDE algorithm. Notably, subtype C2 was found to be more likely to respond to immunotherapy. In addition, we demonstrated that sorafenib was more sensitive to the subtype C2. The sorafenib has been used in the treatment of metastatic RCC cases, with considerable progress (27). Since the subtype C2 corresponded to more advanced stage patients, therefore, we inferred that subtype C2 could more likely be utilized for treatment by the sorafenib. This study has a few limitations: Firstly, our risk signature was constructed based on the public dataset, some important clinical information was not complete and unavailable for analysis, this could introduce potential basis or errors. Secondly, the ICGC dataset sample size was small compared to the TCGA dataset, and may have contributed to the non-significant results, such as in three molecular subtype survival analysis. Thirdly, the EMTrelated lncRNA signature should be validated in future studies through in vivo and in vitro experiments. In conclusion, the present work developed and validated an EMT-related lncRNA signature that can be utilized as a reliable tool for predicting individualized prognosis and for decisionmaking when treating ccRCC patients. In addition, three molecular subtypes were revealed, which may contribute to understanding the molecular mechanism of ccRCC and provide references for clinicians to develop an individualized treatment for ccRCC patients. DATA AVAILABILITY STATEMENT Publicly available datasets were analyzed in this study. The datasets generated for this study can be found in https://portal. gdc.cancer.gov/ and https://dcc.icgc.org. AUTHOR CONTRIBUTIONS YL and JH designed the study. CH and FZ collected the clinical information and gene expression data. WZ analyzed the data and wrote the manuscript. YL and JH revised the manuscript. All authors contributed to the article and approved the submitted version. ACKNOWLEDGMENTS We appreciated TCGA database for providing the original study data. SUPPLEMENTARY MATERIAL The	2020-11-25T14:10:19.529Z	2020-11-25T00:00:00.000	{ "year": 2020, "sha1": "91c2ffe86dfb9fa6397d40f178bcb47e03860095", "oa_license": "CCBY", "oa_url": "https://www.frontiersin.org/articles/10.3389/fonc.2020.591254/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "91c2ffe86dfb9fa6397d40f178bcb47e03860095", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Medicine", "Biology" ] }
235561341	pes2o/s2orc	v3-fos-license	New Insights Into the Promising Antibacterial Activity of Thiophene or Chromene Moiety Containing Aryl Sulfonamide Background: Aryl sulfonamides bearing thiophene and chromene moieties have been reviewed for their antibacterial activity and their synthetic methods. Heterocyclic moiety containing aryl sulfonamide compounds are dispersed in nature and are crucial for life. Diverse investigational strategies towards a structural relationship that cognizance upon the exploration of optimized candidates have grown to be extremely crucial. Method: Literature research tells that for a series of thiophene or chromene moiety containing aryl sulfonamide compoundsare vital in medicinal and industrial chemistry.Aryl sulfonamidescontainingheterocyclic moieties display pharmacological activities against pathogenic microbes. Result:Recent various disciplinary reported articles had been cited in this review article to dene the potentialantibacterial properties of thiophene-arylsulfonamide and chromene-arylsulfonamide. Conclusion: The nding of this review conrms the importance of aryl sulfonamidescontaining thiophene or chromene moiety as potential antibacterial agents. These nal resultswill give ideasto the synthesis and developmentof reactions leading to the potential derivativesfor betterpharmacological applications. Introduction The millions of people are affected by infectious diseases caused by bacteria which lead to the death of many people worldwide. 1 In developing countries, bacterial infections caused by pathogenic microorganism lead to life-threatening diseases. 2 Presently, antibacterial agents which are commonly in use are Fidaxomicin, Cethromycin, Daptomycin, Torezolid and Solithromycin. 3 Antibacterial agents/Antibiotics are used to kill or control the growth of pathogenic bacteria but the long term usage of antibacterial agents there are chances that bacteria develop resistance against antibacterial/antibiotics. [4][5] In the current situation, multidrug-resistant microbial pathogens have emerged which affect the treatment of infectious diseases. 6 Some of the antibacterial agents like tetracycline are persistent in nature, owing to which they kill bene cial bacteria and cause gastrointestinal distress, yellowing of teeth 7 , increased intestinal peristalsis related to erythromycin therapy 8 ,cause discolouration of skin 9 , condrotoxicity 10 , ototoxicity 11 , retinopathy 12 , lactic acidosis and serotonin syndrome 13 , aplastic anemia 14 , hepatitis 15 , neuromuscular blockade 16 and neoplasia. 17 Numerous pathogenic bacteria occur in nature but the available antibacterial agents or antibiotics have limited inhibitory or killing action on bacteria. Few antibiotics have been found to exhibit activity against Gram-negative and Gram-positive bacteria. [18][19] Organic chemistry has a signi cant impact on our lives, as organic compounds are extensively distributed and play a vital role in various biological elds. This approach has emerged and come to the centre stage in the form of Green Chemistry. The class of sulfonamide is an emerging group of research and most of the organic chemists are engaged in the synthesis of novel sulfonamide. [20][21][22][23] Sulfonamide derivatives are usually prepared by the reaction of ammonia or primary or secondary amines with a sulfonyl chloride. 24 Sulfonamide derivatives have been obtained by the reactivity of sul nic acid salt with highly electrophilic nitrogen from various organic sources. [25][26] Intermolecular free radical reactions of penta uorophenylvinyl sulfonate with subsequent aminolysis reaction are a very convenient route to synthesize sulfonamides. 27 Among the broad variety of medicinal compounds, the functional groups of sulfonamide pharmacophore play a unique role in drug design. 28 Sulfonamidesdisplay a broad range of biological activities,including activity against microbial infection. [29][30] Derivatives of sulfonamides used for the treatments of infectious diseases caused by pathogenic microorganisms due to their growthinhibiting tendency. 31 Compounds carrying sulfonamide core moieties, which construct them signi cantly in drug development, have demonstrated signi cant biological activities. 32 Sulfonamides an antibiotic drug are linked with medicinal activities. Sulfonamides have lately exposed them to be distinctly competent synthons within the practice of diverse treasured biologically active compounds. 33 The derivatives of 9-sulfonylamino scaffold were enhanced antibacterial property against an array of minocycline and tetracycline-resistant pathogens as an example, staphylococcus aureus developed resistant from methicillin, carbapenems, penicillin, quinolone, cephems etc [34][35][36] , while sulfonamides and their combination healing procedures are gaining more attention by means of the day in antimicrobial drug research. 37 As a result, it is viable to expand a sequence of sulfonamides with alkyl amide side chain amendment on carboxyl for enhanced antibacterial e ciency. Sulfonamides have medicinal applications and many of them are broadly utilized in therapeutic as antimicrobial 38 , antitumor [39][40] , anticancer 41 77 developed method for the preparation of alkyl aryl sulfonamide derivatives (10). Aryl hydrazine, hydrazines and potassium metabisul te (K 2 S 2 O 5 ) react together in methylcynide under an air atmosphere at 40 o C. In this process, aryl hydrazine was oxidized by air into aryl radicals and potassium metabisul te was used to generate sulfur dioxide to the formation of sulfonamide derivatives. Some sulfonamide moieties containing compounds are available for commercial importance and widely used in the clinic. Sulfonamides antibiotics are clinically approved drugs were used to prevent many infectious diseases including pathogenic bacteria known as sulfa drugs Antibacterial drugsul soxazole enhanced antibacterial activities such as p-aminobenzoic acid and sulfamethazine. 92 Sulfamehazineand tiamulin mixture possessed synergistic antibacterial properties against pathogenic bacteria were isolated from pigs. 93 The chlorothiazide and hydro chlorothiazide are the class of sulphones series were evaluated as antibacterial and antifungal. 94 Sulfonamide core moiety scaffolds have been found to exhibit diverse signi cant biological properties also found in some natural compounds. Generally, sulfonamide moiety compounds were isolated from marine actinomycetes. Cytotoxic monoterpene-alkaloid (-)-altemicidin were isolated from active fractions of Streptomyces sioyaensis. 95 (-)-Altemicidin showed inhibitory activity against bacterial strains (Muralidharan andDeecaraman, 2017) 96 were reported mild growth inhibition against Xanthomonas species. The bromotyrosine-cysteine derivatives, psammaplin A and C were extracted from the sponge Psammaplysilla purpurea are e cient antibacterial agents. 97 Nucleocidin are uorinated sugar structure used as an antibiotic and it were obtained from soil microbe Streptomyces calvus 98 andStreptomyces albo avus. 99 In addition, potent antibacterial activity against both Gram (positive and negative) pathogenic bacteria strains. Thiophene moiety containing sulfonamide derivatives For the preparation of thiophene moiety containing sulfonamide derivatives via Suzuki cross-coupling reaction, a convenient approach was published in the literature. 100 Thiophene moiety containing compounds are excellent bioactive agents. They have been found to exhibit various biological activities such as antimicrobial [101][102] , anti-HIV 103 , anti-in ammatory 104 , anticancer. 105 Cytochrome inhibition has been observed in some thiophene derivatives. 106 Thiophene-2-sulfonamides derivatives are also regarded as inhibitors of carbonic anhydrase, and the literature indicates that diuretic activity is also seen in many of its simple derivatives. The reaction of benzenesulfonamide and (benzothiophen-3-yl)-3-chloropropan-1-one using THF as solvent and K 2 CO 3 as catalyst affords substituted benzothiophene (N-aryl) sulfonamide (4) in good yield. 111 The synthesized thiophene aryl sulfonamide (5) was evaluated in-vitro antibacterial activity against pathogenic bacteria such as B.subtilis, E.coli, B. megaterium and P. ourescens. 112 Thiophene moiety containing N-substituted aryl sulfonamides (6) have been synthesized from the reaction of 3-(thiophene-2-ylmethylamino)propanenitrile with p-toluenesulfonyl chloride using Et 3 N as catalyst and DCM as a solvent. 113 Thiophene-sulfonamide compound does not exist in a natural source. Thiophene exists in natural sources with promising biological applications but the compounds of thiophene moiety sulfonamide scaffold only available in synthesized form. Chromene moiety containing sulfonamide derivatives The core entity of several biologically vigorous natural products, as well as synthetic therapeutic agents, consists of unsaturated 1-benzopyran derivatives, widely known as chromenes. 122 The chromene nucleus is, thus, widely known in medicinal chemistry as a privileged scaffold. 123 4H-chromene and 2H-chromene are an isomeric form of each other, which depending on their substitution pattern, can be differentiated by the site of unsaturation and show a range of pharmacological properties. Chromene moieties play a signi cant role with structural aryl sulfonamide scaffold which can be easily transformed into functionalized diverse biologically active molecules. 124 Naturally occurring 125 and synthetic chromene compounds possess antibacterial properties. 126 A number of pathways are available for their synthesis, but there is a signi cant need for approaches that include chromenes from precursors that are readily available. 127 Some synthesized sulfonamide derived chromenes were investigated for antibacterial, antifungal and cytotoxic. 128 Chromone is known as a single molecule that can be paired with various receptors groups. 129 Due to its useful activities and low toxicity, chromone is regarded as a desirable source for the synthesis of new drugs. Chromene-sulfonamide compound does not exist in a natural source. Chromene and its derivatives exist in various natural sources with promising biological applications but the compounds of chromene moiety sulfonamide scaffold only available in synthesized form. Conclusion Aryl sulfonamides bearing thiophene or chromene moieties have been reviewed for their antibacterial activity and their synthetic methods.The review will provide insight to readers,to tap the potential of these classes of compounds. More compounds of these classes need to be synthesized with wide structural variation in order to establish structure-activity relationship and nd out new compounds having promising antibacterial properties. Declarations Con ict of Interest There is no con ict of interest between coauthor. Funding No funding was supported by any organizations for this work. Figure 2 Advance method for the preparation of aryl sulfonamides. Natural bioactive sulfonamide compounds Figure 5 Preparation methods of thiophene-aryl sulfonamides Figure 6 Preparation methods of myriad thiophene containing aryl sulfonamides Bioactivethiophene-sulfonamides agents Figure 9 Preparation methods of chromene containing aryl sulfonamides Figure 10 Preparation methods of myriad chromene containing aryl sulfonamides Supplementary Files This is a list of supplementary les associated with this preprint. Click to download.	2021-06-22T17:55:33.927Z	2021-04-27T00:00:00.000	{ "year": 2021, "sha1": "92e8c4d5a9b88894ad4c7d2201b93f5b42fec179", "oa_license": "CCBY", "oa_url": "https://doi.org/10.21203/rs.3.rs-452693/v1", "oa_status": "GREEN", "pdf_src": "Adhoc", "pdf_hash": "d849757902e94dc761ca0783be1c66adc6a6eadb", "s2fieldsofstudy": [ "Chemistry", "Medicine" ], "extfieldsofstudy": [ "Chemistry" ] }
235446362	pes2o/s2orc	v3-fos-license	Probing two-electron multiplets in bilayer graphene quantum dots We report on finite bias spectroscopy measurements of the two-electron spectrum in a gate defined bilayer graphene (BLG) quantum dot for varying magnetic fields. The spin and valley degree of freedom in BLG give rise to multiplets of 6 orbital symmetric and 10 orbital anti-symmetric states. We find that orbital symmetric states are lower in energy and separated by $\approx 0.4 - 0.8$ meV from orbital anti-symmetric states. The symmetric multiplet exhibits an additional energy splitting of its 6 states of $\approx 0.15 - 0.5$ meV due to lattice scale interactions. The experimental observations are supported by theoretical calculations, which allow to determine that inter-valley scattering and 'current-current' interaction constants are of the same magnitude in BLG. Graphene quantum dots (QDs) are considered promising candidates for spin-based quantum computation, as the low spin-orbit and hyperfine coupling provides long spin coherence times [1][2][3][4][5]. In addition to the spin, graphene offers the valley degree of freedom, which gives rise to a rich energy spectrum and creates the opportunity for the implementation of valley and Kramer's qubits [5][6][7]. Recent experimental progress on electrostatically confined bilayer graphene (BLG) QDs, demonstrating single-electron occupation [8][9][10], gate-tunable valley g-factors [11] and low spin-orbit coupling [12][13][14], brings graphene based qubits within reach. As twoelectron states are particularly interesting for the implementation of well-controllable qubits, such as exchange and singlet-triplet qubits [15,16], which offer various advantages over single-electron qubits, a detailed understanding of the two-particle spectrum in BLG QDs is becoming increasingly important. This is all the more true since the spin and valley (K + , K − ) degrees of freedom in BLG yield a total of 16 twoparticle states where the wavefunction-dependent valley g-factors give rise to a rich level spectrum. The total two-particle wavefunction in BLG can be factorized into an orbital, a spin and a valley term [17,18], resulting in 6 states with an anti-symmetric spin-valley and a symmetric orbital wavefunction, and 10 states with a symmetric spin-valley and an anti-symmetric orbital wavefunction. This gives rise to the symmetric and anti-symmetric multiplet structure of the two-electron spectrum in BLG. In this letter, we report on the experimental observation and detailed description of the symmetric and antisymmetric two-electron multiplets in BLG QDs. We confirm the Berry curvature induced wavefunction dependence of the valley magnetic moment, observing different valley g-factors for single-particle, symmetric and antisymmetric orbital wavefunctions, which gives rise to a rich magnetic-field dependent transition structure. Additionally, we show that the energy splitting between the multiplets (∆ Orb ) can be tuned in the range of 0.4 − 0.8 meV and that the splitting of the orbitalsymmetric multiplet allows to quantify inherent lattice scale interaction constants in BLG [19][20][21]. We reconstruct the two-electron spectrum of the BLG QD from finite bias spectroscopy measurements of the N = 1 → 2 transition. Therefore, it is necessary to consider both single-and two-particle states, which are presented in Figs. 1(a,b). Fig. 1(b) shows the energy of the four single-particle states of the lowest orbital as a function of an out-of-plane magnetic field, B ⊥ . At zero magnetic field, the four states are split into two Kramers pairs by Kane-Mele spin-orbit coupling, ∆ SO , [12][13][14]22]. At finite magnetic field, the states shift linearly in energy according to their spin and valley Zeeman effect ∆E(B ⊥ ) = 1 2 (±g s ± g (1) v )µ B B ⊥ , with the spin g-factor g s = 2, the single-particle (wavefunction dependent) valley g-factor g (1) v and the Bohr magneton µ B [17,[23][24][25]. Note that the valley g-factor is usually around one order of magnitude larger than the spin g-factor. Measurements confirming the well-understood single-electron spectrum in our single QD are shown in Supplementary Fig. S2. The 16 two-particle states are shown in Fig. 1(a), which, at B ⊥ = 0, group into the orbital sym- 1. (a, b) Single-, N = 1, and two-particle, N = 2, energy spectrum of a BLG QD in an out-of-plane magnetic field. The two-particle states are split into two multiplets with (anti-) symmetric orbital wavefunction, (φa) φs, consisting of (10) 6 states. They are separated by an orbital splitting, ∆ Orb , while the symmetric multiplet is split again, denoted with δ1,2. The valley g-factor is different for single-particle, symmetric and anti-symmetric orbitals. (c) Schematic of the device with BLG encapsulated in hexagonal boron nitride and multiple gate layers that create the QD via soft-confinement. (d) Differential conductance dI/dV b with respect to applied bias and FG voltage, showing diamond shaped regions of Coulomb blockade. Finite bias measurements are performed along the yellow arrow to inspect the excited state spectrum in the single and two-particle regime. White arrows highlight excited states of the two-particle spectrum. metric (φ s ) and anti-symmetric (φ a ) multiplets separated by the energy ∆ Orb . For both the valley and spin wavefunction components, there are three symmetric and one anti-symmetric two-particle states, namely the triplets \|T s,v 0± and the singlet \|S s,v , where s and v refer to spin or valley. There are 6 (10) anti-symmetric (symmetric) combinations of the spin and valley components, which need to be combined with an (anti-)symmetric orbital component for the total wavefunction to remain antisymmetric [26]. The orbital energy of the two-particle states comprises the occupied single-particle orbitals' energies and Coulomb interactions between the two particles, which lead to a splitting of symmetric and antisymmetric orbitals by ∆ Orb , depending on the size of the QD, surrounding screening and applied displacement field [17]. Additionally, the orbitally symmetric states are affected by short-range Coulomb interactions inherent to BLG [19][20][21], since their orbital wavefunctions have nonzero density at the relevant short inter-particle distances. BLG exhibits local density fluctuations which are determined by the lattice symmetry. Mutual interactions between these fluctuations induce the formation of states with spontaneously broken symmetries in sublattice and valley space, introducing splittings δ 1,2 proportional to the strength of the corresponding short-range interactions [18]. The energy of the spin-and valley-dependent part is determined by the coupling to the magnetic field and the spin-orbit coupling. Kane-Mele spin-orbit coupling induces opposite spin splittings in the two valleys, which only affects states in the anti-symmetric orbital which are both valley and spin polarized. A perpendicular magnetic field couples to spin-and/or valley polarized states \|T v,s 0± and shifts their energies according to their corresponding g-factors. Note, that the two-particle valley g-factors (g s v , g a v ) are in general also different from the single-particle g-factor (g (1) v ) [27]. Fig. 1(c) shows a schematic of our QD device, which consists of a BLG flake encapsulated in hexagonal boron nitride (hBN). This heterostructure is placed on a graphite back gate (BG) and has two Cr/Au gate layers evaporated on top, a set of split gates (SG) and a finger gate (FG). To form a QD, we follow previous works [8-11, 13, 24, 28, 29]: A narrow p-doped channel is created utilizing SG and BG. The FG locally overcompensates the potential applied by the BG to form an n-type QD. It is separated from source and drain by two tunneling barriers, where the Fermi energy resides within the band gap. With a simple plate capacitor model [30], we estimate the radius of our QD to be ≈ 80 nm, which is compatible with the dimensions of the gate layout. Fig. 1(d) shows the differential conductance, dI/dV b , through the QD as a function of the bias, V b , and FG voltage, V G , for an electron occupation, N , between zero and two. Within the Coulomb diamonds, N is fixed and transport is suppressed by Coulomb blockade. The outline of the conducting region between N = 1 and N = 2 is defined by the ground state (GS) transition entering and leaving the bias window. Here, 'GS transition' refers to the QD being in its GS before and after the tunneling of the second electron onto the QD. Additional transitions involving excited states (ES) of the N = 1 and/or N = 2 spectrum are possible within that region, as highlighted by the white arrows. Apart from QD transitions, we also observe additional features in the differential conductance in Fig. 1(d) [31], which most likely originate from density of states effects in the leads [32, 33] (see white asterisk) and have previously been observed in similar devices [24]. To experimentally investigate the two-particle spectrum, we measure line cuts along the dashed arrow in Fig. 1(d) and inspect the magnetic field dependence of the GS and ES transitions. In Fig. 2(a), we plot the transconductance, dI/dV G , at V b = 3 mV as function of V G and B ⊥ . The gate voltage along the dashed arrow Fig. 1(b). Transitions to states with symmetric and anti-symmetric orbital wavefunction with different g-factors give rise to a rich spectrum. The inset shows a zoom-in for low magnetic field with reversed bias, where the multiplet splitting is visible. Prominent features are highlighted by dashed lines, obtained from theoretically predicted features in (b): Calculated differential conductance map as in (a), assuming g , g a v = 43, ∆SO = 80 µeV, δ1 = 66 µeV, δ2 = 306 µeV and an asymmetry of the tunnel barriers to source t S and drain t D of t S /t D = 0.7. (c,d) Transition scheme from one to two electrons in the QD at zero (c) and finite (d) B ⊥ , with spin-orbit coupling and spin Zeeman splittings omitted for simplicity. The length of the transition arrows corresponds to the chemical potential necessary to enter the second electron in the QD, which also corresponds to the y-axis in panel (a) and (b). in Fig. 1(d) was converted to electro-chemical potential (displayed on the right axis of Fig. 2(b)) and the raw data was corrected for magnetic field dependent oscillations in the lever-arm, which are due to Shubnikov-de-Haas oscillations in the lead region [25]. With increasing V G , first the GS transition enters the bias window, then additional transitions follow, each appearing as feature of increased differential transconductance, giving rise to a rich spectrum. Eventually, the GS transition leaves the bias window, which Coulomb blockades the QD and appears as a feature of strong negative transconductance (white dashed line). The inset shows the reversed bias direction for low magnetic fields [34]. From the linewidth of the features, we may estimate a lower bound for possible spin-and/or valley mixing of < 100 µeV, in good agreement with recent measurements on BLG double QDs [13]. We reproduce the features of Fig. 2(a) in theoretical calculations, which are shown in Fig. 2(b). Tunnelling transport through the QD is described by solving the rate equations for the single-and two-particle states presented in Fig. 1(a) and (b) [35]. We obtain the tunnel rates for a single-electron sequential tunnelling process to first order in the tunnelling Hamiltonian and by applying Fermi's golden rule. Using the tunnel rates we compute the occupation probabilities for the different QD states in the stationary limit and the resulting sequential tunnelling current. Employing this procedure and adjusting the free parameters g (1) v , g s,a v , δ 1,2 , t S,D and ∆ SO , we obtain differential transconductance maps [18] as in Fig. 2(b). Now, we assign the most prominent features in Figs. 2(a) and (b) to their corresponding transitions from single-to two-particle states. Fig. 2(c) shows all possible transitions at zero magnetic field, where the spinorbit splitting was neglected for simplicity. The length of each transition arrow directly corresponds to the chemical potential required for that transition. Identifying transitions with the features in Fig. 2(a) allows to extract the involved energy scales: The orbital splitting is ∆ Orb ≈ 700 µeV, with the symmetric orbital providing the GS transition (5,6) and the anti-symmetric orbital appearing at position (I) with transitions 1,3,8 and 10. From the inset in Fig. 2(a), we can extract δ 2 ≈ 350 µeV, while δ 1 ≈ 0 within the measurement resolution. Fig. 2(d) shows the level scheme at finite B ⊥ . We can identify three different slopes that match neatly with the data. The first ones are caused by transitions where only the single-particle states shift with g (1) v and the two-particle states remain constant (3)(4)(5)(6)(7)(8), while the v , as a function of applied perpendicular displacement field, evaluated from transitions to orbitally symmetric (6) or anti-symmetric (8) two-particle states with vanishing valley magnetic moment (labeling as in Fig. 2(c)). (b) Two-particle g-factors, g s,a v , evaluated from the transitions to valley triplet states in the (anti-)symmetric orbital. Symmetric and anti-symmetric orbitals exhibit an average g-factor difference of ≈ 8. (c) Energy splitting between symmetric and anti-symmetric orbital, ∆ Orb , and multiplet splitting, δ2, in the symmetrical orbital as a function of displacement field. other two are due to transitions from the single-particle states to the valley polarized triplets in either the symmetric (2,9) or anti-symmetric (1,10) orbital [36]. The change in chemical potential for each transition is given by respectively. Taking these different slopes into account, it can be understood how the GS transition evolves with magnetic field. Transitions 5 and 6 are the GS transition for zero magnetic field. For increasing positive magnetic field, transition 6 requires more chemical potential, while transition 5 needs less. Still, transition 6 remains the GS transition (which defines the Coulomb blockaded region), since \|K − is the single-particle GS. Instead, transition 5 becomes a 'negative' excited state, which manifests as a decrease in transconductance (see arrow 5 in Fig. 2(a)). At position (II), transition 9 becomes the GS transition, when \|S s T v − becomes the two-particle GS. This reveals that the symmetric orbital is lower in energy than the anti-symmetric one, since only the multiplet splitting of the symmetric orbital gives rise to this change in the GS transition. When further increasing the magnetic field, transition 10 eventually becomes the GS transition (see position (III)), as soon as \|T s 0± T v − is lower in energy than In order to better understand the energy scales involved in the two-particle spectrum, the measurement of Fig. 2(a) is performed for different displacement fields, D, applied to the BLG, which changes the band structure and also the confinement potential of the QD [17]. For each displacement field, we evaluated ∆ Orb , δ 2 , and the valley g-factors of the most distinct transitions. For transitions 3 -8 in Fig. 2, the slope only arises from the single-particle states, since the two-particle states do not shift with magnetic field. Fig. 3(a) shows that the single-particle g-factor evaluated from transitions 6 and 8 are the same within the error margins, even though they target states in two different orbitals. There is a decrease of g (1) v for higher displacement fields, which was also observed earlier [11]. This decrease is also visible in Fig. 3(b), where the g-factors of the symmetric and antisymmetric orbitals g s,a v , are evaluated from transitions 9 and 10. They show an average difference of g a v − g s v ≈ 8, confirming that the symmetric and anti-symmetric orbital comprise of different single-particle orbital states. The decrease of all g-factors in Fig. 3 indicates a shift in the wavefunctions' composition in k-space. This conclusion is supported by the fact that both the orbital splitting as well as the short range interaction contribution δ 2 show the same trend in Fig. 3(c), since they also scale with the shape of the wavefunction [17]. The orbital splitting, ∆ Orb , decreases from ≈ 0.8 to 0.4 meV for increasing displacement fields, while δ 2 decreases from ≈ 0.5 to 0.15 meV. For all displacement fields, δ 1 ≈ 0 within the measurement resolution. In summary, we have experimentally observed both orbitally symmetric and anti-symmetric two-particle states in a BLG QD using finite bias tunneling spectroscopy. We identified that the 16 possible states are split into orbitally symmetric and anti-symmetric two-particle states separated by ∆ Orb ≈ 0.4 − 0.8 meV. The orbitally symmetric multiplet is further split with δ 2 ≈ 0.15−0.5 meV by lattice scale interactions which are equally related to inter-valley scattering and "currentcurrent" interactions. We find that the two-particle ground state is a spin-triplet at B = 0 but can be tuned to be a spin-singlet for finite out-of-plane magnetic field. This is in contrast to semiconductor QDs, where the spin-singlet is the two-particle ground state for common magnetic field strengths and is utilized for qubit read-out in double QDs via Pauli spin blockade. Understanding the two-particle spectrum in BLG QDs is thus an essential step for further investigating the influence of the valley on Pauli blockade in double QDs and eventually for identifying a suitable regime for qubit operations. valley g-factors would be the same for the symmetric twoparticle and single-particle wavefunctions. Conversely, a differing g-factor indicates that also higher energy singleparticle orbital states are included in φs. [36] \|K + is an ES of the QD at N = 1 and finite B ⊥ , which makes its occupations probability smaller than \|K − . Thus, most transitions originating from \|K + are not clearly visible in the data, as their contribution to the transport current is much smaller. [37] We cannot determine the sign of the coupling constants from comparison to the transport calculations. Changing the sign of g ⊥ , gz0, g0z reverses the order of states in the multiplets shown in Figs. 2(c,d) but leaves the pattern intact and hence leads to the same differential conductance maps as in Figs. 2(b). I. DETAILS ON CALCULATIONS We outline the structure of the bilayer graphene quantum dot's single and two-particle states that we have identified in the experiment. We characterise a one-electron state in the n-th dot orbital with the spin (σ =↑, ↓) and valley (τ = +, −) degree of freedom and write \|n, σ, τ = d † nστ \|0 , where d † nστ is the electron creation operator and \|0 is the empty dot state. The single-particle states' energy in a perpendicular magnetic field, B ⊥ , is Here, E n is the energy of the n-th orbital dot state at zero magnetic field, ∆ SO is the Kane-Mele spin-orbit splitting (enhanced by zero-point vibrations [39]), g s = 2 is the free electron spin g-factor, g v is the valley g-factor (a consequence of gapped bilayer graphene's nontrivial Bloch band Berry curvature entailing an topological orbital magnetic moment with opposite sign in the two different valleys [40][41][42][43], and µ B is the Bohr magneton. We account for the finger gate with capacitance C G in the experimental setup, which, at gate voltage V G , changes the dot's electrostatic potential by, with N being the dot occupation number and C the total capacitance of the dot. In the experiment, we identify the lowest energy two-particle state to be orbitally symmetric (φ s ), while the first two-particle state features an antisymmetric (φ a ) orbital wave function. There are six combinations of spin (s) and valley (v) -singlet (S) and -triplet (T) states and an orbitally symmetric two-particle state of orbital n: The energies of this two-particle ground state multiplet in a finite perpendicular magnetic field B ⊥ are [17], Here, E s nn , comprises the energy of the n-th single-particle orbital and the screened longrange electron-electron Coulomb interaction. Since any symmetric orbital wave function has finite density at zero or small inter-particle real space distances, these states are further affected by short-range (lattice-constant-scale) electron-electron interactions [19,20]. The factor, (for all combinations of τ i corresponding to inter-and intra-valley scattering channels), is computed from the wave functions, ψ B nστ , on the occupied bilayer graphene sublattice B (a finite diplacement field forces layer polarization), and hence depends on the experimental dot state characteristics, i.e., dot shape, gap, and mode number. The valley g-factor, g s v , of the two-particle states is given by the sum of the single-particle g-factors in the two valleys. For the first excited two-particle dot state that we observe in experiment, the ten possible two-particle states with orbitally antisymmetric (a) wave function are, Hence the excited two-particle state energies are given by, Here, E a nm is the energy of the orbitally antisymmetric states of two screened interacting electrons in single-particle orbitals n and m (akin to the orbitally symmetric state described above), and g a v is the valley g-factor of the two-particle multiplet. We model the leads in the experiment as single channels and a corresponding lead-dot tunnelling Hamiltonian, where c † lkστ creates a lead electron with momentum k, energy l k , spin σ, and valley quantum number τ ), and t L (t R ) is the tunnelling amplitude for the left (right) lead. The experimentally observed transport features can be described by single-electron sequential tunnelling. We obtain the corresponding tunnelling rates by expanding to first order in the tunnelling Hamiltonian: With f being the Fermi function and µ l the electro-chemical potential of lead l. We consider the case of symmetric bias, µ L/R = ±eV B , with respect to the equilibrium electro-chemical potential. Depending on the dot occupation number (N = 1 or N = 2) χ indexes the states from the single-particle or two-particle multiplets, respectively. We solve the master equation of the probabilities, P N:χ , for the state, \|N : χ , to be occupied at a given time,Ṗ N:χ = N :χ (W N:χ←N :χ P N :χ − W N :χ ←N:χ P N:χ ), (11) in the stationary limit,Ṗ N:χ = 0, using the normalization condition N:χ P N:χ = 1. From the probabilities we compute the total particle current flowing from the dot to lead l, and the differential conductance, dI/dV G . II. ADDITIONAL SIMULATED DATA We compute the charge current across the dot, Equation 12, and the differential conductance for various system parameters of the dot and the leads. We take values for the spin-and valley g-factors, the orbital energies, and the short-range interaction coupling parameters from the experiment, as explained in the main text. Figure S1 shows calculated differential conductance maps for different values of tunnelling amplitudes to the left (source contact in our convention) and right (drain) contacts. = 65 ± 10 µeV < 100 µeV (1) g = 33 ± 2 (1) S2. (a, b) Differential transconductance dI/dV G as a function of V G (transformed to electrochemical potential) and perpendicular magnetic field, measured across the N = 1 Coulomb peak at V b = 3 mV similar to Fig. 2(a). For the first electron, the electro-chemical potential of the transition directly corresponds to the energy of the single particle state occupied, which confirms that the single particle spectrum presented in Fig. 1(c) is indeed observed. (a) shows data from the device presented in the main manuscript, (b) shows data from a different device where the spin-orbit splitting could be resolved. In this section we provide additional experimental data. In Fig. S2 we show differential transconductance (dI/dV G ) measurements as a function of the chemical potential (obtained from V G and the experimentally extracted gate lever-arm) and B ⊥ for two different BLG QD devices. The data are recorded across the N = 1 Coulomb peak at V b = 3 mV (see also Fig. 1(b) in the main manuscript). In Fig. S3 we show the measured raw data of Fig. 2(a) in the main manuscript, where the Shubnikov-de-Haas oscillations at high B ⊥ are well visible. Fig. 2(a). Differential transconductance dI/dV G as a function of FG voltage and perpendicular magnetic field, measured across the N = 2 Coulomb peak at V b = 3 mV. Shubnikov-de-Haas oscillations of the density of states in source and drain lead to periodic changes of the lever-arm, which oscillates with 1/B [25]. every single data point. Then, we measure across the second Coulomb peak (N = 2) at finite bias (V b = 3 mV) as a function of out-of-plane magnetic field to obtain data sets like Fig. S4(a)-(c). Next, we determine the lever-arm by the outline of the conducting region, which corresponds to ∆µ = eV b (see white lines in Fig. S4(a)). Finally, we obtain the various g-factors from the different slopes (see dashed lines in Fig. S4(c)) and we extract ∆ Orb as well as δ 2 as shown by the black arrows in Fig. S4(c). correspond to the applied bias via eV b = ∆µ, which allows to compute the lever-arm of the FG. The dashed lines and arrows in (c) illustrate the evaluation of the energy scales and g-factors. The slight trend of larger ∆ Orb and δ 2 for smaller displacement fields is nicely visible.	2021-06-17T01:15:51.806Z	2021-06-15T00:00:00.000	{ "year": 2021, "sha1": "538cbb276081855517cc3b1b60c3d68ed97c7030", "oa_license": null, "oa_url": "http://arxiv.org/pdf/2106.08405", "oa_status": "GREEN", "pdf_src": "Arxiv", "pdf_hash": "538cbb276081855517cc3b1b60c3d68ed97c7030", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics", "Medicine" ] }
268521903	pes2o/s2orc	v3-fos-license	Prospective cold metal working and analysis of deformation susceptibility of CuMg alloys with high magnesium content Metal alloys designated for cold metal working exhibit much higher strength properties than pure materials due to solid-solution hardening. However, with the increase of mechanical properties its plasticity and workability decreases. Constant development and demand in this area has led to research on many copper alloys, such as copper alloys with high content of magnesium which were never tested before. The limitations regarding cold metal working of CuMg alloys is the main objective of this paper. Here we show that the tested materials exhibit much higher mechanical properties than currently used as electric conductors and carrying-conducting equipment materials such as pure copper, aluminum, M63 brass or CuNiSi alloy. The results were obtained using Hollomon relation, Considére criterion, Gubkin method and hardness measurements. It lead to assessing the prospective cold metal working of CuMg alloys with 2 wt% of magnesium up to 4 wt% of magnesium. The test range included upsetting with 10–50% of cold deformation. It provided the results on evolution of mechanical properties and deformability of tested alloys. Additional information was provided based on the alloys subjected to 50% of strain. The results have proven that as the amount of magnesium increased so did the assessed values, however, it was also linked with increasing friction coefficient. Measured hardness was 2 times higher and calculated Ultimate Tensile Strength (UTS) was even 2.5 times higher in reference to pure copper in the as-cast state. However, with magnesium content at 3.6 wt% or higher, the elevated amount of α + β phase causes brittleness making it impossible to subject these materials to cold metal working processes. We anticipate our assay to be a starting point for more sophisticated models and experimental research concerning cold metal working processes of CuMg alloys of high-strength, which may lead to developing novel and promising set of alloys. In order to significantly lower the manufacturing cost cheaper alloys such as CuMg have been introduced.As for now only materials with maximum of 0.7 wt% of magnesium are being produced on a large industrial scale and are mainly used for overhead contact wires in railway systems and for overhead power transmission cables 21,22 .Among the researchers from around the world these materials are mainly discussed regarding the materials with low magnesium content as well (not exceeding 2 wt%) [23][24][25] .However, published papers mostly discuss proof strength and electrical properties of alloys subjected to heat treatment.In recent times, a few research papers have been published regarding copper alloys with higher amount of magnesium.The authors of these papers claim that with proper heat treatment it is possible to obtain Yield Strength of over 1000 MPa 26 and that two-phase materials may maintain electrical conductivity of approximately 50 IACS % (International Annealed Copper Standard) 27 .Based on these works it can be stated that metallurgical synthesis and heat treatment of these alloys is possible and suggest that CuMg alloys may be subjected to precipitation hardening while maintaining a satisfactory level of electrical conductivity.Nevertheless, for the copper alloys with high magnesium content to be a prospective substitute material for other, more expensive alloys it is necessary to confirm their susceptibility to cold metal working, which will further increase their mechanical properties due to strain hardening.Despite copper magnesium alloys being promising materials the lack of information regarding their metal working is a huge knowledge gap which needs to be clarified, thus, making this research essential for further development of materials science considering copper alloys. The aim of this paper is to provide information on the UTS, strength coefficient, strain exponent and friction coefficient of CuMg alloys.The results were obtained in upsetting tests which is an absolute novel approach in terms of research regarding CuMg alloys and will determine whether the materials may be subjected to metal forming processes.This type of procedure was chosen as all necessary results could be obtained from one test stand and small samples which tensile testing could not provide.What is more the materials were in the as-cast state and subjecting them to tensile testing without prior homogenization might result in false conclusions.Additionally, hardness of tested materials was evaluated to assess how much the alloys strengthen with applied strain.Obtained results were confronted with widely used materials for electrical purposes such as pure aluminum (EN AW 1050), CuZn alloy (M63 brass) and CuNiSi alloy which were obtained and subjected to analogical tests as CuMg alloys in question.What is more the employed methods were fully explained and along with the obtained results may function as a guideline for other works regarding deformability of metals and alloys. Experimental procedure The crucial feature of metals and alloys, especially designed for conductive applications, is among others their susceptibility to metal working processes.This is mainly caused by the fact that conducting materials originate mainly from continuous or semi-continuous casting processes and further extrusion, forging or drawing.The tested copper magnesium alloys with the magnesium content ranging from 2 to 4 wt% have been obtained using laboratory horizontal continuous casting stand.The metallurgical synthesis was conducted at 1473 K with casting velocity of approximately 180 mm/min (2 s of standstill for each 6 mm of pull).The casted rod was axi-symmetrical with a diameter of 14 mm.The furnace consisted of graphite crucible and graphite crystallizer.Primary cooling system was made of copper with closed circulation of cooling medium provided at the constant velocity of 0.5 l/min.Further secondary cooling was supplied directly onto the cast rod's surface.Additionally, the reference material of pure copper was obtained with the same casting parameters.Materials in the as-cast state were subjected to chemical composition analysis using arc spark spectrometer Spectrotest and Scanning Electron Microscopy (SEM) observations using Hitachi SU-70 microscope.The main part of the proposed research analysis consisted of subjecting the obtained materials to upsetting tests in ambient temperature with the deformation of 10%, 20%, 30%, 40% and 50%.The samples were cylindrical with diameter of 10 mm and height of 15 mm.The longer dimension was parallel to the casting direction.The upsetting tests were conducted using Zwick/ Roell Z100 testing machine.The force load accuracy of the machine is 0.5% and the velocity was 1 mm/min.The tests were carried out with petroleum-based mineral oil as lubrication.The strain hardening was expressed as an average value of 5 hardness measurements obtained using Brinell's method with Nexus 3001 tester, which test force accuracy is 0.5% and display resolution is 0.1 HB.Upsetting stress during the cold deformation process was also analyzed.Additionally, the samples subjected to 50% of strain were further analyzed in order to calculate: • The friction coefficient employing the Gubkin's method as presented at Fig. 1 and Eq.(1) 28 , • UTS using Considére criterion 29,30 , which allows its assessment as the value projected at the Y axis from the tangent (red line) drawn from the negative one value to the true stress/strain upsetting curve as presented at Fig. 2C, • Strain hardening exponent and strength coefficient as proposed by Hollomon 31 which was collectively pre- sented at Fig. 2 on the exemplary upsetting stress/strain relation and Eq. ( 2). The microstructure of the samples subjected to 50% of strain was also analyzed in order to determine visible cracks.Electrical conductivity was measured using the eddy current device which determines the electrical conductivity of nonferrous metals expressed in MS/m.The tests were conducted on all samples after placing them for 24 h at ambient temperature in order to stabilize their thermal state which prevents the influence of temperature on measurements.10 measurements were performed on each sample with a frequency of 60 kHz, which provides maximum depth of signal penetration.The accuracy of the equipment is 0.5% of measured values.The average values were calculated and presented as final results, where µ is the friction coefficient, θ is the dimensional barreling ratio equal to (d 2 -d 1 )/d 2 , ε is the engineering strain and d 0 , d 1 , d 2 , h 0 and h 1 as marked at Fig. 1, (2) www.nature.com/scientificreports/where σ s is the deformation resistance, K is the strength coefficient, ε t is the true strain and n is the strain hardening exponent. Results and discussion The obtained results made it possible to determine the cold metal working potential of newly developed alloys which is to be presented and discussed in the following part of this work.Since magnesium boiling temperature (1364 K) is nearly identical to the melting point of copper (1357 K) it was fundamental to determine whether the obtained alloys have the assumed chemical composition as magnesium tends to evaporate easily during the metallurgical synthesis.The chemical compositions were selected to provide 3 single-phase (Cu2Mg, Cu2.4Mg and Cu2.8Mg) and 3 two-phase materials (Cu3.2Mg,Cu3.6Mg, Cu4Mg).Additional alloy with 3 wt% of magnesium (the amount where at 999 K α + β phase should occur according to 32 ) and pure copper as reference material were produced.Thus, 8 materials were obtained in the continuous casting process.Each material was subjected to chemical composition analysis, which results were presented in Table 1.Providing, that the chemical composition does not vary more than 0.05 wt% (500 ppm) from the nominal values it would appear that the alloys were obtained with the amount of magnesium as intended.The amount of impurities which were present at the cast material did not exceed 50 ppm per individual element.SEM observations presented in Fig. 3 prove that as the amount of magnesium increased (Table 1) the amount of α + β-phase (dark area) 27 increased as well.It is evenly distributed among the volume of all of the materials.Using ImageJ software the amount of α + β-phase was estimated and the percentile results were placed in Table 1.The data received show that the amount of α + β-phase quadruples as the amount of magnesium increases from 2 to 4 wt%.Rapid increase starts when the amount of magnesium is higher than 3 wt%, i.e. when the alloys become two-phase materials, according to CuMg phase diagram 32 .The magnesium is not present in pure copper and it was confirmed both in chemical composition and microstructure analysis.Additional SEM observations were conducted regarding the materials subjected to 50% of strain which was presented at Fig. 4 and as presented the α + β-phase is less random and more aligned than in the as-cast state due to the applied force in a set direction. Alloys with 2 and 4 wt% of magnesium were presented with higher magnification at Fig. 5 in order to better display the cracks resulting from compression force at a microscale.There are less than 10 cracks across Cu2Mg sample and around 40 across Cu4Mg sample.Cracks are mostly vertical, parallel to the direction of applied force and do not appear in areas where there is no α + β-phase.This would suggest that excessive amount of magnesium causes brittleness during cold metal working. The samples were subjected to 5 different deformation values as discussed above.An exemplary photograph of one set of samples after upsetting tests with a starting sample on the left were presented at Fig. 6 along with the strain values.The picture shows how much the samples deformed during the tests. Since pure copper does not contain any additives it should exhibit significantly lower mechanical properties than the tested alloys.Bearing that in mind the values obtained from the upsetting tests and hardness measurements of pure copper were presented separately in Table 2.The results were presented for each of the applied upsetting strains and both values increased quasi-linearly.Since pure copper strengthens when being subjected to strain the increase in measured values of hardness was expected.This is a result of strain hardening which exhibits during metal working conducted below the temperature of recrystallization of materials 33 . As there was much more data regarding the tested alloys than in the case of pure copper both maximum recorded stress in upsetting tests and hardness values were presented in the form of 3-axis graphs.Maximum upsetting stress increased significantly for each material as the strain increased with the values exceeding 1000 MPa at 50% of applied cold deformation as presented at Fig. 7.The reference value of pure copper was only 524 MPa meaning that the tested alloys needed two times more force to obtain the same height reduction during compression.Remarkable was the fact that regarding single-phase materials (up to 3 wt% of magnesium) when the strain increased from 10 to 50% the recorded stress increased 3 times, whereas in the case of two-phase alloys (above 3 wt% of magnesium) it was only 2 times.However, it was notable that when the cold deformation of 50% was applied there was no difference in recorded values in the case of two-phase materials.This might mean that the amount of α + β-phase and discussed cracks is so high across the volume of the material that it could lower the stress needed to obtain the 50% height reduction.Even though, the cracks are not visible with naked eye they lower the integrity of the material. The average values of hardness of samples after upsetting with given strains measured with Brinell's method were complemented with hardness values of non-deformed material as presented at Fig. 8.The obtained results proved that there is a significant solid-solution strengthening as the hardness values of the base material increased by 80-150% (depending on the magnesium content) in reference to pure copper.The standard deviation of the obtained results was between 0.832 and 1.871 for materials in the as-cast state and up to 20% of applied strain.At higher amounts of deformation the standard deviation has risen to the range between 2.012 and 3.143.Further increase of the measured values as the deformation was applied proved that as typical copper alloys the newly developed materials may be subjected to strain hardening and the hardness increases quasi-linearly regardless of the magnesium content.Further analysis of the stress/strain relations was conducted on the compression curves obtained from the tests with 50% of applied strain, as there was the most data to analyze and the conducted calculations would reflect the actual state of material.The data obtained from the analysis as exemplary presented at Fig. 2 and collectively at Fig. 9 was presented in Table 3.For comparison, other common materials used as electric conductors and carrying-conducting equipment such as M63 brass, CuNiSi alloy and aluminum (EN AW 1050) also in the as-cast state were subjected to upsetting tests with 50% of strain.Just by analyzing the data presented at Fig. 9 it is visible that regarding Cu3.6Mg and Cu4Mg alloys (marked with the red color at the graph), when the maximum true stress is achieved its value decreases over time.Such decrease was not recorded in terms of CuMg alloys with magnesium content of up to 3.2 wt% (marked with the green color).This data correlates well with the results presented at Fig. 7 where at 50% of applied strain the maximum stress values did not increase at higher magnesium content.This may be caused as mentioned above by excess amount of α + β-phase which makes the material too brittle to sustain cold deformation of such high strain.Materials of lower magnesium content more or less remain at a constant level of true stress when reaching its maximum value.Regarding the reference materials which are marked with the black color at the graph it is visible that all materials exhibit much lower levels of true stress at 50% of applied strain than the tested CuMg alloys.www.nature.com/scientificreports/ The assessed value of UTS of pure copper in the as-cast state was at the level of experimentally obtained values in 33,34 .Similarly in terms of K and n factors which regarding pure copper were close to values obtained by Bowen et al. 35 and El-Domaity et al. 36 .Meaning that chosen methods for assessing these parameters were correct and should provide proper results for other materials.Regarding CuMg alloys the assessed values using Considére criterion proved that adding just 2 wt% magnesium copper might increase this value almost two times, and by further increasing the magnesium content the UTS value would increase as well.The obtained results correspond very well with the hardness measurements presented at Fig. 8 and might suggest that the assumptions are correct.Regarding the values calculated using Hollomon relation (K-strength coefficient; n-strain hardening exponent) it must be noted that as the magnesium content increased the n-value decreased.It means, that formability of these alloys lowers and if given enough strain pure copper might obtain the same mechanical properties (if technologically it would be possible to do so without interoperation annealing).However, calculated K-value shows that if the applied true strain would be equal to 1 the true strength of tested alloys would be approximately 65-115% higher than this of pure copper.It proves yet again that the tested CuMg alloys with high magnesium content despite relatively lower formability would surpass pure copper in terms of mechanical properties.The calculated R 2 -value (coefficient of determination) is at relatively high level of above 0.9 (the range is between 0 and 1) meaning that the assumptions made with the trend lines predict the actual outcome quite well.However, with increased mechanical properties come disadvantages such as increased friction coefficient.That would mean that for instance in the drawing process, according to Avitzur's analysis of the upper bound method 37 the drawing force and thus the drawing stress would increase significantly.As copper and its alloys are mostly used as conductors of electrical energy it was necessary to measure the electrical conductivity values (ρ) and calculate the IACS %.The results presented in Table 3 show that as the amount of alloy additive increases, the electrical conductivity decreases in regard to pure copper.The standard deviation of the obtained results of electrical conductivity in MS/m was approximately 0.097 for copper, 0.083 for aluminum and between 0.035 and 0.073 for copper alloys.Regarding other commonly used copper alloys it was proven not only that they have lower electrical conductivity but also significantly lower calculated UTS, and K-value than CuMg alloys.It suggests that CuMg alloys, regardless of the magnesium content (among the tested range) are prospectively much better materials than currently used alloys.As for pure aluminum, the electrical conductivity is higher than the tested copper alloys, but its calculated mechanical properties are even 4-7 times lower than those of tested CuMg alloys.Thus, when mechanical properties may be neglected and the materials shall function only as a conductor then pure aluminum and pure copper are sufficient.However, when the final elements are carrying-conducting equipment, then CuMg alloys are superior to currently used copper alloys and pure materials. The most common elements which require satisfactory levels of mechanical and electrical properties are among others overhead contact lines, railway contact wires, cable terminals and connectors, brackets and handles for contact wires or railway catenary systems.These are all products of wire drawing or forging processes, where the batch material may be the outcome of casting processes or casting + rolling/extrusion 38 .As of March of 2024 the market price of magnesium is around 3450 $/mt.At the same time the prices of currently used alloy additives in materials designed for the above-mentioned applications are: around 17,500 $/mt for nickel (CuNiSi alloys), around 2650 $/mt for zinc (brass) and around 28,200 $/mt for tin (bronze) 39 .This shows that only the price of zinc is noticeably lower, while nickel and tin are significantly more valuable.However, there are 2 problems with zinc as alloy additive.One is that the properties of brass regarding UTS and electrical conductivity are approximately two times lower than tested CuMg alloys as discussed above.Secondly, most brass materials contain lead in its chemical composition (some of the grades even up to 2 wt%).And as it is now generally known lead is toxic and manufacturers try to avoid it as much as possible.Last factor that needs to be discussed is the weight of the final product.The density of magnesium (1.738 g/cm 3 ) is much lower than nickel (8.908 g/ cm 3 ), zinc (7.14 g/cm 3 ) or tin (7.31 g/cm 3 ).Hence, the final products will be slightly lighter, depending on the amount of magnesium used 40 . Presented research results were obtained in laboratory conditions for the materials in the as-cast state.Since copper alloys tend to respond significantly to heat treatment one must assume that the results would vary when the homogenization, supersaturation or artificial aging would be applied. Conclusions Taking all of the above into account the following conclusions were made: 1. Newly developed copper alloys with high amount of magnesium may be obtained in the continuous casting process and might be subjected to cold metal working.These factors make its prospective applications versatile as both solid-solution strengthening and strain hardening occurred when the alloys were subjected to laboratory tests in ambient temperature with no cracks visible with naked eye.However, in terms of materials with 3.6 wt% of magnesium content and higher it was noted that the amount of α + β-phase might be too high, causing brittleness and thus making it impossible to subject these materials to cold metal working processes.This may of course change if the materials would be subjected to supersaturation before applying cold deformation.2. Maximum recorded upsetting force was 2-3 times higher than in the case of pure copper depending on the magnesium content, however, two-phase materials at 50% of applied strain did not exhibit increase in recorded force.It might be the result of brittleness coming from excessive amount of α + β-phase.3. Regardless of the assessing method the mechanical properties of the newly developed CuMg alloys increased as the applied strain or magnesium content increased and were several times higher than those of pure copper, however, the side effect was the increase of the coefficient of friction. 4. When compared with other commonly used as electric conductors and carrying-conducting equipment materials such as M63 brass, CuNiSi alloy and aluminum (EN AW 1050) in the as-cast state it was proven that currently used copper alloys have both electrical conductivity and mechanical properties lower than those of CuMg alloys, regardless of the magnesium content.Pure aluminum despite having better electrical conductivity has 4-7 times lower mechanical properties than CuMg alloys.5.The tested materials were mostly proven to withstand 50% of cold deformation in the state as such may be used as batch material in metal working processes.Prospectively these materials may substitute currently used copper alloys such as M63 brass or CuNiSi alloy, which properties both mechanical and electrical were proven to be inferior to CuMg alloys. Figure 1 . Figure 1.The shapes and dimensions of the cylindrical sample subjected to upsetting tests (Gubkin's method of assesing the friction coefficient). Figure 3 . Figure 3. Backscatter electron SEM analysis of copper magnesium alloys; in the as-cast state; magnification ×500. Figure 6 . Figure 6.Exemplary photograph of samples subjected to upsetting tests with specified strain values. Figure 7 .Figure 8 .Figure 9 . Figure 7. Maximum recorded upsetting stress for each of the tested alloys and applied strains. Table 1 . Chemical composition analysis of the obtained alloys and estimated amount of α + β-phase at SEM images.Significant values are in [bold].	2024-03-20T06:17:35.194Z	2024-03-18T00:00:00.000	{ "year": 2024, "sha1": "c968d784c13b4ee17665861b5ba4594645460c08", "oa_license": "CCBY", "oa_url": null, "oa_status": null, "pdf_src": "PubMedCentral", "pdf_hash": "b6b8ca581f2c70ea6a1aec2765e9e4f99b5d1625", "s2fieldsofstudy": [ "Materials Science", "Engineering" ], "extfieldsofstudy": [ "Medicine" ] }
261726794	pes2o/s2orc	v3-fos-license	Predictors of Escalation to Intensive Care Unit Level of Care Among Admissions for Alcohol Withdrawal According to the 2019 National Survey on Drug Use and Health, 14.5 million people ages 12 and older had alcohol abuse disorder. Alcohol withdrawal syndrome (AWS) can be defined as a collection of physical symptoms experienced due to abrupt cessation of alcohol after long-term dependence. In instances where regular inpatient management fails to control AWS symptoms, patients are shifted to intensive care units (ICUs) for closer monitoring and prevention of life-threatening complications like withdrawal seizures and delirium tremens (DTs), labeled as severe alcohol withdrawal syndrome (SAWS). Although this represents a significant healthcare burden, minimal studies have been conducted to determine objective predictors. In this study, we aim to determine the effect of patient demographics, socio-economic status, biochemical parameters, and clinical factors on the need for escalation to ICU level of care among admissions for AWS. Our study showed that factors such as a history of DTs or alcohol-related seizures, the initial protocol of management, degree of reported alcohol usage, activation of rapid response teams, mean corpuscular value, alcohol level on admission, highest Clinical Institute Withdrawal Assessment Alcohol Revised (CIWA-Ar) scored during the hospital stay, and the total amount of sedatives used were significantly associated with escalation to ICU level of care. Clinicians must use these objective parameters to identify high-risk patients and intervene early. We encourage further studies to establish a scoring algorithm incorporating biochemical parameters to tailor management algorithms that might better suit high-risk patients. Summary box What is already known on this topic: The current gold standard for the management of alcohol withdrawal syndrome (AWS) is the Clinical Institute Withdrawal Assessment Alcohol revised (CIWA-Ar) scale and treatment with benzodiazepines (BZDs). CIWA-Ar scoring involves discussion with the patient, which negates the objectivity of the symptoms. BZDs are administered either using a symptom triggered regimen (STR) or fixed dose taper regimen (FDTR). STRs are favored as studies have shown STRs to be associated with a shorter length of stay, decreased rates of hospital-acquired infection and lower cumulative dose administration. Despite the wide acceptance and utilization of these management protocols, several patients with severe symptoms of AWS are transferred to the intensive care unit (ICU) for more aggressive care and closer monitoring. Introduction A ccording to the 2019 National Survey on Drug Use and Health, 14.5 million people ages 12 and older had alcohol abuse disorder. Alcohol withdrawal syndrome (AWS) can be defined as a collection of physical symptoms experienced due to abrupt cessation of alcohol after long-term dependence. 1 It is common in patients with alcohol use disorder (AUD), affecting up to 40% of patients being hospitalized. Critically ill patients with AWS are often associated with poor outcomes, such as higher rates of infection and sepsis. 2 Medical consensus has classified the symptoms of withdrawal based on the time of onset or severity. Early withdrawal symptoms (such as diaphoresis, tremor, hyperactivity, insomnia, and headache) can start around 6 h after the last drink and last up to 48 h. Perceptual disturbances such as visual, auditory, and tactile hallucinations can be classified as moderate withdrawal and may last up to 6 days. However, the most severe consequence that we aim to avoid when treating patients for AWS includes withdrawal seizures and delirium tremens (DTs). This last for up to 2 weeks with anticipated mortality of 37% without treatment. [3][4][5] Considering the fatality of AWS, clinical guidelines suggest a standardized instrument to measure the severity of AWS. The center of attention has shifted from the reactive management of AWS to the implementation of different screening and assessment tools. Clinical Institute Withdrawal Assessment Alcohol revised (CIWA-Ar) is a widely used and corroborated 10-item assessment tool developed over the last few decades to objectively quantify the severity of AWS with a score based on history from the patient and pertinent physical exam. In an inpatient setting, the CIWA-Ar score is often ordered preventatively in suspected cases of alcohol withdrawal. 1,6 The 10 items used to monitor What this study adds: Studies evaluating associations between objective parameters such as demographic, clinical, and biochemical parameters and escalation to ICU care in AWS are lacking. Our study contributes in this aspect. We attempt to validate the use of CIWA-Ar score and evaluate the outcomes of STR and FDTR in our institution. We identified that parameters such as a history of delirium tremens or alcohol-related seizures, the initial protocol of management, degree of reported alcohol usage, activation of rapid response teams, mean corpuscular value, alcohol level on admission, single highest CIWA-Ar scored during the hospital stay, and the total amount of sedatives used are significantly associated with escalation to ICU level of care. How this study might affect research, practice, or policy: Objective cut-offs including a blood alcohol level on admission <13 mg/dl and a CIWA-Ar score >16.5 offer a good balance of sensitivity and specificity in reliably predicting patients at a high risk of requiring ICU level of care. Clinicians can use these objective parameters to stratify patients and guide management. Additionally, our results encourage further research to create an objective tool using the parameters explored and design management algorithms that might be better tailored to patients at higher risk. the progression of AWS include agitation, anxiety, auditory disturbances, clouding of the sensorium, headache, nausea/vomiting, paroxysmal sweating, tactile disturbances, tremor, and visual disturbances. 7 CIWA-Ar assigns scores of 0e8, 9e15, and 16 or more to indicate the severity of AWS (mild, moderate, and severe, respectively). A pharmacotherapeutic regimen involving benzodiazepines (BZDs) that is either a symptom-triggered regimen (STR) or a fixed-dose taper regimen (FDTR) is generally used to manage patients. 3,8 The STR is defined as the delivery of adequate pharmacotherapy on an as-needed basis while monitoring patients every 4e6 h based on symptom severity. In comparison, an FDTR involves administering a tapering dose of medication based on a predetermined schedule every 4e6 h. STRs are ordered concurrently with FDTRs to administer extra doses as needed. 9 Previous studies have depicted STR dosing to be superior to FDTR, having a shorter length of stay and decreased rates of hospital-acquired infection in patients receiving a shorter course of BZDs with lower cumulative doses. 10 However, no studies evaluate if STRs compared to FDTRs decrease escalation to ICU level of care. BZDs are regarded as the drug of choice in treating AWS. Diazepam and lorazepam are the most used drugs. However, in different inpatient settings, other agents such as anti-convulsants (such as haloperidol and olanzapine) and adrenergic drugs (dexmedetomidine and clonidine) are also utilized in case of failure of BZDs to control AWS symptoms. The National Institute of Health and the National Institute for Health and Care Excellence guidelines continue to recommend CIWA-Ar as the gold standard tool for the management of AWS, with STR being the superior choice. 3 One prominent drawback of the CIWA-Ar tool is its subjective nature. Scoring involves discussion with the patient, which negates the objectivity of the symptoms. Primary factors contributing to the unreliability of the assessed score include the language barrier, the need for frequent reassessment by different scorers, and altered mentation in patients. 6 In instances where regular inpatient management fails to control AWS symptoms, patients are often shifted to intensive care units (ICUs) for closer monitoring and prevention of life-threatening complications like withdrawal seizures and DTs, labeled as severe alcohol withdrawal syndrome (SAWS). Several single-center studies were done in the past to determine predictors of SAWS. Higher baseline blood pressure, lower potassium level, lower platelet count, higher initial level of alanine aminotransferase (ALT), and high gamma-glutamyl transpeptidase were a few of the significant predictors. 11 Risk factors for developing DTs, the most severe form of AWS included prior history of SAWS, tachycardia, and higher systolic blood pressure. 12 However, few studies have been conducted to further determine other objective parameters to predict admission to the intensive care unit (ICU). In this study, we aim to study the association of a wide range of patient characteristics such as demographics, biochemical parameters, and clinical factors on the need for ICU admission. Additionally, we establish significant cut-off values of clinical parameters such as the admission blood alcohol level, CIWA-Ar scores and the dosage of benzodiazepines administered in predicting ICU admissions. Design and data source In this retrospective cohort study, we included all adult hospitalizations with a discharge diagnosis of alcohol withdrawal, alcohol dependence with withdrawal, or alcohol use taken from the medical records department of Monmouth Medical Center, Long Branch, United States of America. We collected details of all inpatient adult admissions for alcohol withdrawal between January 1, 2021, and December 31, 2021. Patients were searched using the International Classification of Diseases, 10th revision, and Clinical Modification/Procedure Coding System (ICD-10-CM/PCS) codes. We used all subbranches of the ICD-10-CM/PCS codes -F10.1, F10.2, and F10.9 to identify patients who were admitted for alcohol withdrawal and its complications. After individual chart review, patients who were admitted to the inpatient medicine service for the management of alcohol withdrawal without other co-existing illnesses complicating the hospital course were included. Patients who left prematurely against medical advice, had polydrug intoxication, altered mental status attributable to other causes, and those admitted to the ICU directly from the emergency room or admitted to the ICU for reasons other than the management of AWS were excluded. Medical record numbers corresponding to the diagnoses were obtained from the medical records department. Non-identifiable data and details of hospitalization were extracted from the electronic health record system, Cerner, and entered into the REDCap survey software. Any dose of lorazepam in our collected data was converted to diazepam using a 1:5 ratio (1 mg of Lorazepam ¼ 5 mg of Diazepam). Alcohol consumption was recorded as heavy drinking when >14 drinks were consumed per week and as binge drinking when 5 drinks for men or 4 drinks for women were consumed on one occasion as a recurring pattern. This ensured that outcomes were analyzed in a blinded fashion as all patient identifiers were removed. Data on patients' demographics, comorbidities, and biochemical and clinical parameters were collected and compared to look for factors with a positive association with the incidence of ICU admission. Any missing data were filled by using the patient's medical record number to revisit the patient chart and acquire the missing values. Statistical analysis All other benzodiazepines were also converted to diazepam equivalents when needed. Statistical analysis for the data was done using the IBM Statistical Package for the Social Science (SPSS) Statistics for Windows, Version 20.0. Armonk, New York. Numeric variables including age, alcohol level on admission, single highest CIWA-Ar scored by the patient during ward stay, the total amount of sedatives received during medical wards stay, along with biochemical parameters (platelets count, albumin, mean corpuscular volume (MCV), international normalized ratio (INR), alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase (ALP)) were compared between these two groups using a t-test. Categorical data such as gender, race, insurance status, prior ICU admissions for alcohol withdrawal symptoms, history of delirium tremens or alcohol-related seizures, initiation of CIWA-Ar protocol within the first 24 h of admission, the initial protocol of management, the degree of alcohol usage, psychiatric or substance abuse history, and rapid response team activation frequency were also compared using the chi-square test or a Fisher's exact test. Multiple regression analysis was then done on variables with significant association with ICU admission. Additionally, cutoff points with sensitivity and specificity levels were determined using ROC curves for significant variables. A p-value of <0.05 was considered significant in this study. Results A total of 123 patients were included in this study, with a mean age of 50.8 (±12.8) years. 70.7% (87) were males, and 74% (91) were insured. Only 13% 16 had a history of prior ICU admission for alcohol withdrawal symptoms, with an average of one admission per patient. About one-third of the patients (31.7%) had a history of DTs or alcoholrelated seizures. CIWA-Ar evaluation was done for 95.9% (118) within the first 24 h of their presentation. Half of all patients (52.8%) received initial STR with lorazepam. The rest of the patients received either STR with diazepam (18.7%), an FDT of lorazepam (17.9%), or an FDT of diazepam (10.6%). Most of the patients were either heavy drinkers with >14 drinks a week (43.9%) or binge drinkers (25.2%), while only 16.3% were moderate drinkers (7e14 drinks a week), and 14.6% with 1e7 drinks a week as defined by our study protocol. About 60% had a history of liver cirrhosis or steatosis, while 45% had a history of either psychiatric or other substance use disorders. In this cohort, 16 patients (13%) of the total sample were admitted to the ICU during their admission. Most of these patients (62.5%) were admitted to the ICU on either hospital day one or two. The rapid response team (RRT) was activated at least once for 15 patients of the sample. To study the effect of different variables on the risk of ICU admission, the sample was divided into 2 groups; those who were admitted to the ICU, 16 and those who did not need ICU management (107). Table 1 lists the various categorical variables analyzed along with the calculated odds ratio. Table 2 lists the p-values obtained from comparing continuous variables across the two groups with the associated mean values. Receiver operator curves (ROC) were plotted for the following parameters -alcohol level on admission ( Fig. 1), maximum CIWA-Ar score (Fig. 2), and total sedatives (diazepam) received (Fig. 2) to determine cut-offs significant for ICU admission. Table 3 describes cut-off values with associated sensitivity and specificity obtained from the ROCs. Discussion Studies of hospitalized patients have shown the incidence of alcohol withdrawal is between 11% and 32%. 13,14 Among those with alcohol use disorder, 50% of patients have been noted to undergo withdrawal at some point in their lifetime. 15 Patients withdrawing from alcohol are often agitated, significantly burdening physicians, nurses, and other hospital staff. It is when their agitation cannot be managed on the medicine floors that they are transferred to the ICU where they may necessitate the need for a dexmedetomidine drip. Through our study, we hope to identify patients at risk for progression to severe withdrawal and intervene early thereby decreasing ICU burden. Our study found that men were most often the predominant sex for AUD admissions. The mean age was around 50 ± 12.8 years pointing towards a wide population sample. Among the patients who came into our hospital, 16 (13%) were transferred to the ICU. It is commonly observed clinically that it is often the same patients who come back in for treatment of alcohol withdrawal and its complications. Fortunately, however, most patients (95.9%) admitted for alcohol withdrawal were identified early and started on medication. Most patients were started on lorazepam, rather than diazepam at admission. Lorazepam is chosen by most clinicians for its better hepatic tolerability and shorter half-life compared to diazepam. Despite early medication initiation, 13% of patients were admitted to the ICU. Factors significantly associated with escalation to ICU level of care were a history of delirium tremens or alcohol-related seizures, the initial protocol of RESEARCH ARTICLE management, degree of reported alcohol usage, activation of rapid response teams, mean corpuscular value, alcohol level on admission, single highest CIWA-Ar scored during the hospital stay, and the total amount of sedatives used. A history of DTs and alcohol-related seizures is expected to be associated with higher ICU admissions as a prior history increases subsequent risk. However, a history of prior ICU admissions did not carry a similar correlation. This may be because a history of DTs might convey a more severe disease burden. Additionally, patients with a history of prior ICU admissions possibly received more intensive care right from their admission as it might be readily apparent to clinicians based on chart review of their electronic medical records. Among patients that were transferred to the ICU, only 16 patients (13%) had a previous history of ICU admission for alcohol withdrawal whereas 31.7% had a previous history of DTs or alcohol-related seizures. The positive correlation with the activation of rapid response teams and ICU admissions points to an opportunity for early intervention in high-risk patients. Among biochemical parameters, interestingly, a lower blood alcohol level was associated with a higher risk. This might be because a lower alcohol level at presentation would indicate that the patient is further along their abstinence from alcohol and closer to withdrawal. Clinically we administer similar amounts of benzodiazepines to patients irrespective of their admission blood alcohol levels. This possibly underdoses patients further into their abstinence with lower blood alcohol levels than those who had their last drink more recently and presented with higher blood alcohol levels. A higher MCV was also noted to be associated with a higher risk for ICU admission possibly because this represents more chronic alcohol use and hepatic function impairment. Prior studies have established that MCV is elevated in chronic alcohol use, however, a relationship between higher MCV values with higher levels of drinking is lacking. 16 Among clinical parameters, a higher maximum CIWA-Ar score during ward stay was associated with ICU admission. This helps us further validate the CIWA-Ar scale which is widely used. As a corollary, higher sedatives administered to patients were also found to be significantly associated with ICU admission. This can be expected as higher CIWA-Ar scores call for higher doses of sedatives. Also, before ICU admissions, large quantities of sedatives are administered to control agitation during rapid response team activations for agitated patients. We also found that patients on an FDTR compared to STR were associated with an increased risk. This is likely because patients on a fixed dose taper had high-risk features such as a history of ICU admission for withdrawal or DTs which prompted the clinician to choose this regimen in the first place. It is unclear if this points to the possibility that the FDTR at our institution is not adequate to prevent impending withdrawal among high-risk patients. The FDTR and STR at our institution are detailed in Appendix 1. Statistical analyses using ROCs showed us that an alcohol level of <13 g/dl and a CIWA-Ar score of >16.5 significantly predicted ICU admission. Both these values are objective parameters clinicians may use to identify such high-risk patients during their management. We attempted to find a dose of sedative that prevented ICU admissions and a linear response with CIWA-Ar score was noted. Through our study, it is not possible to suggest a threshold dose of benzodiazepine that might reliably prevent an ICU admission. Conclusion We found a significant association between ICU admissions and parameters such as a history of delirium tremens or alcohol-related seizures, the initial protocol of management, activation of rapid response teams, mean corpuscular value, alcohol level on admission, single highest CIWA-Ar scored during the hospital stay, and the total amount of sedatives used. Clinicians must be aware of these biochemical and clinical parameters and monitor patients at high risk closely. Objective cut-offs including a blood alcohol level on admission <13 mg/dl and CIWA-Ar score >16.5 offer a good balance of sensitivity and specificity in reliably predicting patients at a high risk of requiring ICU level of care. Our results encourage further research to create an objective tool using the parameters explored and explore management algorithms that might be better tailored to patients at higher risk. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.	2023-09-05T15:03:08.684Z	0001-01-01T00:00:00.000	{ "year": 2023, "sha1": "e3e0f51a7521d1f8f9ed0bebe35b86d3585879cd", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "ScienceParsePlus", "pdf_hash": "2a7964e67e907edf3bdc1bfff9b8f55fb15f1079", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [] }
209071731	pes2o/s2orc	v3-fos-license	Applied structured database in a small project The growth of software application today has a lot of variety. The ease for software developers which provide an interest for all is to do self-taught learning. However, not all application developers are able to apply database theory in developed database systems. Database development needs to be normalized. Famous theoretical normalization for the database is known as BCNF Boyce-Codd normal form. This theory is actually very easy to understand but in the execution, application developers for small project application do not use the theory in the implementation of the database in their system. One of the reasons is the lack of understanding of normalization process. The simplest way to verify database normalization in a small project is with the use of ERD (Entity Relationship Diagram). This can be known by the same representation of the number of entities in the ERD by the number of tables in the database schema. Introduction With the rapid development of information technology [1], various types of software that facilitate users to develop their own applications are also growing. Observer of this application developer is not only from people who have a computer skill background, but also everyone who has the interest and expertise gained from self-taught learning. With the internet technology [2], application development tutorials with various kinds of software make learning easier. In addition, the development of information technology supported by increasingly good features and inexpensive supporting equipment triggers various institutions [3], government [4], schools [5] and corporations [6] to always improve in the recruitment of information technology experts to integrate their data and systems in order to run more effectively and efficiently. For example, in the banking sector [7], all banks provide online services in doing financial transactions; online stores [8] are able to make transactions to purchase and sell goods with online system without having to meet their customers; and employee recruitment [9] can also be done online. Therefore, all existing data must be stored properly, for example in a bank, the data of customers who do the saving must be kept secure; online stores must have updated data on the prices of the goods they sell; or agencies that do recruitment must maintain the data validity of prospective employees. These data should be managed by database experts in order to speed up the process and decision making accurately, quickly and precisely. Therefore, learning of the database technology development starts from institutions or educational institutions through teaching and learning activities, research and dedication to educate and train the resources that they have through teachers, staff and students. Moreover, learning of system or application integration with database technology has been developed through small projects written in scientific work both at national and international levels. Unfortunately, the standardization of projects that integrate with the database cannot be applied. This situation occurs especially when students carry out the final project by doing simulation project or building application system that is integrated with the database. Their database conditions which integrate with project or application are usually not in normal conditions. This condition can be said as a normal condition when the database does not undergo any repetition and there is certainty of data dependencies (the data is in the right table). Database Database [10] is a data collection that is interrelated. This relation is indicated by the relationship between tables which is related to the key in the table. Each table has many attributes that are used to describe themselves as unique identities. Terms in Database In the database there are several terms that must be understood by database developers, including (a) Attributes that can be referred to as data elements, data fields, or data item used to explain detailed information of an entity; (b) Data value that is the contents of the actual data stored on the attribute; (c) An entity that is a collection of attributes that are unique; (d) Record/Tuple that is a collection of elements that are interconnected to inform detailed information about an entity. Each record usually represents one data or information; (e) File that is a collection of similar records with the same element length, the same attributes, but with different data values. Criteria of Database Database development should have the following criteria: (a) Data stored does not experience repetition or redundancy to prevent anomalies and save space; (b) Databases can be utilized by many applications without having to change their structure; (c) Database security is prioritized; (d) Relationships between tables must be clear; (e) Database is easy to manage. Objective of the Database Database development has the following objectives: (a) The use of databases makes the speed of data access in processing large amounts of data more efficient; (b) Data accuracy for large amounts of data using a database more structured; (c) The sharing of data for different applications can be sharebility; (d) Eliminating repetition of data; (e) Providing data flexibility. Component of Database Components as supporting databases include (a) database support hardware that is usually in the form of computer devices that have the specification standards of minimum storage media required for the database; (b) Operating System that is a system software that regulates the performance of hardware and basic syntax operations including programs integrated with the computer; (c) Database: a collection of structured data from various types of facts stored in a storage medium that is represented in the form of a schema; (d) DBMS stands for DataBase Management System. DBMS is a system used to manage large amounts of data supported by data performance and integrity that can operate independently so as to produce data that is in accordance with the user's request; (e) Users that are people who use database facilities so they can insert, update or delete data; and (f) Database Developer Software that is a software that allows database developers to create databases. Normalization Database normalization [11] is a systematic approach used in building logic database relation design by applying a number of standard rules and criteria to minimize data redundancy in a database so that the database is able to work optimally. This design logic database can be defined by using functional dependencies to state normalization rules. The objective of normalization is to produce a normal structure. The table structure can be said to be in the normal form if the rules imposed on relations or 3 tables in the database meet the stages at normalization levels. The stages of normality of a database include the form of a database that is not normal (still according to pure data from field data), the 1 st Normalize Form or called 1NF, the 2 nd Normalize Form or called 2NF, the 3 rd Normalize Form or called 3NF, and the Boyce-Codd BCNF form. Unnormal Form Unnormal Form is a collection of data that will be processed without having to follow a certain format. Data in this form is obtained from a variety of formats so that there is still duplication of data because the data is in accordance with field facts that are sometimes imperfect or incomplete. In relational databases there should be no duplication of data because it can affect anomalies. This anomaly occurs if in one line/tuple there is data to be deleted so the consequences are other data may lost. 1 st Normalize Form The first normal level or also called 1 NF if each attribute only has a single value of record. In other words, data in an abnormal condition must be changed to the first normal form by creating a row containing the same number of columns with each column containing only one value. This is to confirm that each field contains only one meaning and is not a data set that has a double meaning. 2 nd Normalize Form In the design of relational databases there should be no partial functional dependency to the primary key, because it can have an anomaly. Therefore the first normalization stage will produce a second normal form or what is called as 2 NF. It is considered as 2 NF form if and only if the relation in the database meets the criteria of the 1 st Normalize Form. Morever, each of non-key attribute in 2 NF must functionally dependent on with the Primary key. So that when the 2 nd Normalize Form has been formed then the key in the defined field has also been determined. 3 nd Normalize Form Anomalies can occur if in a relational database design there is a transitive functional dependency. The functional dependency transitive occurs because there is a functional dependence of an attribute on another attribute through another attribute as well. So that the normalization of the third form or often referred to as 3 NF is a database relation that meets the rules of 2 NF form and all of the attributes are not primary keys that do not have a transitive functional dependency relationship to the primary key. This means that each non-key attribute must depend only on the primary key as a whole, and the third form of normalization already has an optimal table. The 3NF form is a database normalization technique that has a very close relationship with the normal form of BCNF Boyce-Codd form. At this stage of normalization, BCNF or 3NF is considered sufficient to minimize problems in database design (redundancy, lossless, dependency, and preservation). ERD ERD [12] is a diagram technique used to model data during system development. This technique provides a basic description for carrying out the relational database design specifications that underlie the information system developed. Components of Entity Relationship Diagram ERD has 4 main components: (a) Entity, an object that is a representation of a table, which is usually referred to as a single noun and is represented as a rectangle; (b) Relation that comes in the form of a rhombus to express the relationship of two entities connected by a straight line; (c) Attributes, that provide more detailed information about the details (characters) of the entity and can be optional; (d) Outline that serves as a link between relations and entities or relations and entities with attributes. Cardinality/Degree of Relations Cardinality or degree of relation is used to describe the maximum entity relationship in a set that can relate to entities in other sets and vice versa. Cardinality relations that occur between two sets of entities can be: (a) One to One relation: each entity in entity X is only related to one entity in entity Y and vice versa. (B) One to Many relations: each entity in entity X can have a relationship with many entities in entity Y, but not vice versa. (c) Many to One relation: each entity in entity X can have a relationship with at most one entity in entity Y, but not vice versa. (D) Many to many relations: each entity in entity X can relate to many entities in entity Y and vice versa. Discussion The steps in implementing this normalization are indeed used as a first step to understanding the development of a database so that data repetition does not occur. However, when this normalization process is applied to a project that has case studies, its application becomes more difficult. To that end, understanding is to make it easier to normalize by implementing Entity Relationship Diagrams (ERD). To find out more clearly how the application of ERD in the database concept, it can be seen in the sample of the small project that was done in the final project in case study basis. The samples presented were ERD and Database Schema applied in Schedule simulation for Department of Computer Science at the Universitas Negeri Semarang, Indonesia. ERD for Schedule Simulation The detail relation description of ERD for Schedule Simulation in the Department of Computer Science can be seen in Figure 1. The business rules that can be seen in Figure 1 are: 1. Each lecturer is able to teach more than one course and each course can be taught by many lecturers so that many to many relationships arise. If there is many to many relation between those two entities there must be a new entity whose attributes consist of foreign keys originating from these two entities ( Figure 2). 2. Each lecturer in a particular subject has 1 status that is as Chairperson of the Lecturer or becomes Assistant of Lecturer. If there is only one lecturer, it means that the lecturer is the Chairperson of the lecturer who is considered as a single lecturer. 3. The lecture_course entity can have many teaching schedules influenced by semester entity and room entity. Database Schema for Schedule Simulation Database Schema of the Schedule Simulation in Department of Computer Science at the Universitas Negeri Semarang in Figure 2 that is a detailed description of the entity in the ERD for the Schedule Simulation that can be seen in Figure 1. By using the database scheme, relations between tables can be figured out by the linking the keys between the tables. There are seven tables for the detail description of Database Schema of the schedule simulation and each table have their relation. Every lec Lecturer table represents detailed information of lecturers' data including lecturer_id, lecture_ein (ein:employer id number) and employer_name. Course table consists of course_id, course_name, ucu (university credit unit), course_type (theory or practice), course_practice_code. Therefore, these two tables have many to many relationship. Due to that condition, there must be a new table between those two tables called lecturer_course table consisting of keys from lecturer table and lecturer course as the foreign key. Lecturer_course table also includes status_id as a foreign key from the table status which represent status of lecturer as a chairperson or assistant and schedule_id from schedule table. Furthermore, schedule data are stored in schedule table containing foreign key from semester, room and course tables. so they should have model databases following the rules applied in the database structure. The easiest way after normalization is to see relations between tables by matching entity relationships in ERD, especially for small projects that are usually developed by students in a thesis or dissertation to stimulate their small project. The term of normal condition occurs if database entities and schema have the same number of expectations.	2019-11-22T01:31:28.177Z	2019-10-01T00:00:00.000	{ "year": 2019, "sha1": "725e6b2a3ed36860af70f43ec5687f22abc0c580", "oa_license": null, "oa_url": "https://doi.org/10.1088/1742-6596/1321/3/032130", "oa_status": "GOLD", "pdf_src": "IOP", "pdf_hash": "e06e49f31444b6f2cccab985bece1dc9fb27ea3f", "s2fieldsofstudy": [ "Computer Science" ], "extfieldsofstudy": [ "Physics", "Computer Science" ] }
265658438	pes2o/s2orc	v3-fos-license	Investigation of effects of interlayer interaction and biaxial strain on the phonon dispersion and dielectric response of hexagonal boron arsenide In this study, the effects of interlayer interaction and biaxial strain on the electronic structure, phonon dispersion and optical properties of monolayer and bilayer BAs are studied, using first-principles calculations within the framework of density functional theory. The interlayer coupling in bilayer BAs causes the splitting of out-of-plane acoustic (ZA) and optical (ZO) mode. For both structures, positive phonon modes across the Brillouin zone have been observed under biaxial tensile strain from 0 to 8%, which indicate their dynamical stability under tensile strain. Also, the phonon band gap between longitudinal acoustic (LA) and longitudinal optical (LO)/transverse optical (TO) modes for monolayer and bilayer BAs decreases under tensile strain. An appreciable degree of optical anisotropy is noticeable in the materials for parallel and perpendicular polarizations, accompanied by significant absorption in the ultraviolet and visible regions. The absorption edge of bilayer BAs is at a lower energy with respect to the monolayer BAs. The results demonstrate that the phonon dispersion and optoelectronic properties of BAs sheet could as well be tuned with both interlayer interaction and biaxial strain that are promising for optoelectronic and thermoelectric applications. exhibits a high thermal conductivity of approximately 1300 W m -1 K -127 , semiconducting band structure with a suitable bandgap between 1.5 and 2.0 eV 25,28,29 , and good mechanical performance 30 .Moreover, among all the III-V compounds, BAs is the most covalent III-V compound 31 and it is stable against chemical decomposition 32 .These intriguing properties motivated researchers to investigate electro-optical, thermal, and transport characteristics of BAs sheets in greater depth 20,[33][34][35][36] .The studies have shown that BAs monolayer exhibits semiconducting properties with a direct band gap, which can be modulated by applying the strain and it is dynamically stable, as evidenced by the phonon mode behavior 37,38 .The h-BAs monolayer has a smaller bandgap than other III-V group materials such as BN, AlN, and BP monolayer.This suggests that the BAs monolayer has a higher electrical conductivity than these other materials, which could make it a promising material for applications in alkali metals-ion batteries 39 .The BAs monolayer is a promising material for ideal alloy systems due to its greatest covalent character, stability against chemical decomposition and dissolution, and well-aligned valence bands [40][41][42][43] .In addition, BAs monolayer possess high carrier mobility, high thermal conductivity comparable to graphene, and exhibits photoactivity under illumination by visible and UV light 23,38,44,45 .Research on the electrical transport properties of hexagonal BAs has revealed behaviors suitable for short-wavelength optoelectronic devices 38 .Theoretical studies have shown that BAs sheet has the potential to be used as gas sensor with high sensitivity and selectivity 35,46 .The combination of these useful properties makes BAs monolayer a promising 2D material for next-generation photonic and electronic applications. Applying the strain on 2D materials is a highly effective method for modifying their properties [47][48][49][50] .By deforming the lattice parameters of a 2D material, strain can alter the electronic, optical and phononic properties of the material in various ways.For instance, strain can induce changes in the bandgap, carrier mobility, and thermal conductivity of a 2D material, leading to enhanced or new functionalities.The impact of strain on the phonon dispersion can affect the thermal conductivity of the material. Bilayer systems are of great interest as they offer additional control over properties compared to monolayers, through the interlayer interactions which can be tuned via stacking orientation, interlayer spacing, external electric fields, etc. [51][52][53] .In this paper, we investigate the phonon dispersion, electronic, and optical properties of bilayer BAs under strain, which has not been done before.By comparing the results with the corresponding monolayer cases, we provide a comprehensive description of the physical outcomes. Computational model and method We have used density functional theory (DFT) as implemented in SIESTA 54 code to study the electronic and optical properties.For exchange-correlation functional, the Perdew-Burke-Ernzerhof (PBE) type of the generalized gradient approximation (GGA) was implemented 55 .In order to determine band gaps with higher accurately, we have used the Heyd-Scuseria-Ernzerhof (HSE06) hybrid functional with the HONPAS package 56,57 .The selection of the double zeta polarized (DZP) basis set was accompanied by setting the orbital confining cut-off at 0.01 Ry.The value of the mesh cutoff for real space projection is 500 Ry.The threshold value for the total energy and forces during structural optimization is set to 10 -6 eV and 0.01 eV/Å, respectively.The first Brillouin zone is sampled with a 10 × 10 × 1 Monkhorst-Pack grid of k-points.For optical properties, a sufficiently dense k-point grid of 150 × 150 × 1, within the Monkhorst-Pack scheme is used.Optical broadening of 0.15 eV was used for optical spectra. To structural optimizations and phonon dispersion calculations were performed using the JDFTx package.Incorporation of the van der Waals (vdW) interaction was achieved by implementing the dispersion correction through the use of the DFT-D2 method 58 .The DFT calculations with a plane-wave basis set with a 50-hartree kinetic energy cutoff and, throughout, Coulomb truncation is included.During the process of structural optimizations, the relaxation of atoms was carried out until the force acting on each atom reached a convergence value of 0.1 mHa bohr −1 .To capture the phonon properties, a 4 × 4 × 1 supercell is used.To prevent interlayer interactions, a vacant space of 20 Å separates the sheets. To simulate the in-plane tensile strain, the formula ε = (a − a 0 )/a 0 is used, where a 0 represents the unstrained lattice constant and a represents the strained lattice constant. The absorption coefficient α(ω) and reflectance R(ω) can be calculated by the following formula 59 : Results and discussion The optimized geometric structure of BAs monolayer is similar to graphene where alternating B and As atoms replace carbon atoms in the hexagonal lattice with a space group of P-6m2.The relaxed lattice parameters of BAs are a = b = 3.39 Å with B-As bond length of 1.95 Å and bond angle of 120° which coincide with previous results 19,23,41,60,61 . In order to find the most stable stacking mode of the BAs bilayer, five stacking patterns, namely AA, AA 1 , AB, AB 1 and AB 2 were considered (see Fig. 1).In the AA (AA 1 ) stacking pattern, the B atoms of top layer are located above the B (As) atoms of bottom layer.For AB stacking pattern, the B atoms on top layer are located above the As atoms on bottom layer, while As atoms on top layer centered above the hexagon of the bottom layer.In the AB 1 (AB 2 ) stacking pattern, the B (As) atoms on top layer are located above the B (As) atoms on bottom layer, while As (B) atoms on top layer centered above the hexagon of the bottom layer.After geometry optimizations, (1) the interlayer distances of bilayer BAs for the AA, AA 1 , AB, AB 1 and AB 2 stacking patterns are 4.12, 3.78, 3.37, 3.57 and 3.85 Å, respectively.The binding energy (E b ) was obtained as the difference between the total energy of the BAs bilayers (E tot ) and the isolated BAs monolayers (E m ), as follows 40,61,62 : The calculated binding energy for the AA, AA 1 , AB, AB 1 and AB 2 stacking patterns are − 40.55, − 63.86, − 95.51, − 84.48 and − 56.19 meV, respectively.The shortest interlayer distance and the smallest binding energy of the AB stacking pattern confirm that the AB stacking pattern is the most energetically favorable.After the structural optimizations, the planer bilayer structure is converted into buckled one, due to the interlayer interactions.For bilayer BAs, the optimized value of buckling height and B-As bond length is 0.96 Å and 1.96 Å, respectively. The calculated electronic band structures along with the corresponding partial density of states (PDOS) of monolayer and bilayer BAs is presented in Fig. 2. The BAs monolayer exhibits direct gap semiconducting behavior with a band gap of 0.79 eV using PBE method and 1.19 eV using HSE06 method.Both the conduction band minimum (CBM) and valence band maximum (VBM) of BAs monolayer lie at the K point.The appreciable band www.nature.com/scientificreports/gap of BAs could make it suitable for use in nanoelectronics applications and the direct band gap characteristics of this material enables it to have several applications in the optical devices field because of the lower energy required to form excitons.The bilayer BAs is an indirect gap semiconductor.The CBM is situated at the K point while the VBM is slightly far from the K point.The calculated band gap is 0.63 eV using PBE method and 0.96 eV using HSE06 method.The calculated PDOSs show the contributions of individual orbitals of monolayer and bilayer BAs.The VBM is mainly contributed by p orbitals of As, and CBM mainly comes from the p orbitals of B. The calculated band structures of monolayer and bilayer BAs subjected to different biaxial strains is demonstrated in Fig. 3. Up to 8% of strain, the monolayer (bilayer) BAs demonstrates direct (indirect) band gap semiconducting behavior.For both structures, by increasing the tensile strain, conduction band edge at Γ point shifts downward relative to the Fermi level.Applying a strain in the range of 0-8% results in a slight increase of the band gap value of monolayer BAs, from 0.79 (1.19) to 0.89 (1.36) eV using PBE (HSE06) method, while the band gap value of bilayer BAs remains almost unchanged. Figure 4a shows the phonon dispersion of monolayer BAs.No imaginary modes are observed in the first Brillouin zone for monolayer BAs, indicating its dynamical stability.There are three acoustic and three optical phonon modes corresponding to the two atoms in one unit cell.The LA and TA modes exhibit linear dispersions in the vicinity of the Γ point, while the ZA mode has a quadratic dispersion relation because of the 2D character of monolayer BAs.The phonon energy of the longitudinal optical (LO) and transverse optical (TO) degenerate modes at Γ point for BAs is 111 meV.The LO and TO branches are also degenerated near the K point, indicating strong covalent bond between the B and As atoms.Also, a direct phonon gap of 62 meV between the acoustic braches and LO and TO branches is observed. Figure 4b shows the phonon dispersion of the bilayer BAs.There are 12 phonon modes, including 9 optical modes and 3 acoustical modes in the bilayer structure corresponding to the four atoms in one unit cell.If two layers are placed together without interaction, we expect that their phonon dispersions completely overlap.The phonon dispersion of bilayer BAs is similar to the one of monolayer BAs and most of the branches in the bilayer can be viewed as the ones split from the monolayer.Very small changes are observed in the in-plane doubly degenerated bands (the in-plane LA and LO 1 , TA and TO 1 , LO 2 and LO 3 , TO 2 and TO 3 ) implying that the main effect of the interlayer interactions is due to the ZA modes.The weak interlayer coupling causes the splitting of out-of-plane ZA and ZO modes to ZA and ZO 1 and ZO 2 and ZO 3 .The splitting of ZA and ZO 1 and ZO 2 and ZO 3 has an appreciable magnitude near the Γ point. Next, we study the effect of in-plane biaxial strain on the phonon dispersion of monolayer and bilayer BAs.The structural optimizations show that the value of buckling is decreased under tensile strain.Figures 5 and 6 display the phonon spectrum of the monolayer BAs under biaxial tensile strain.There are no negative phonon frequencies under tensile strain which indicates that the monolayer and bilayer BAs are dynamically stable under tensile strain. For monolayer BAs, by increasing the tensile strain the LO/TO, LA and TA phonon modes become soften which can be attributed to the weakness of atomic bond strength by increasing the lattice constant.The phonon band gap between LO/TO and LA modes decreases by increasing the tensile strain.A hardening of the ZO and ZA modes is also found for monolayer BAs under biaxial tensile strain and the hybridization between LA/TA and ZO mode is removed.For all applied tensile strains, the acoustic TA/LA/ZA phonons remain degenerate at Γ-point. For bilayer BAs, by applying the tensile strain the )LO 2 , LO 3 /TO 2 , TO 3 ) phonon modes shift to lower mode frequencies leading to phonon softening and the phonon band gap between (LO 2 , LO 3 /TO 2 , TO 3 ) and LA and LO 1 decreases.The ZA mode becomes harden, similar to the monolayer structure. In order to examine how interlayer interaction and strain affect the BAs sheet, we calculated the most important optical parameter, namely, the complex dielectric function which is represented by: ε(ω) = ε 1 (ω) + iε 2 (ω) where ε 1 (ω) and ε 2 (ω) are the real and imaginary terms, respectively.By counting the matrix elements of interband optical transitions between occupied and unoccupied states using the random phase approximation (RPA), it is possible to identify the imaginary part.The Kramers-Kronig relations can then be applied to compute the real part.With both the real and imaginary components, it becomes feasible to calculate other optical parameters of a material, including its absorption and reflectivity spectra. The frequency dependent absorption coefficient α(ω), real ε 1 (ω) and imaginary part ε 2 (ω) of dielectric function, and reflectance R(ω) for monolayer and bilayer BAs without and with strain are investigated and demonstrated in Figs. 7, 8, 9, 10, 11, 12, 13, 14.Anisotropy of optical spectra for parallel and perpendicular polarizations is obvious from Figs. 7, 8, 9, 10, 11, 12, 13, 14.The anisotropy of the optical spectra can be clearly observed for parallel and perpendicular polarizations as depicted in Figs. 7, 8, 9, 10, 11, 12, 13, 14.Figures 7 and 8 show the absorption spectrum of monolayer and bilayer BAs under biaxial tensile strain, respectively.For light polarized along the x direction, the unstrained monolayer BAs shows a pronounced absorption in the UV and visible regions.In particular, a strong peak is observed at 6.32 eV, while a weaker peak is visible at 2.66 eV within the visible region.The absorption spectrum of unstrained bilayer BAs contains some shoulders and additional weak peaks with respect to the monolayer BAs.The bilayer BAs has two main peaks at 2.30 and 6.59 eV, respectively (see Fig. 8a).The absorption onset energies for monolayer and bilayer BAs are observed to be approximately 0.69 eV and 0.55 eV, respectively.These values are consistent with the presence of a direct band gap at the K point, with corresponding energies of 0.79 eV and 0.63 eV for monolayer and bilayer BAs, respectively.The shift of the absorption edge and first main absorption peak to lower energies in bilayer BAs is due to the interlayer interactions.These findings suggest that the absorption of light by BAs occurs through the excitation of electrons from the valence band to the conduction band, and that this process is more effective in bilayer BAs compared to monolayer BAs. When light is polarized along the z direction, the unstrained monolayer BAs sheet exhibits an almost negligible absorption in the visible region.However, it demonstrates a robust absorption feature in the UV region with significant absorption peaks at 6.92, 9.80, 11.11, 12.45, 13.10, 14.76, and 16.42 eV.The absorption spectrum of the unstrained bilayer BAs has a small peak at visible region and several absorption peaks at UV region.The unstrained bilayer BP has main absorption peaks at 3.22, 6.67, 10.19, 13.62, 16.15 and 17.32 eV.It can be observed that the number of peaks in the bilayer structure is more than the monolayer structure and the peaks become broader, due to the interaction between the layers along the z direction. The absorption spectra of monolayer and bilayer BAs show an effective absorption of light in both the visible and UV regions.This property makes them promising candidates for the development of optoelectronic devices that operate at short wavelengths, where high absorption efficiency is critical.Additionally, their ability to absorb UV light could make them useful in applications where protection against UV radiation is important, such as in sunscreen or protective coatings.Overall, the demonstrated ability of BAs to absorb light in multiple regions of the electromagnetic spectrum highlights their potential for a range of applications in the field of materials science and optoelectronics. The calculated imaginary part ε 2 (ω) of dielectric function for monolayer and bilayer BAs under biaxial tensile strain is demonstrated in Figs. 9 and 10.For light polarized along the x direction, the ε2(ω) for monolayer BAs shows three significant peaks at 0.92, 2.56 and 6.24 eV in the near-IR, visible, and UV regions of the electromagnetic spectrum, respectively.For bilayer BAs, there are two strong optical absorption regions with multiple peaks in the range 0.6-3.4 and 5.8-9.2eV.The main absorption peaks occur at 0.76, 1.49, 2.23 and 6.24 eV.It can www.nature.com/scientificreports/be observed that monolayer BAs exhibits no ε 2 (ω) peak up to 6 eV when subjected to z-direction polarization, which suggests that the material has no optical absorption within this spectral range.For bilayer BAs, a multiple peak is observed in the visible region from 1.9 to 3.4 eV.The real part of the dielectric function, ε 1 (ω), provides insight into both the electronic polarizability and dispersion effects of the material.Figures 11a and d and 12a and b show ε 1 (ω) for light polarized along the x and z directions for monolayer and bilayer BAs, respectively.The key parameter of interest in this part is the static dielectric constant, which is determined by the magnitude of ε 1 (ω) at ω = 0.The static dielectric constant of the unstrained monolayer BAs has been calculated to be 4.45 and 1.31 along the x and z directions, respectively.The calculated ε 1 (0) for bilayer BAs, is found to be 8.26 and 2.23, along the x and z directions. The calculated reflectance spectra R(ω) for monolayer and bilayer BAs under biaxial tensile strain is demonstrated in Figs. 13 and 14.For light polarized along the x and z directions, the reflectance spectra static values are observed at some low values of 0.13 and 0.01 (0.23 and 0.04) for monolayer BAs (bilayer BAs), respectively.For light polarized along the x direction, the main reflectance peaks occur at 0.84 , 2.66 and 6.32 eV (0.71, 1.46, 2.32, 3.24 and 6.64 eV) for monolayer BAs (bilayer BAs) with a reflectance of about 23.1%, 27.7% and 33.9% (34.3%, 29%, 38.1% and 41.4%), respectively.These results are consistent with the calculated ε 2 (ω) values. For light polarized along the x direction, the main absorption peaks of monolayer and bilayer BAs become red-shifted and their intensities decreases as the applied strain changes from ε = 0% to ε = + 8%.Also, for monolayer and bilayer BAs the first peak of ε2(ω) becomes blue-shifted, while the second and third peaks become red-shifted as the strain changes from ε = 0 to ε = + 10%.Similar behavior is observed in the reflectance spectrum.The absorption edge for monolayer and bilayer structures remains almost unchanged under application of tensile strain.Therefore, application of compressive strain decreases the absorption of monolayer BAs by decreasing the intensity and width of absorption.The calculated ε 1 (ω) and ε 2 (ω) for monolayer and bilayer BAs using HSE06 method are shown in Figs.S1 and S2. Conclusion Using the first-principles calculations based on DFT, the impact of interlayer interaction and biaxial strain on the band structure, phonon dispersion and optical characteristics of BAs sheet was systematically studied.Based on the phonon dispersion curve calculations, both the monolayer and bilayer BAs are dynamically stable under zero strain.The phonon dispersion of bilayer BAs is very similar to the one of monolayer BAs.Almost no change is observed in the in-plane doubly degenerated bands while, the interlayer coupling causes the splitting of outof-plane modes.When the tensile strains is applied, both the monolayer and bilayer structures are dynamically stable.For light polarized along the x direction, the main absorption peaks of monolayer and bilayer BAs become red-shifted and their intensities decreases by increasing the tensile strain.This study implies the potential applications of BAs sheet for optoelectronic and thermoelectric devices in nanoscale. ( 3 )Figure 1 . Figure 1.Top and side views of five different stacking patterns of bilayer BAs. Figure 2 . Figure 2. Electronic band structure and correspnding partial density of states (PDOS) for (a) monolayer and (b) bilayer BAs. Figure 7 . Figure 7. Calculated absorption coefficient α(ω) for light polarized along the (a) x and (b) z directions for monolayer BAs under biaxial tensile strain. Figure 8 . Figure 8. Calculated absorption coefficient α(ω) for light polarized along the (a) x and (b) z directions for bilayer BAs under biaxial tensile strain. Figure 9 . Figure 9. Imaginary part ε 2 (ω) of dielectric function for light polarized along the (a) x and (b) z directions for monolayer BAs under biaxial tensile strain. Figure 10 . Figure 10.Imaginary part ε 2 (ω) of dielectric function for light polarized along the (a) x and (b) z directions for monolayer BAs under biaxial tensile strain. Figure 11 . Figure 11.Real part ε 1 (ω) of dielectric function for light polarized along the (a) x and (b) z directions for monolayer BAs. Figure 12 . Figure 12.Real part ε 1 (ω) of dielectric function for light polarized along the (a) x and (b) z directions for bilayer BAs. Figure 13 . Figure 13.Reflectance spectra R(ω) for light polarized along the (a) x and (b) z directions for monolayer BAs under biaxial tensile strain. Figure 14 . Figure 14.Reflectance spectra R(ω) for light polarized along the (a) x and (b) z directions for bilayer BAs under biaxial tensile strain.	2023-12-06T06:17:50.054Z	2023-12-04T00:00:00.000	{ "year": 2023, "sha1": "1f21c68f9d7c16f0d93b93d9ec2d3abc5e8617f3", "oa_license": "CCBY", "oa_url": "https://www.nature.com/articles/s41598-023-48654-9.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "53a1ba5d7b37259554a1f1eecf50697cdb6f0747", "s2fieldsofstudy": [ "Physics", "Materials Science" ], "extfieldsofstudy": [ "Medicine" ] }
9062465	pes2o/s2orc	v3-fos-license	Quark-gluon plasma phenomenology from the lattice The FASTSUM Collaboration has calculated several quantities relevant for QCD studies at non-zero temperature using the lattice technique. We report here our results for the (i) interquark potential in charmonium; (ii) bottomonium spectral functions; and (iii) electrical conductivity. All results were obtained with 2+1 flavours of dynamical fermions on an anisotropic lattice which allows greater resolution in the temporal direction. Introduction The Particle Data Book [1] is a repository of particle physics knowledge, and yet it contains no entries on the deconfined phase of QCD. We present here some lattice calculations of phenomena in the quark-gluon plasma (QGP) phase with the ultimate aim of addressing this omission. The lattice approach is based on the non-perturbative study of correlation functions of operators calculated in a background of glue and dynamical (sea) quarks using the imaginary time (i.e. Euclidean) formulation 1 . For instance, to study the η c system, correlators, C(τ ), of the operator J = cγ 5 c are determined (c is the charm quark field). Each state i which has the same quantum numbers as J (and therefore can be excited by it) contributes a term ∼ e −M i τ to C(τ ), where M i is the mass of state i. For large τ , only the ground state remains, and its properties can be determined. While the lattice technique has had many successes in calculating phenomenologically relevant quantities at zero chemical potential, µ, particularly in the confined phase of QCD, it has a significant limitation for µ > 0. This is the well-known "sign problem" which has haunted efforts to extend our knowledge into the entire (T, µ) plane 2 . In this talk, I will discuss three lattice results obtained by the FASTSUM Collaboration at µ = 0 and non-zero temperature, T , above and below the deconfining temperature, T c , all produced on our 2+1 flavour, anisotropic lattices. These topics are a calculation of the potential in the charmonium system (Sec.2), bottomonium spectral functions (Sec.3), and a determination of the electrical conductivity of QCD as a function of T (Sec.4). This work uses lattice simulations with the parameters in Table 1 and a carefully crafted action with reduced lattice artefacts [6]. Anisotropic rather than isotropic lattices provide a distinct advantage due to the finer temporal resolution and correspondingly larger sampling of the correlator, C(τ ). This is particularly significant in the T > 0 case where the temporal extent is limited to 0 ≤ τ ≤ 1/T . Our T c value is obtained from the Polyakov Loop. Table 1. Lattice parameters used where N (s)τ is the number of sites in the spatial (temporal) direction. Our spatial and temporal lattice spacings are a s = 0.1227(8) and a τ = 0.0351(2) fm. Charmonium potential The interquark potential in charmonium is of great interest to phenomenologists modelling the QGP phase in heavy ion collisions. Knowledge of this potential aids our understanding of the J/ψ system, and the extent to which states become unbound with increasing temperature. Lattice calculations of the finite temperature interquark potential have, until recently, been restricted to the static (i.e. infinitely heavy) quark limit [7]. On the other hand, the finitemass interquark potential has been calculated at T = 0 [8] using the method developed by the HAL QCD collaboration for internucleon potentials [9]. This method first calculates the wavefunction of the two-particle system, ψ(r), which is then used as input into the Schrödinger equation, yielding the potential, V (r), as output. In our case, we consider non-local operators of (charm) quark fields, J(x, r) = c(x)ΓU (x, x + r)c(x + r), where Γ is an appropriately chosen Dirac (gamma) matrix, and U (x, x + r) is the gauge connection between x and x + r [10]. The correlation function of these operators is where the second sum is over the states i. Following [11], C(r, τ ) satisfies the Schrödinger equation which can then be used to extract the potential. As usual, the potential can be decomposed into the central, V C (r), and the spin dependent, V S (r), terms, V Γ (r) = V C (r)+s 1 ·s 2 V S (r), where s 1,2 are the spins of the charm quarks. In Fig.1 (Left), we show the charmonium central potential, V C (r) thus obtained at various temperatures above and below the deconfining temperature. We use N s = 24 in this work. Also shown is another determination of this potential, but obtained in the static quark limit [12]. As can be seen, there is a clear temperature dependence in our calculated potential which becomes less confining as T increases. In Fig.1 (Right) V S (r), is plotted showing a repulsive core with a temperature variation at large distances. Bottomonium spectral functions In this section we discuss our studies of bottomonium at T > 0. Recent results from CMS show that the 1S and 2S/3S Υ states have different relative multiplicities in Pb-Pb compared to p-p collisions at the LHC [13]. This result confirms the picture, originally proposed in the charmonium system, in which the higher mass states in quarkonium are the first to become unbound as the temperature increases beyond T c [14]. We have performed lattice simulations of the bottomonium system using an O(v 4 ) NRQCD lattice action [15] to represent the b-quarks. This extends our earlier work [16]. NRQCD is an approximation obtained from QCD as a velocity expansion in v/c, and is thus applicable for bquarks. The advantages of NRQCD over the (exact) relativistic quark formulation are two-fold. There is no periodicity in time which complicates the correlation function: in the relativistic case, backward movers effectively half the number of time points that carry independent information. The lack of these thermal boundary effects means that the NRQCD quarks should be viewed as test colour charges moving in a thermal bath of dynamical light quarks and gluons. Secondly, the solution of the NRQCD propagator is an initial value problem which is much easier to solve than the matrix inversion required in the relativistic case. Thus, for a given computer resource, much higher statistics can be achieved in the NRQCD case. The spectral function, ρ(ω), can be defined from the two-point correlation function, C(τ ), of bottomonium operators via where in the NRQCD case the kernel is K(τ, ω) = exp(−ωτ ). In principle, the spectral function contains complete information on the states in the channel considered. For a stable particle of mass M , ρ(ω) ∝ δ(ω − M ). For a resonance, this δ-function broadens acquiring a non-zero width, and if the state becomes unbound, the spectral feature disappears. We have used the Maximum Entropy Method (MEM), a Bayesian technique, to de-convolve eq.(2) to extract ρ(ω) [17]. The results for the S-wave (Υ) channel and P -wave (χ b1 ) channels are shown in Fig.2, all obtained with N s = 24. There is a distinct temperature dependence in both channels. While the Υ ground state (1S) peak is seen to decrease with T , it remains a distinct (i.e. bound) feature up to T ≈ 1.90T c . However, the excited state (2S) peak seems to disappear for T ∼ > T c . This agrees with the experimental result obtained by the CMS collaboration [13]. These results contrast with the χ b1 case where the ground state (1P) appears to melt at T ≈ T c , see Fig.2 (Right). Figure 4. The conductivity, σ, as a function of temperature. As well as the results from our work, quenched (N f = 0) [19,20] and dynamical results [21] are also shown. We now discuss our calculation of the electrical conductivity, σ [18]. Transport coefficients such as σ are of special relevance for understanding the bulk properties of the fireball in relativistic heavy-ion collision experiments. Typically, they can be determined from the zero energy limit of the appropriate spectral function, i.e. lim ω→0 ρ/ω. In the case of the conductivity, the correlator to consider is that of the electromagnetic current of quark flavour f , where e is the electron's charge, q f = 2 3 , − 1 3 , is the fractional charge and j f i ∼ ψ f γ i ψ f the number density current of quark field ψ f . In our case [18] we use the conserved lattice vector current (i.e. ∂ µ j f µ = 0) which means we do not have to renormalise our results. We again use MEM to invert the correlator to extract the corresponding ρ em (ω), see Eq(2), but this time with the relativistic kernel K(τ, ω) = cosh[ω(τ − 1/2T )]/ sinh[ω/2T ]. The conductivity is obtained via the Kubo relation, Fig.3 shows ρ em (ω) around ω ≈ 0 for three temperatures spanning T c . There is a clear non-zero intercept at ω = 0 for T ∼ > T c , indicating a conductivity signal. In Fig.4, we present our σ values for all the T values we studied. The dimensionless ratio σ/T increases across the transition temperature. Results from three other published works are also shown [19,20,21]. Conclusion In this talk, I have summarised our collaboration's lattice calculations showing that the charmonium potential becomes less binding with higher T , that the Υ(1S) state survives above T c but the Υ(2S) and χ b1 don't, and that the conductivity increases with T across the transition.	2013-10-18T19:13:17.000Z	2013-10-18T00:00:00.000	{ "year": 2014, "sha1": "844e7c01a5c1080386b5971026d6a4d88359eb6f", "oa_license": null, "oa_url": "https://doi.org/10.1088/1742-6596/509/1/012015", "oa_status": "GOLD", "pdf_src": "Arxiv", "pdf_hash": "ec53fc7adf975d34c574abe9eeb0c07db00897a4", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics" ] }
385293	pes2o/s2orc	v3-fos-license	Robotic telesurgery for achalasia The craft of surgery has always relied on the use of instruments. Innovations in surgery have paralleled innovations in instrumentation. Advances in surgical instrumentation continue today and have enabled huge strides in surgical procedures and outcomes during this generation. Computers and related technology are now changing the interface between the surgeon and the patient, and are poised to improve patient outcomes by enhancing the surgeon’s skills and training. The application of computer enhanced telemanipulators, or “robots”, may specifically enhance operations, for example Heller myotomy, that require good visualization and precise careful dissection of delicate structures. This review covers the pathophysiology of achalasia and its history of medical and surgical treatment, leading to modern robotic telesurgical approaches. Improvements in outcome from medical to standard surgical to robotic telesurgical approaches are discussed. Current operative technique for robotic telesurgical treatment of achalasia is described and the authors conclude with a glimpse of where, in the future, current research endeavors will lead us in the treatment of achalasia. Introduction Achalasia is a swallowing disorder that affects one in 100,000 individuals. Most people are diagnosed between the ages of 25 and 60 years. It initially presents with difficulty swallowing and is a chronic progressive condition that usually does not resolve [1]. Although the specific cause is unknown, the primary pathology involves the inability of the lower esophagus to relax, resulting in delayed passage of food and liquid into the stomach. Diagnosis is based on a combination of clinical symptoms and objective evaluations. Clinical symptoms range from initial dysphagia to heartburn, a feeling of fullness in the chest and sometimes weight loss. The objective work-up includes barium radiography, esophagoscopy, and esophageal manometry. In patients with achalasia barium radiography often reveals a classic ''bird's beak'' appearance ( Fig. 1) in which the dilation tapers into the lower esophagus, because of the chronic buildup of food contents. Esophagoscopy often shows retained food and possible fermentation. Manometric evaluation reveals nonperistaltic movement within the esophagus during swallows. Long-term sequelae include an increased risk of the development of esophageal cancer in patients diagnosed with achalasia. The objective of treatment is to resolve symptoms in this usually progressive disease. Conservative maneuvers have included the use of muscle relaxing medications or injections and dilation of the lower esophagus with endoscopically guided balloons. These usually result in temporary relief only rather than achieving long-term resolution of the problem. Surgical intervention involving splitting of the restrictive external muscle of the lower esophagus and upper stomach with subsequent wrapping of the upper stomach partially around the lower esophagus results in symptom relief in 85% of people 10 years after surgery and in approximately 65% of people 20 years after surgery [1]. Achalasia was first described, with its initial treatment, by Thomas Willis in 1679 [2]. Treatment and instrumentation have rapidly evolved since his first description of dilation of the esophagus using a tapered whale bone. In 1887, Russel described initial balloon dilatation using a silk bag over a rubber balloon [2]. Treatment over the last century has mainly focused on endoluminal dilatation or surgical myotomy. The first esophageal myotomy was described in 1913 by Heller. This surgical technique involved splitting the outer muscle of the lower esophagus to relieve internal narrowing and associated swallowing difficulties [3]. This procedure was subsequently popularized by approaches through the abdomen or through the chest. In the early 1990s, during the beginning of the era of minimally invasive surgery, thoracoscopic [4] and laparoscopic [5] Heller myotomy were described. Prompt recovery and good outcomes quickly enabled clinicians to conclude that surgical treatment was the optimum primary treatment of choice for amelioration of symptoms in patients with achalasia [5]. Laparoscopic Heller myotomy using standard laparoscopic techniques has evolved into an extremely safe and efficient technique of reducing symptoms in the treatment of achalasia. Preliminary case series demonstrated that early outcomes were favorable and the perioperative course was much improved compared with an open laparotomy or thoracotomy [5]. Perioperative complications were reduced and the complications secondary to the trauma of the incision were largely eliminated. Laparoscopic Heller myotomy has now been widely reported with large case series with relatively good outcomes in both early and late results [6][7][8]. In addition, favorable outcomes have been achieved in some series in which laparoscopic Heller myotomy was offered as the first-line therapy [9]. Although patient satisfaction, relief of symptoms, and recovery time were categorized as favorable, these large case series still reported a significant number of intraoperative esophageal perforations ranging between 1 [10] and 15% [5]. In these series, the mucosal perforations were identified to have occurred as a technical failure during the course of the division of the muscle fibers from the underlying esophageal mucosa. The clinical relevance of these complications may be uncertain, and when the perforation is recognized at the time of surgery and repair performed immediately the outcomes are still usually good. These perforations expose the patients to a prolonged hospital stay and the risk of an esophageal leak, however. Delayed diagnosis of an esophageal perforation is a potential fatal complication and is one of the most significant and serious risks of Heller myotomy. Management in this situation requires broad-spectrum antibiotic coverage, proximal bowel decompression and rest, potential operative drainage, debridement and reconstruction of the injured tissues, and sometimes intravenous nutritional support for a prolonged period of time. Different surgical techniques have been described to avoid the perforation of the underlying mucosa and have included the injection of epinephrine into the submucosal plane, blunt dissection, sharp dissection, and intraoperative esophagoscopy to evaluate the esophageal mucosa and detect underlying perforations. Operative technique and outcomes The da Vinci surgical system (Intuitive Surgical, Inc; Sunnyvale CA, USA) is currently the only robotic telesurgical device marketed and approved for general surgery in the United States. The device is a computer interface that interacts with the operating surgeon who sits at a control panel connected to a bedside actuator with instrument holders and a camera manipulator (Figs. 2, 3). da Vinci has instruments that rotate and articulate around three axes. Motion is enhanced by filtering fine motor tremor and providing motion scaling between the hand-piece and the activator arm, The robotic device has the unique ability to enhance surgical technique and, therefore, optimize surgical outcome. This enhancement may make more of a difference in some surgical procedures. These advantages specifically are fine motor control, three-dimensional magnification, and motion scaling. These may be significant in operative procedures such as Heller myotomy or in cases when optimum completion of the operation requires these specific surgical techniques. It is this enhancement of surgical performance that may have improved the surgical outcomes mentioned above regarding recent case series comparing robotic telesurgery with historic laparoscopic approaches to Heller myotomy [11]. Using this model it may be reasonable to identify some surgical procedures and interventions, for example Heller myotomy, that can specifically benefit from this type of therapy. At the author's institution robotic telesurgery for achalasia is performed in the following manner. Preoperative preparation of patients with achalasia may include a prolonged period of fasting to completely empty the esophagus. Clear liquids for 48 h preoperatively are routine, especially in cases of severe esophageal dilatation. Preoperative hospital admission may be necessary for patients to enable them to maintain fasting with intravenous hydration. Preoperative antibiotics and oral anti-fungal medications should be included to help reduce the amount of contamination within the esophagus. Deep venous thrombosis prophylaxis is given routinely. The patient is positioned on a split leg table or in a modified lithotomy (legs apart and gently angled down) position after the induction of general anesthesia. The bedside component of the robot is positioned over the patients left shoulder after accessing the abdomen through four ports. The patient is placed with the head of the bed angled upwards and the patient's motion must be completed before docking to the robot ports. Two working ports are used with downward traction of the upper part of the stomach by the assistant or the fourth arm of the robot. The esophageal hiatus of the diaphragm is identified and the crura are circumferentially dissected free of peritoneal attachments. The distal esophagus is dissected, and a soft rubber Penrose drain is usually placed around the distal esophagus for traction, enabling dissection up into the lower mediastinum. The esophageal myotomy is created using the electrocautery hook of (Fig. 4). The articulation of this arm provides precise control and helps keep the end of the hook directly parallel to the esophageal mucosal wall. This dissection is aided by a curved dissector on the opposite arm, to help completely separate the muscle fibers. When the submucosal is plane is developed between the outer esophageal musculature and the inner mucosa, the division of the muscle fibers is carried 8-10 cm proximal and at least 2 cm distally from the junction of the esophagus and stomach. The gastric dissection is clearly identified by the disordered character of the vasculature of the gastric submucosa. The mucosa is carefully inspected for injury and for complete division of all muscle fibers. The muscular bundles bordering the esophageal diaphragmatic hiatus are reapproximated behind the esophagus using nonabsorbable sutures (Fig. 5). A partial gastric fundoplication is then created using a 90°anterior Dor or a 270°p osterior Toupet fundoplication (Fig. 6). The robot is disengaged from the port sites, which are closed with sutures. Patients are allowed liquids post operatively and then are maintained on a soft diet for several days. Postoperative patients should receive routine follow-up consisting of surveillance endoscopy every 3 years in anticipation of early detection of asymptomatic pathology. Patients who complain of recurrent symptoms should undergo formal evaluation and workup as outlined for initial patients suspected of having achalasia. Long-term postoperative results for achalasia patients are good, particularly when patients are selected with preoperative lower esophageal sphincter pressures greater than 35 mmHg [12]. For a group of 174 patients Ruffato et al. reported overall symptomatic relief of dysphagia in 87% of patients 15 years after Heller myotomy and Dor fundoplication with 9% of patients requiring medical/surgical treatment of symptomatic reflux and 4% of patients requiring esophagectomy for malignant disease or recurrent disease not amenable to repeat myotomy and fundoplication [13]. In patients who require re-operation for recurrent dysphagia following esophagomyotomy symptoms often present within the first postoperative year and the root cause is usually inadequate or healed myotomy. Although success can usually be achieved, results from repeat esophagomyotomy are less impressive than adequate initial operations [14]. Subtotal esophagectomy for patients with achalasia is infrequently required. It is indicated after failed medical therapy, pneumatic dilation, non-resecting surgical, and redo procedures. Gockel et al. performed subtotal esophagectomy for eight patients for whom previous therapy had failed. Dysphagia was resolved and oral alimentation was restored in each patient [15]. Discussion Although laparoscopic Heller myotomy has emerged as the treatment of choice for achalasia, the intraoperative esophageal perforation rate of 1-15% has remained a significant problem that has prompted investigation of improvements of this operative technique. In 2000, the da Vinci surgical system (Intuitive Surgical, Inc; Sunnyvale CA, USA) a robotic telesurgical device, entered the market. The advantages reported include a stable work platform, a magnified three-dimensional image, fine motor control of articulated instruments, and motion scaling, which are uniquely designed to facilitate surgery requiring fine tissue manipulation. Disadvantages of robotic telesurgery systems include the cost of purchase and maintenance. They are large and may limit access to the patient during surgery. They provide a narrower field of view of the operative site and they provide the surgeon with essentially no tactile feedback [16]. With these advantages and hurdles in mind, a variety of surgical procedures have been reported using robotic telesurgical devices. Initial data obtained before Food and Drug Administration approval was obtained in clinical trials performing cholecystectomy and gastroesophageal fundoplication. After Food and Drug Administration approval brief clinical reports described cholecystectomy [17], fundoplication [18], colon resection [19], pancreatic resection [20], Heller myotomy [21], and other general surgery procedures [19,22]. With experience, case series were presented which eventually demonstrated the safety and feasibility of robotic general surgery procedures [22,23]. Reports have also been presented that look at traditional lapa- [24]. Both techniques were safe and associated with low occurrence of complications; no clear benefit was seen in the robotic group, however. Use of the da Vinci robotic telesurgical device to perform a Heller myotomy was first reported in 2001 [21] and a subsequent small case series was then reported [23,25]. The academic robotic surgery group, comprising surgeons from Ohio State University, Columbus, OH, USA, Johns Hopkins University, Baltimore, MD, USA, and the University of Illinois at Chicago, Chicago, IL, USA, has prospectively collected data on 104 robotic telesurgical Heller myotomies, the largest reported case series, which were performed without a single esophageal perforation. Similar results favoring the robotic telesurgical approach were obtained in three subsequent published series comparing robotic telesugical Heller myotomy with laparoscopic Heller myotomy [26][27][28]. This improvement over standard laparoscopic studies which reflected the esophageal perforation rate of 1-15% may be related to the advantages of the robotic telesurgical technique and surgical enhancement the device offers compared with laparoscopic techniques [11]. Future directions The surgical enhancement of robotic telesurgery is being applied to increasingly complex procedures. One pilot study evaluated the technique of pancreaticojejunostomy after pancreatic head resection using the robotic surgical device. Five patients were enrolled and underwent a computer-enhanced pancreaticojejunostomy using a 6-0 suture in an open abdomen after pancreatic head resection. Good results were obtained and all the anastomoses were completed [29]. Robotic assistance also has the potential to contribute to the evolving technology of natural orifice transluminal endoscopic surgery (NOTES) in which therapeutic approaches to abdominal surgical diseases are being pursued via an endoscope that is advanced from an intraluminal position within the stomach, vagina, bladder, or rectum into the abdominal cavity to perform a variety of diagnostic and therapeutic maneuvers. This was first attempted with simple peritonoscopy and has recently advanced to organ retrieval, for example appendectomy [30,31]. Although no formal treatments for achalasia by a NOTES approach have been published, a treatment involving transgastric peritonoscopy with retroflexion and endoscopic myotomy with potential intraluminal gastroesophageal plication may prove to be feasible for treatment of achalasia with end results similar to Heller myotomy and fundoplication.	2016-05-12T22:15:10.714Z	2007-01-20T00:00:00.000	{ "year": 2007, "sha1": "35a754fc5552ca13535ca38fe4dd6406b220a76c", "oa_license": "CCBYNC", "oa_url": "https://link.springer.com/content/pdf/10.1007/s11701-006-0001-y.pdf", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "35a754fc5552ca13535ca38fe4dd6406b220a76c", "s2fieldsofstudy": [ "Engineering", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
6614442	pes2o/s2orc	v3-fos-license	Improving performance of mammalian microRNA target prediction Background MicroRNAs (miRNAs) are single-stranded non-coding RNAs known to regulate a wide range of cellular processes by silencing the gene expression at the protein and/or mRNA levels. Computational prediction of miRNA targets is essential for elucidating the detailed functions of miRNA. However, the prediction specificity and sensitivity of the existing algorithms are still poor to generate meaningful, workable hypotheses for subsequent experimental testing. Constructing a richer and more reliable training data set and developing an algorithm that properly exploits this data set would be the key to improve the performance current prediction algorithms. Results A comprehensive training data set is constructed for mammalian miRNAs with its positive targets obtained from the most up-to-date miRNA target depository called miRecords and its negative targets derived from 20 microarray data. A new algorithm SVMicrO is developed, which assumes a 2-stage structure including a site support vector machine (SVM) followed by a UTR-SVM. SVMicrO makes prediction based on 21 optimal site features and 18 optimal UTR features, selected by training from a comprehensive collection of 113 site and 30 UTR features. Comprehensive evaluation of SVMicrO performance has been carried out on the training data, proteomics data, and immunoprecipitation (IP) pull-down data. Comparisons with some popular algorithms demonstrate consistent improvements in prediction specificity, sensitivity and precision in all tested cases. All the related materials including source code and genome-wide prediction of human targets are available at http://compgenomics.utsa.edu/svmicro.html. Conclusions A 2-stage SVM based new miRNA target prediction algorithm called SVMicrO is developed. SVMicrO is shown to be able to achieve robust performance. It holds the promise to achieve continuing improvement whenever better training data that contain additional verified or high confidence positive targets and properly selected negative targets are available. Background MicroRNAs (miRNAs) are single-stranded non-coding RNAs with about 22 nucleotides in length [1] known to mainly inhibit target translation or cleave target mRNA by binding to the complementary sites in the 3′ untranslated region (UTR) of targets. miRNAs have been shown and are speculated to play many important post-transcriptional regulatory roles in a wide range of biological processes and diseases including development, stress responses, viral infection, and cancer [2]. Despite rapid advance in miRNA research, the detailed functions and regulatory mechanisms of most of miRNAs are still poorly understood. To gain better understanding, an important task herein is to identify miRNAs' regulatory targets. However, the current knowledge about the known targets is disproportional to that of the known miRNAs. In miRBase, 969 human miRNAs are annotated; in contrast, only 815 targets of 121 human miR-NAs are recorded in the most up-to-date target database miRecords [3]. Given that targets of each miRNA could be hundreds, the reported number of verified targets accounts for only a very small fraction of the potential targets. This fact greatly underscores the urgent need to develop effective target identification methods for genome-wide target discovery. Considerable advances have been made in computational target prediction [4] and many algorithms have been proposed including TargetScan [5], PicTar [6], miRanda [7], PITA [8], DIANA-microT [9], RNAhybrid [10], microInspector [11], MovingTargets [12], rna22 [13], NBmiRTar [14] and Nucleus [15]. These algorithms make predictions mainly based on various important features of miRNA-target nucleotide sequence interaction. Although different algorithms utilize different sets of features, a few important features including "seed region complementary", "binding free energy", and "sequence conservation" are among the most common ones. Using different features will result in different prediction performance and a central goal of various algorithms concerns the selection of most discriminative features that can lead to better prediction accuracy. A promising direction there is within the data driven framework, where the features are partially or entirely determined by training using the training data composed of validated positive and negative targets. Algorithms including MirTarget [16], miTarget [17], and TargetBoost [18] are data driven algorithms and developed based on training. Given sufficient training data, the data driven algorithms hold the promise to provide accurate prediction, since they have the ability to uncover important features from data that cannot be easily observed otherwise. However, the existing algorithms have poor prediction specificity and sensitivity [19,20]. The performance deficiency is partially due to the poor understanding of the precise mechanisms underlying miRNA-target interaction [1,21], and therefore, the adopted features of the rules are not yet as specific and sensitive as needed. Besides, high quality training data essential for training data-driven algorithms is greatly lacking. For many algorithms, the positive training data are based on a very small number or even one of validated targets and thus hardly include important features relevant to different aspects of miRNA-target interactions; these problems hamper the ability of the data-driven algorithms to select discriminative features. Many others also select the positive targets from down-regulated genes in an mRNA microarray data of over-expressing a miRNA. However, since protein inhibition is considered as the primary function, any reduction at the mRNA level measured by microarray is likely due to the secondary effect of miRNA regulation. Consequently, these training data are neither specific since many under-expressed genes may not be targets, nor sensitive since many targets might not be under-expressed at the mRNA level. It is apparent that constructing a richer and more reliable training data set is the key to improve the performance of the current data-driven algorithms. The aim of this paper is to improve the sensitivity and specificity of target prediction by constructing a comprehensive training dataset and developing a support vector machine (SVM) [22,23] algorithm that exploits extensive binding features. Particularly, we take advantage of the most updated miRNA target depository called miRecords to construct positive training data set. In addition, we derive the negative data set based on 20 microarray data, each generated by over-expressing a different miRNA. With this more diverse, higher quality, and larger quantity training data set, we develop a more sophisticate two-stage SVM based algorithm called SVMicrO. In SVMicrO, 113 and 30 features were extracted to survey the potential binding sites and the UTR characteristics, respectively. A feature selection step is introduced to select most discriminative features for site and UTR. Comparison with several popular target prediction algorithms are performed based on training data and results from high confidence experiments including IP pull down and proteomics experiments. All these investigations indicate that SVMicrO achieves consistently improved sensitivity, specificity, and precision, which proves it to be a competitive alternative to the existing sequenced-based algorithms. The algorithm of SVMicrO The structure of SVMicrO is shown in Figure 1, which includes three steps. First, a site filter is applied, which uses the miRNA sequence to scan through the 3′UTR sequence for the potential binding sites of the probing miRNA. This filter is introduced to improve the efficiency of SVMicrO. The goal is to select the potential sites with high sensitivity since the sensitivity of the entire algorithm is upper-bounded by the sensitivity of the filter. In contrast, false positive sites identified by the seed match rules can be reduced by site-SVM and UTR-SVM. Many of the exiting algorithms also include a filter step, most of which rely on the presence of a 6-mer match in the seed region. However, by testing against the true binding sites and target pairs in the training data (see Construction of training data section for details), we find that more than 20% miRNA-site and 20% miRNA-target pairs do not possess the 6-mer seed match. This also implies that the existing target prediction algorithms relying on perfect seed match will result in a reduction of sensitivity regardless how good the later prediction is. As a result, a looser seed match rule should be used to gain higher sensitivity. However, the seed match rules should be not too loose to introduce too many false positives; otherwise they would increase the computation burden of the subsequent SVMs. Out of these considerations, we examine the different combinations of nucleotides match statuses in seed region by considering TargetScan seed match rule [5], the results from [24] and our own investigation on the experimentally validated sites as well as targets in miRecords. Among different combinations, the following 5 seed match rules achieve near 96% sensitivity on both experiment validated targets and sites, while introducing less false positive sites than other combinations. Thus, regions of the 3′UTR sequence that obey one of the seed match rules are considered as potential sites. (1) There are more than 4 continuous W-C matches, or (2) There are more than 5 continuous matches (including G:U pair) and more than 2 continuous W-C matches, or (3) There are more than 6 matches in total and 3 continuous W-C matches; no gap allowed, or (4) 2~4 nucleotides of miRNA are W-C match, there is more than 3 W-C matches and more than 4 matches in total; no gap allowed, or (5) There are more than 5 matches and 5 W-C matches, and only one gap is allowed on either miRNA sequence or 3′UTR. Specifically, W-C match stands for Watson-Crick match, while match denotes Watson-Crick match or G: U wobble pair. The sensitivity of proposed filtering rules is evaluated by the training data and shown to be around 96% for both site and UTR (Additional file 1 Table 1), thus satisfying our goal of achieving higher sensitivity for the filter. In the subsequent step, the potential sites identified by the filter are subjected to the Site-SVM, which extracts features from each site, and assigns a score to indicate the prediction confidence of the site as a true site. in the final step, the site scores together with other UTR features are considered by the UTR-SVM to produce the final prediction of the UTR as a target. Features extraction Two different types of features representing important and possible binding characteristics were extracted for the Site-and UTR-SVM. A total of 113 site features and 30 UTR features are extracted. Binding Structure definition To define features, we first provide the mathematical definition of miRNA-site binding as in Figure 2. For a given miRNA sequence of length M, let p = {p 1 , ..., p m , ..., p M } denote its nucleotide composition, where p m NT represents the nucleotide content at the mth position from its 5′ end and NT = {A, C, G, U}. For a binding site of length N, let q = {q 1 , ..., q n , ..., q N } indicate its nucleotide composition from 3′ end, where q 1 is the nucleotide corresponding to p 1 in miRNA. Naturally, {q 0 , ..., q -1 , ...} stands for the 3′ context of binding region. Since based on the current consensus, the seed sequence of miRNA complements with UTR much Figure 2 Binding structure and regional definition of miRNA and target site. Figure 1 The block diagram of SVMicrO. SVMicrO includes three steps. First, a site filter is applied to find the potential binding sites of the probing miRNA. Second step, Site-SVM extracts features from each potential site and assigns a score to indicate the prediction confidence of the site as a true site. Final step, the site scores together with other UTR features are considered by the UTR-SVM to produce the final prediction of the UTR as a target. better than the rest, we divide the entire binding site into two sub-regions, which are the seed binding region and the 3′ binding region. From miRNA point of view, {p 1 , ..., p 8 } belong the seed binding region while {p 9 , ..., p 20 } belong to the 3′ binding region. From mRNA's perspective, {q 1 , ..., q n } belong to the seed binding region and {q n+1 , q n+2 , ...} belong to the 3′ binding region, respectively, where q n is the nucleotide of UTR that pairs with p 8 of miRNA. We also use {r 1 , r 2 , ...} to denote {q n+1 , q n+2 , ...} in the following description. Moreover, only first 20 nucleotides of miRNA are consider in the feature extraction step in our algorithm, and thus the last nucleotides in {r 1 , r 2 , ...} pairs with p 20 of miRNA. Site Features 7 groups of site features are extracted to describe the characteristics of target recognition within a site (See Additional file 1 Table 2). Perfect seed match features Perfect seed match is widely used for binding site prediction in many target prediction algorithms. We survey 6 types of perfect seed matches (Table 1) and define the corresponding features as the existence of the respective match in the site. Pair-wise binding structure features Past research shows that miRNA binding varies according to the position of the binding structure. In order to observe these characteristics, a modified version of RNAduplex [25] named miRNAbind is developed, which can be found in the SVMicrO package on the provided paper website. miRNAbind uses RNAduplex to generate the required secondary structure of miRNA binding similar to that in Figure 2 and it provides additional information including binding energy of seed region, binding energy of 3′ region, exact boundaries of each regions, etc. Based on the secondary structure, four types of nucleotide matches are defined as W-C match, G-U match, mismatch, and gap. Subsequently, the match status of each nucleotide, represented by integer 1 to 4, as well as the content of each 2 mer, represented by integer 1 to 16, from p 1 to p 20 are extracted. There are totally 39 pairwise binding structure features. Regional binding structure features To investigate the local binding characteristics, the sub-regions defined above and the total regions are used. For each region, the total numbers of W-C matches, G-U matches, mismatches, and gaps are counted as regional features. Additionally, to the reveal the bulge structure on mRNA, the numbers of bulged structures and bulged nucleotides in each binding region are also counted as 2 additional features. There are totally 18 features in this group. Conservation features To investigate the conservation characteristics of sites, the binding region is again divided into 3 sub-regions, which are the seed binding region, 5′ context region ({r 1 , ..., r 10 }), and 3′ context region ({q 0 , ..., q -9 }). The conservation score of each nucleotide in the site is then obtained from the phast-Cons28way table (See Additional file 1 S.2) in the UCSC Gene Table. The conservation scores of each region are then defined as the average conservation scores of the respective regions. Energy features It is believed that miRNA-target binding forms a stable low energy hybrid. Hence, the more stable the binding structure is, the more likely the site is to be a true binding site. The binding energy features of the seed region, 3′ region and total region are thus evaluated by miRNAbind. Moreover, the accessibility defined in PITA [8] is also adopted as another energy feature, which is introduced to evaluate the openness attribute of the secondary structure of a potential site. Seed context features Context region stands for the contiguous upstream and downstream sequences of the seed region. It has been reported that seed region preferentially resides within a locally AU-rich context [5]. To this end, two 10-nt long sequences, {r 1 , ..., r 10 } and {q 0 , ..., q -8 }, on both ends of seed binding regions are isolated as context regions. The single nucleotide and 2-mer compositions are then recorded for each context regions to obtain 20 context features. Moreover, the nucleotide compositions of all positions in these 2 regions are regarded as another 20 context nucleotide type features. Site location features It is also reported that binding sites are more frequently observed at the two ends of a 3′UTR but not too close to the stop codon [5]. To reflect this point, we define 3 features including the distance from the potential site to stop codon, the distance from the potential site to the nearest end of 3′UTR, and the ratio of the distance from the potential site to the nearest end over the length of 3′UTR. UTR Features 3 groups of UTR features are extracted to describe the characteristics of target recognition within the 3′UTR (See Additional file 1 Table 3). Length of 3′UTR Since a target 3′UTR includes multiple binding sites, the length of the 3′ UTR presumably affects miRNA targeting. We investigate the length of 3′ UTR in our training data set and the result shows that [5,24]. To reflect this fact, global site density feature is calculated as the ratio of the number of potential/positive binding sites over the length of 3′UTR. Also, a 100-nt long window is used to identify a region with the maximum number of potential/positive binding sites and these maximum numbers are recorded as 2 features. Binding site score features The score produced by the Site-SVM for each candidate site can be regarded as the prediction confidence for this site. Consequently, the higher the confidence of site predictions, the more likely the UTR is to be a target. Again, the potential sites are the sites identified by the filter, and the positive sites are the potential sites predicted positive by the Site-SVM (SVM score >0). The total score of positive sites, the number of potential sites, the number of positive sites, etc as defines in Additional file 1 Table 4 are defined as 25 features. Construction of training data Data used for training and testing should include both positive and negative miRNA-site and miRNA-target pairs for Site-and UTR-SVM. RefSeq information for human, mouse and rat have been downloaded from UCSC Genome Browser mySQL database. The sequences of 3′UTRs and the conservation scores of each nucleotide in the 3′UTR have been retrieved from the UCSC Genome Browser either. A local database for the sequences of 3′UTRs and the conservation scores have been built. A local miRNA sequences database has also been created based on miRBase V12.0. The positive and negative data sets are obtained as follows: Positive data The positive data are obtained from miRecords, which records most up-to-data experimentally verified targets. Since our goal is to predict targets of mammalian miRNAs, we only focus on the records of human (1020 records), mouse (166 records), and rat (133 records). To ensure the fidelity of training data, all the miRNA sequences are mapped to miRBase and all binding site sequences are aligned to the corresponding 3′UTRs; site records with irresolvable problems are removed. For some miRNA-target pairs that share the same miRNA and 3′UTR region, only one record is retained. Finally, 324 miRNA-site pairs are obtained from 187 miRNA-target pairs, and 709 additional miRNA-target pairs are also extracted but without site information. Negative data Currently, the negative data are almost nonexistent in any annotated database. In this case, miRNA overexpression microarray data are consulted and we assume that negative targets are less likely to be underexpressed under miRNA over-expression. To provide diversity of the negative data, we collected 20 miRNA over-expression microarray data from NCBI Gene Expression Omnibus (Additional file 1 Table 4). To generate the high quality negative data, we only consider the most confident up-regulated genes by restricting the differential expression p value, the fold change, and consistency of the samples over time whenever available. After the negative miRNA-target pairs are derived, the negative miRNA-site pair data are generated by the site filter. In the end, we obtain 3542 negative miRNA-target pairs. (See Additional file 1 S.5 for detailed discussion) Feature selection and training For both Site-and UTR-SVM, RBF is chosen as the kernel function. 5-fold cross validation is carried out to train the parameters and select features for both SVMs. Due to the imbalance between the positive and negative data, the cost ratio factor is introduced in the SVMs. To measure the prediction accuracy, F score is adopted, which is a unified measurement of the prediction precision p and sensitive r, i.e. where p and r represent the prediction precision and sensitivity, respectively, and b ≥ 0 is a pre-specified weight that defines the relative importance between precision and sensitivity. In this case, b = 1, which puts equal importance on precision and sensitivity in defining the final prediction accuracy. In each round of cross validation, a sequential forward search algorithm is implemented for feature selection based on the ranked features by minimal redundancy maximal relevance (mRMR) algorithm [26]. In a nutshell, the mRMR algorithm is designed to choose a subset of features that have the highest relevance to the target class while maintaining the minimal redundancy. Particularly, the redundancy measures the correlation among features. Given a feature set S with m features {x i }, i=1, ..., m, the relevance D with the target class c is defined by and the redundancy R among features in S is given by where I(·,·) is the mutual information and defined by where p(x), p(y) and p(x, y) denote the marginal and jointly probability density functions, respectively. mRMR selects the minimal redundant maximal relevant feature set that maximizes the objective Φ(D, R) = D-R. The optimization can be achieved by a greedy search that iterates among features individually. A side product to the final optimal set is a feature rank list. Based on this rank list, a sequential forward search algorithm is applied to site-and UTR-SVM through cross validation on the training data to determine the 21 optimal site features ( Table 2) and 18 optimal UTR features (3). Based on these optimal features, SVMicrO can achieve the best prediction accuracy in terms of F score. At the same time of feature selection, a 2-D grid search is carried out to optimize the parameters of SVM. Specifically, two parameters need to be optimized including the penalty constant C of SVM and the parameter γ of the RBF kernel. Refer to [27] for a detailed discussion regarding the definition of these parameters. The entire cross validation is implemented by C language based on SVMlight v6.01 http://svmlight.joachims.org/. At the end of cross validation, the optimized SVM and the feature sets are recorded. Investigation of site features The 21 optimal site features are resulted from the sequential forward search feature selection applied to the site training data (Table 2). Close examination of these features lead the following observations. 7mer and 8mer seed matches are sufficient for miRNA site recognition Overall, seed region is clearly the most important region as 13 out of the 21 optimal features are related to the seed region. Among the different categories of features, seed match features account for a large portion of the optimal features with 6-mer seed match ranked at the top. The histograms of various seed match features are plotted in Additional file 1 Figure 3. The most discriminative feature 6mer seed match is present in around 80% true target sites but absent in 95% negative sites. All those seed type match features are also among the top ranked. These results echo the general belief that the seed matches are among the most important mechanisms for miRNA target recognition. However, the histograms of 7mer and 8mer matches are much less distinct between positive and negative data than those of the 6mer. Interestingly, almost no negative target sites possess these seed type match features; this implies that using these features on top of 6mer seed match reduces the false positive rate, although they are not as nearly sensitive as 6mer seed match. This fact is also demonstrated by the ROC curve of site-SVM in Additional file 1 Figure 10. From the perspective of miRNA site recognition mechanism, these data suggest that the 7mer and 8mer matches are unique to miRNA site recognition; however, miRNA does not always employ these mechanisms in target recognition. Conservation of 3′context region downstream of seed is of higher importance As expected, all conservation features including those of the seed, 5′ and 3′ context regions are important features. Unexpectedly, the conservation of the 3′ context region, or 10 nts downstream of the seed region, ranks the 2nd and plays more important roles than the seed conservation, which ranks only the 13th in the list. Many existing algorithms including TargetScan rely on seed conservation but none of them consider the conservation of 3′ context region. This finding points to the importance of downstream and upstream regions of the binding sites. Sequence motifs of Argonaute protein, ALG-1, binding have been revealed by cross-linking IP to preside in these regions. We hypothesize that the significance of the conservation features in the context regions is a result of Argonaute binding to UTR. Accessibility energy and seed binding energy are important features Energy features include accessibility energy and binding energies of the context region, seed, and total region. Among them, only accessibility energy and binding energy of seed regions are determined to be among the optimal features. It is not surprising to see the seed binding energy in this list, which recapitulates the importance of the seed region. However, accessibility energy feature is determined to be more important (ranked the 8 th ); this fact stresses that the 2 nd structure of potential site can considerably influence the ability of miRNA binding. Currently, only PITA assesses the accessibility energy. This finding advocates the inclusion of this feature for target prediction. Predicted site characteristics of miR-1 To further demonstrate the ability of site-SVM to reveal insights about miRNA binding, we apply the site-SVM to predict the binding sites of miR-1 in 75 validated positive targets obtained in miRecords. Although they are reported positive targets, no binding sites are reported. This is a common scenario especially prevalent for high throughput screening of miRNA targets and the question often concerns the binding characteristics of a miRNA. For the 75 miR-1 targets, a total of 155 sites (or 2.07 sites/UTR) are predicted by site-SVM. A binding matrix is constructed based on the predicted sites with the ij-th element being 1 if the j-th nucleotide of miR-1 is paired to the i-th site, and 0, otherwise. Based on the binding matrix, the empirical probability of binding can be obtained for each of the nucleotides in miR-1 sequence and a binding sequence logo is generated by the Tar-Logo program in the SVMicrO suite and plotted in Figure 3 to depict these binding probabilities. Figure 3 reveals two regions in miR-1 sequences that are likely to be responsible for binding to its target. The first region corresponds to the 6-mer seed from nt 2-6, and 100% probabilities in this region suggests miR-1 has perfect 6-mer pairing with every sites. The second region stretches from nt 12-20. The binding probabilities are not as high as those in the seed region but there is still a relative high chance of binding compared with the rest of the sequence. A close look into the secondary structure of binding at each predicted sites reveals that there is an average of 7 bulges and mismatches in this region of the site; the largest number of bulges and mismatches is 25 and the smallest is 0, indicating that miR-1 binds perfectly with some sites in this region. Investigation of UTR features There are totally 30 features for the UTR-SVM and the feature selection process chooses 18 features as the optimal UTR features (Table 3). We summarize some the observations in the following. UTR length is not a factor that influences miRNA target recognition Among the three groups of UTR features, the length of the UTR is left out by feature selection, suggesting an ill correlation between the length of UTR and miRNA target recognition. The histograms of UTR length (Additional file 1 Figure 11) also reflects this finding. The more accurate the sites are predicted, the more accurate the UTR prediction is The top Site-SVM score is the most important feature in UTR-SVM, and the higher the site score of a candidate site is, the higher probability the 3′UTR is predicted to be a real target. This observation agrees with those of other algorithms [9,28], which all accept a 3′UTR to be the target if one potential site has a score more than the cut-off score. Moreover, the third top feature is the number of positive sites. Note that the positive sites are the candidate sites that are predicted to be true by the Site-SVM. The histograms of the number of positive sites (Additional file 1 Figure 13) reveal that more than 80% of the negative targets are not predicted to have positive sites by Site-SVM, which partially explains the good specificity achieved by our Site-SVM. Maximum number of positive sites within 100 nts also plays an important role in the UTR-SVM, which is consistent with the fact that the effectiveness of binding sites will be enhanced if they are close (Additional file 1 Figure 12). All the 5 features about 8mer_A1 are important features and ranked within the top 15 in the UTR-SVM features. As shown in Additional file 1 Figure 13, almost no false targets have 8mer_A1 and 8mer_m1 seed matches. Performance evaluation of SVMicrO To investigate the performance of SVMicrO, the ROC performance are obtained from the cross-validation compared with several other popular target prediction algorithms including TargetScan, PITA, PicTar and miRanda (Figure 4-(a)). Except PITA, for which predictions are obtained by the provided algorithm program, the TPR and FPR of the other algorithms were calculated based on the predictions published on their website. Notice that the curves for the compared algorithms are partially in broken lines at various TPR. This is because for TargetScan (v5.1), PicTar (2006), MirTar-get2 (v3.0), PITA (v6), and miRanda (2008 Sept), the prediction scores for only a subset of mRNAs can be retrieved, while the scores of rest of mRNAs were assumed to be assigned by random predictors. There are two reasons for an mRNA to not receive a score. First, all these five existing algorithms apply different filters to remove unlikely sites/targets before proceeding to target prediction. mRNAs removed by the filter therefore receive no score and conceptually are predicted to be negative targets. Filtering does help reducing the search space for subsequent target prediction but at the price of reduced sensitivity; the reduction varies depending on the sensitivity of the filter employed by each algorithm. The filters of the four existing algorithms all rely heavily on the existence of the 6-mer seed match, and as discussed before, which result in around 20% reduction in sensitivity for the entire compared existing algorithms due to this filter. The no-score mRNAs for PITA are solely the result of the filter and it can be noted the sensitivity of the solid curve ends at around 80%. In fact, conceptually, the sensitivity of these algorithms should be capped at the sensitivity of the filter. In this case, assuming a random filter actually lends a performance advantage to these existing algorithms since it provides them a chance to increase beyond the capped sensitivity. In addition to the filter, TargetScan, PicTar and miRanda also apply a threshold to the prediction score to determine positive from negative. Since only positive targets are reported on the algorithm websites and we have neither access to the programs, nor the scores of the negative targets. Had the scores been available, the performance of the algorithm might be better or worse than a random predictor. However, we want to point out that regardless if these scores are available, SVMi-crO will have the best sensitivity among the group since the filter designed for SVMicrO has only about 4% reduction in filtering step. Overall, SVMicrO has the largest AUC (Area-Under-the-Curve) and its ROC almost wraps around the curves of all other algorithms. Although PITA has the second largest AUC, it has the worst performance at low false positive rate. In the low False Positive Rate (FPR) region, the algorithms except PITA have similar performance for FPR < 0.01 while MirTarget has a slight edge over the rest. For 0.01 < FPR <0.3, SVMicrO clearly has the best and much larger sensitivity. At a practical FPR value of 0.1, SVMicrO increases the sensitivity about 6% over miRanda and at least 17% over MirTarget, TargetScan, and PicTar. Figure 4-(b) depicts the zoom-in view for FPR < 0.023. In this region, SVMicrO is only inferior to mirTarget. Although mirTarget has better TPR at low FPR, SVMi-crO has much better sensitivity and obtains the best overall balance between TPR and FPR by achieving the largest AUC. To further reveal SVMicrO's performance at low FPR, we gauge the performance by evaluating the prediction precision. Precision represents the percentage of true targets among the predicted targets and can also be considered as the number of true targets among the given number of top ranked predictions. The precision of each algorithm in terms of the number of true targets among the different numbers of top ranked genes is Figure 5. For the top 100 predictions, SVMi-crO and MirTarget produce very similar true positives and achieve better precisions than the other algorithms, while after top 150 predictions SVMicrO starts to set itself apart from the rest by achieving much higher precision. In summary, the validation based on training data indicates that SVMicrO attains the largest AUC, achieves highest TPR especially for low FPR, and has consistently better precisions; these results demonstrate clear performance improvement over many popular algorithms. SVMicrO's overall better performance was further validated next. Test on the proteomics data To investigate the performance of SVMicrO on targets independent of the training data, we carried out the genome-wide target prediction for human miR-1, miR-16, miR-30a, miR-124, miR-155 and let-7b. Before prediction, the positive and negative targets of each miRNA were first removed from the training data and SVMicrO was retrained using the updated training data. The difficulty with validation is due to the lack of measurements of the ground truth. To mitigate the problem, we consulted the high throughput proteomics data in [29,30]. In these two papers, protein fold change due to the over-expression of specific miRNA were measured by stable-isotope-labelling-of-amino-acids-in culture (Additional file 1LAC) and quantified by LC/MS. Since protein inhibition is considered as primary mode of miRNA inhibition, protein down expression can be used as a utility for prediction validation. However, due to the limited coverage of LS/MS and relatively week intensity signals, no genes are declared targets definitively in the paper. Instead, it is only reasonable to assume that the larger down-fold a protein has, the more likely the corresponding gene is a true target. Due to the limitation of LC/MS coverage, only a subset of proteome is identified. Therefore, target prediction performance is only validated among these proteins. Figure 6 depicts the CFC (Cumulative Fold Change) for the top ranked 300 predictions of miR-124 and miR-1. Intuitively, CFC rewards higher confidence prediction with a drop and penalizes false prediction with a raise in the fold change. A better algorithm with higher precision and smaller false positive is expected to show faster drop in CFC. For miR-124, SVMicrO and TargetScan clearly set them apart from the rest, with SVMicrO performing slightly better up to top 100 and TargetScan having a slight edge up to top 200. At top 300, SVMicrO has clear advantage over the rest. For miR-1, SVMicrO is still among the better performing algorithms; instead of Tar-getScan, MirTarget and Pictar emerge to have competitive performance with SVMicrO. However, after top 200, SVMicrO achieves apparently much sharper drops than the others, compared with those of miR124. Moreover, same validation was carried also out for miR-16, miR-30a, miR-155 and let-7b and the significances of CFC prediction by each algorithm are assessed by random permutation (see Additional file 1 S.11). Next, we further investigate the consistency of the prediction performance for each algorithm using the results of 6 miRNAs. A better algorithm should have a cumulative sum curve with two characteristics: 1) it drops faster at the beginning, signifying a higher precision, and 2) it has the highest overall drop. Therefore, we calculate the average area between the cumulative sum curve and the horizontal axis as a measurement of the performance for each algorithm where c(x) denotes the function of cumulative sum curve and n denotes the number of the top ranked prediction. Intuitively, the smaller ℳ(n) is, the better the algorithm. Subsequently, a consistency measurement can be defined as the average value of ℳ(n) of the 6 miRNAs Table 4 and Table 5; they clearly shows SVMicrO is among the highest ranked algorithms at different n. The consistency measure ( ) n was subsequently calculated and shown in Table 6. SVMicrO is the top ranked at all n. Based on these results, it is reasonable to conclude that SVMicrO is the most consistent algorithm that provides among the best prediction performance. Test on the IP pull-down data Although the above experiments demonstrate consistently better performance achieved by SVMicrO, the utility of the evaluation on proteomic data might be limited by the coverage and potential noise in protein quantification. We therefore further validated the prediction of miR-124 and miR-1 on the IP pull-down data [31]. In these experiments, each miRNA was transfected in 293 cell and immunoprecipitation of the ARG-2 protein, an important component of the miRNA effector protein complexes, was carried out; the expression of genes recruited by ARG-2, or most likely the miRNA targets, was analyzed by microarray, and the target genes should be expressed in the microarray. 388 genes for miR-124 and 56 genes for miR-1 were determined in the end to be highly expressed at a stringent FDR level of 0.01. Although this technology has its own limitation, it nevertheless complements the proteomics data for prediction validation. Particularly, we treated the 388 and 56 highly expressed genes as the true targets of miR-124 and miR-1, respectively and investigated the ROC performance of different algorithms (Figure 7). Again, SVMi-crO has the overall best performance supported by the largest AUC. Very similar phenomenon as the performance tested using the training data can be observed; in the low FPR region, SVMicrO, MirTarget, and TargetScan have similar sensitivity. But for FPR > 0.01 the Figure 7 ROC curves for the predictions of miR-124 tested on IP pull-downs. The ROC curves were plotted based on 388 high confidence positive targets determined by IP pull down experiment. existing algorithms cannot achieve satisfied TPR due mainly to the poor performance of their adopted filters. We further investigated the prediction precision ( Figure 8). Clearly, SVMicrO attains the highest number of TPs for all the tested numbers of top ranked predictions, and thus has the best prediction precision. Figure 7 and 8 reflect the similar performance improvement of SVMi-crO over other algorithms as that demonstrated by the proteomics data. Same tests were also carried out for miR-1 (Figure 9 and 10). Again, similar conclusion can be drawn from these figures, which reinstate the consistent higher performance achieved by SVMicrO. In contrast, other algorithms do not show similar consistency; unlike the case of miR-124, TargetScan has worse performance than miRanda this time. Based on these results, we can conclude confidently that SVMicrO achieves improved prediction sensitivity, specificity, and precision than the existing algorithms. Discussion The improved performance of SVMicrO can be attributed to the following three factors. First, a comprehensive training data set including a large number of verified positive targets and derived negative targets for a diverse group of miRNAs was constructed. Compared with the training data constructed for other existing data driven algorithms, this training set contains by far the largest number of verified targets. As a result, this training set possesses a better coverage of different characteristics of miRNA target recognition than any other existing training datasets. Secondly, due to the increased size of training data, we could afford to develop more sophisticated prediction algorithms to better uncover the important targeting characteristics from data. SVMicrO algorithm has a unique two-stage structure, where the miRNA binding sites are first predicted, which is then followed by the prediction of the 3′UTR in the second stage. In each stage, the prediction is accomplished by a SVM algorithm. The performance improvement can be considered as a result of the sophisticated SVM algorithm to properly model not only binding sites but especially their relationship with 3′UTR. Thirdly, the improved performance is also an outcome of the site and UTR feature sets that, when combined, encompasses the largest extraction of features, surveying extensive characteristics of miRNA target binding. In addition, the adopted feature selection algorithm also ensures that only the optimal set of features is chosen for target prediction; this feature selection not only increases the computational efficiency by removing the correlated features but also ensures the best performance by eliminating the potential distortion and noise introduced by the non-effective features. Even though SVMicrO achieves the improved performance, it is evident from the evaluation that further improvement is needed. For a data driven algorithm, further improvement comes at the expense of increased quality and quantity of training data set. Compared with the number of potential genome-wide miRNA targets, our training data set is still relatively small in size and thus cannot cover all features of miRNA binding. Moreover, collecting representative negative targets is also a challenge. On the one hand, there is almost no reported, verified negative target. On the other hand, the number of negative targets is much larger than that of the positive targets, making the training data highly imbalanced. This created enormous difficulty for computational algorithms to learn the features of true targets. Improving the quality of training data especially the negative targets will be an important future research topic. Since SVMicrO is shown to be able to achieve robust performance on the current training data, it holds the promise to achieve continuing improvement whenever better training data that contain additional verified or high confidence positive targets and properly selected negative targets are available. Conclusions We proposed in this paper a new data driven algorithm, SVMicrO, for prediction of mammalian miRNA targets. Comprehensive validation of SVMicrO using a large training data set, the proteomics, and the IP pull down data has confirmed that SVMicrO can produce consistently better sensitivity, specificity, and precision than several popular existing algorithms. Genome-wide prediction of human miRNA using SVMicrO has been carried out. All the related materials including source code of SVMicrO and generation of miRNA binding logo and prediction results are available at http://compgenomics. utsa.edu/svmicro.html. Additional material Additional file 1: Supplementary Information.	2014-10-01T00:00:00.000Z	2010-09-22T00:00:00.000	{ "year": 2010, "sha1": "2ecf81addb32d061bb57f731dafe94cac65498aa", "oa_license": "CCBY", "oa_url": "https://bmcbioinformatics.biomedcentral.com/track/pdf/10.1186/1471-2105-11-476", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "28d9f177b0f962b14a9bd161e043f3870f59f9a7", "s2fieldsofstudy": [ "Biology", "Computer Science" ], "extfieldsofstudy": [ "Biology", "Medicine", "Computer Science" ] }
255500157	pes2o/s2orc	v3-fos-license	Codelivery of TGF-β1 and anti-miR-141 by PLGA microspheres inhibits progression of intervertebral disc degeneration Background Cervical and lumbar pain is usually caused by degeneration of the nucleus pulposus (NP). As a powerful therapeutic strategy, tissue engineering can effectively restore the normal biological properties of the spinal unit. Previous studies suggested that poly(lactic-co-glycolic acid) (PLGA) microspheres are effective carriers of cells and biomolecules in NP tissue engineering. This study aims to explore the therapeutic effect of PLGA microspheres coloaded with transforming growth factor-β1 (TGF-β1) and anti-miR-141 on intervertebral disc degeneration (IDD). Methods PLGA microspheres were characterized by scanning electron microscopy, a laser particle size analyzer, and laser confocal microscopy. The in vitro release rate of biomolecules from the microspheres was analyzed by reversed-phase high-performance liquid chromatography and agarose gel electrophoresis. The rat NP cells (NPCs) treated with the solutions released from microspheres for different lengths of time were assigned to a control group (Ctrl), an empty PLGA microsphere group (Mock microsphere, MS), a TGF-β1-loaded PLGA microsphere group (TMS), an anti-miR-141-loaded PLGA microsphere group (AMS), and an anti-miR-141 + TGF-β1-loaded PLGA microsphere group (ATMS). The proliferation and apoptosis of NPCs were observed by alamar blue and flow cytometry. The gene and protein expression of cartilage markers COL2A1 and ACAN were observed by RT-qPCR and Western blot. The rat model of IDD was established by tail puncture. Rats were divided into a control group (Ctrl), a mock operation group (Mock), a TGF-β1 microsphere group (TMS), an anti-miR-141 microsphere group (AMS), and an anti-miR-141 + TGF-β1 microsphere group (ATMS). The degree of rat tail IDD was assessed in each group through magnetic resonance imaging (MRI), safranin O-fast green staining, immunohistochemistry, and Western blotting. Results PLGA microspheres were stably coloaded and could sustainably release TGF-β1 and anti-miR-141. The results of in vitro cell experiments showed that the release solution of PLGA microspheres significantly enhanced the proliferation of NPCs without inducing their apoptosis and significantly upregulated cartilage markers in NPCs. The effect of microspheres was greater in the ATMS group than that in the TMS group and AMS group. In vivo experiments showed that IDD could be effectively inhibited and reversed by adding microspheres coloaded with TGF-β1 and/or anti-miR-141, and the effect was greatest in the ATMS group. Conclusion PLGA microspheres coloaded with TGF-β1 and anti-miR-141 can reverse IDD by inhibiting the degeneration of NPCs. Introduction Cervical and lumbar pain, a common orthopedic disease, seriously affects the work and life of patients, and intervertebral disc degeneration (IDD) is considered the initiating factor for the occurrence and progression of cervical and lumbar pain [1,2]. The intervertebral disc is composed of the annulus fibrosus, nucleus pulposus (NP), and cartilage endplates. NP degeneration is thought to be the earliest change of IDD as well as the most critical step in the IDD process [3,4]. The NP can keep its homeostatic balance under normal circumstances. In the case of NP degeneration, such homeostasis is broken, thus causing pathological changes that mainly manifest as a decrease in NP cells (NPCs), loss of proteoglycan in the extracellular matrix, and disorders of collagen structure. Ultimately, dehydration, revascularization, and reneuralization of NP occur, leading to loss of function and causing corresponding clinical symptoms [5,6]. Therefore, exploring the degeneration and regenerative repair of NP tissues is the core of the IDD research field [7]. Due to the poor self-repair ability of NP tissues, IDD is rarely self-reversable. An ideal therapy could be to promote the secretion of extracellular matrix by NPCs and inhibit relevant catabolic enzymes through the actions of biological agents [8]. Transforming growth factor-β1 (TGF-β1) has a key regulatory role in IDD [9,10]. Injecting TGF-β1 into the degenerated intervertebral disc can exert a definite repair effect on the intervertebral disc [11]. The in vivo use of TGF-β1 is restricted due to its easy degradation, short half-life, and difficulty of longterm stable expression [12]. miRNAs, a class of endogenous, non-coding, small RNA molecules, can regulate the stability and transcriptional expression of mRNAs. Abnormally expressed miR-NAs are important regulators in the pathophysiological process of IDD [13]. Applying miRNAs to the repair in the field of regenerative medicine is becoming a new research hotspot [14][15][16][17][18]. However, miRNAs are also prone to degradation in vivo, so carriers are required for their efficient and safe transfection. In recent years, growing attention has been paid to microparticle systems in tissue engineering. With excellent biocompatibility and a more uniform degradation rate, poly(lactic-co-glycolic acid) (PLGA) can meet different needs of tissue engineering if its composition ratio and molecular weight are adjusted [19]. In this study, injectable microspheres were constructed based on PLGA, and TGF-β1/anti-miR-141 complexes were loaded on PLGA microspheres for in situ delivery. The microspheres significantly inhibited the pathological process of NP degeneration and exerted a definite therapeutic effect on IDD. This study offers a new strategy for exploring the biological repair of IDD. Preparation of PLGA microspheres coloaded with TGF-β1 and anti-miR-141 PLGA microspheres coloaded with TGF-β1 and anti-miR-141 were prepared by the water-in-oil emulsification/solvent evaporation method [20]. First, 150 mg of PLGA was dissolved in 2 mL of dichloromethane and added to 0.2 mL of 0.5 mg TGF-β1 or anti-miR-141. The mixture was put in an ice bath and emulsified using a homogenizer (8000 rpm, 30 s). Then 0.1 mL of ammonium bicarbonate solution (100 mg/mL) was added to the system. The mixture was further emulsified using the homogenizer (4000 rpm, 30 s), and the emulsion was then added to 50 mL of polyvinyl alcohol solution, followed by homogenization (4000 rpm, 2 min). The resulting product was added to 50 mL of deionized water and magnetically stirred for more than 3 h to volatilize dichloromethane. Finally, after centrifugation (8000 rpm, 10 min), the PLGA microspheres were collected, washed with deionized water three times to remove the free nonsphered components, freeze-dried, and stored at − 20 °C for later use. The empty PLGA microspheres (Mock microsphere, MS), PLGA microspheres loaded with TGF-β1 (TMS) or anti-miR-141 alone (AMS), and PLGA microspheres coloaded with both TGF-β1 and anti-miR-141 (ATMS) were prepared in the above way. Characterization of PLGA microspheres The surface morphological characteristics of PLGA microspheres were observed by scanning electron microscopy under an accelerating voltage of 5 kV. The particle size of PLGA microsphere was measured using a laser particle size analyzer. To determine the distribution of TGF-β1 and anti-miR-141 in the microspheres, fluorescein-labeled PLGA microspheres were prepared and observed by laser confocal microscopy. In vitro biomolecule release assay The release status of TGF-β1 and anti-miR-141 from PLGA microspheres was detected at pH 7.4. First, PLGA microspheres were placed in a dialysis cassette and immersed phosphate-buffered saline (PBS) at pH 7.4 in a large beaker. A certain volume of sample was collected from the external solution chamber at predetermined intervals, and the PBS was replenished with the same volume of pH-adjusted fresh water. The level of TGF-β1 in the sample was detected by reversed-phase high-performance liquid chromatography. The presence or absence of anti-miR-141 in the sample was detected by agarose gel electrophoresis and quantified using a Quant-It ™ PicoGreen dsDNA kit. Cell proliferation assay The spine-derived NPCs of Sprague-Dawley rats were plated in a 96-well plate (9.0 × 10 3 /well) and cultured with Dulbecco's modified Eagle's medium (DMEM) in a 5% CO 2 incubator at 37 °C for 24 h. Twenty microliters of 5-, 10-, 20-, and 40-d PLGA microsphere release solution were added for 24-h treatment. Then the NPCs were incubated with 20 µL of DMEM medium containing 10% Alamar blue and 10% fetal bovine serum for 2 h, and the medium was aspirated (100 µL/well). The absorbance values at 570 nm and 600 nm were measured with a microplate reader. The corresponding sample without cells was inoculated as a blank control in each group. Finally, the relative cell count was determined: cell proliferation rate = absorbance experimental group /absorbance control group . Apoptosis assay The NPCs were plated in a six-well plate (2.5 × 10 5 / well) containing 2 mL of DMEM and incubated at 37 °C overnight. After the medium was discarded, 2 mL of medium containing 100 µL of 15-d microsphere release solution was added. Twenty-four hours later, the NPCs were digested with trypsin, centrifuged, and washed twice with PBS. Then they were resuspended with binding buffer according to the kit instructions and incubated with Annexin V-FITC and PI at room temperature away from light for 10 min. Finally, apoptosis was detected by a flow cytometer (BD Biosciences, Mountain View, USA). RT-qPCR Total RNA was extracted from cells by TRIzol (Invitrogen, USA). The ACAN and COL2A1 mRNA expression level was measured using the SYBR ® Premix Ex Taq ™ II kit (Takara, Japan). The PCR program was as follows: pre-denaturation at 95 °C for 30 s, and 50 cycles of denaturation at 95 °C for 5 s and annealing/extension at 60 °C for 25 s. Relative quantification was performed by the 2 −△△Ct method, with GAPDH as an internal control. The assay was repeated three times. The primer sequences are shown in Table 1. Western blotting The cell or tissue samples were collected, washed twice with PBS, and lysed with radioimmunoprecipitation assay lysis buffer for 2 h on ice, with protease inhibitors added. The lysate was centrifuged at 12,000 rpm for 10 min, and the supernatant was harvested. The proteins were quantified using a bicinchoninic acid kit. Then an equal amount of protein was separated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis, and the protein of a specific molecular weight was electro-transferred onto a polyvinylidine fluoride membrane. The membrane was blocked with PBS containing 5% skim milk powder and 1% Tween-20 (PBST) at room temperature for 1 h, then incubated with primary antibodies (ACAN, COL2A1) at 4 °C overnight. After washing three times with PBST, the membrane was incubated with horseradish peroxidase-labeled secondary antibodies at room temperature for 1 h. Finally, specific protein bands were detected by chemiluminescence solution. Immunofluorescence assay The NPCs were fixed with 4% paraformaldehyde, permeabilized with 0.3% Triton X-100, and blocked with 2% sheep serum, followed by incubation with primary antibodies (ACAN, COL2A1) (1:1000, Abcam, USA) at 4 °C overnight. Following incubation with fluorescent secondary antibodies and DAPI, the NPCs were washed with PBS, and the change in the target protein content was observed under a laser confocal microscope (Leica, Germany). Animal experiments Upon the approval by the Laboratory Animal Committee of the support unit, 40 8-week-old Sprague-Dawley rats were randomly divided into a control group (Ctrl), an operation group (Mock), a TGF-β1 microsphere group (TMS), an anti-miR-141 microsphere group (AMS), and an anti-miR-141 + TGF-β1 microsphere group (ATMS), with eight rats in each group. No operation was performed in the Ctrl group. In the Mock, TMS, AMS, and ATMS groups, the rats were anesthetized with pentobarbital (40-50 mg/kg) and fixed in a supine position. After hair shaving, the abdomen was cleaned, disinfected with iodophor, and draped. Then the rat tail intervertebral disc was punctured till the central NP with a 21-G micropuncture needle (Hamilton, Bonaduz, Switzerland) under X-ray fluoroscopy. The needle was rotated 360° and then withdrawn. The puncture site was sterilized with iodophor, and 50,000 U of penicillin was injected intramuscularly to prevent infection. In the TMS, AMS, and ATMS groups, the rats were additionally injected with TMS, AMS, and ATMS in the operative intervertebral space by a 33G micropuncture needle, after which the puncture site was also sterilized with iodophor and 50,000 U of penicillin was injected intramuscularly to prevent infection. The IDD status was observed by lumbar MRI plain scan at 2 months after modeling and was evaluated by Pfirrmann grading in each group. After all rats were sacrificed, the tail intervertebral disc samples were harvested. The morphological changes in the operative segment of intervertebral disc were observed by safranin O-fast green staining. The expression of COL2A1 in the NP tissues of the operative segment was detected by immunohistochemistry and Western blotting. The intervertebral disc was stained by histologically, and the results were graded and compared according to the criteria of Masuda et al. [21]. Statistical analysis SPSS18.0 software was used for statistical analysis. Normally distributed measurement data are described as mean ± standard deviation. The data were compared for significance using the independent-samples t-test between two groups and using one-way analysis of variance between more than two groups. Before the analysis of variance, the homogeneity test of variance was routinely performed. Tukey's highly significant difference test was used to compare the significance of differences in the case of homogeneity of variances, while Dunnett's T3 test was used in the case of heterogeneity of variances. P < 0.05 was considered statistically significant. Characterization of PLGA microspheres and the loading and release of biomolecules Scanning electron microscopy showed that the PLGA microspheres had a uniform spherical shape and did not adhere to each other. The mean particle size was 100 µm (20-170 µm) (Fig. 1A, B). To directly verify the coloading of TGF-β1 and anti-miR-141 in PLGA microspheres, red TGF-β1 fluorescence and green FITC fluorescence (FITC-labeled anti-miR-141) in PLGA microspheres were observed by laser confocal microscopy. In the overlap images, PLGA microspheres displayed yellow fluorescence, suggesting that both TGF-β1 and anti-miR-141 were uniformly dispersed in the microspheres (Fig. 1C). To determine the release efficiency of TGF-β1 and anti-miR-141 from PLGA microspheres, the release solution of single-loaded and coloaded PLGA microspheres was detected for 40 consecutive days. As shown in Fig. 1D, both TGF-β1 and anti-miR-141 showed a slow release trend, and the biomolecule release rate was roughly the same between the single-loaded and coloaded PLGA microspheres. The release efficiency of TGF-β1 and anti-miR-141 reached a plateau at approximately 15 d, so the 15-d release solution of PLGA microspheres was selected for later experiments. Effect of PLGA microsphere release solution on NPCs The effects of 5-, 10-, 20-, and 40-d release solution of PLGA microspheres on the proliferation of NPCs were observed. Compared with that in the MS group, the proliferation of NPCs was enhanced in the TMS, AMS, and ATMS groups, especially the ATMS group ( Fig. 2A). Flow cytometry showed that the apoptosis did not differ between the MS, TMS, AMS, and ATMS groups (Fig. 2B). Furthermore, the gene and protein expression levels of cartilage markers (ACAN, COL2A1) were higher in the AMS, TMS, and ATMS groups (AMS < TMS < ATMS) compared with those in the MS group ( Fig. 2C-G). The above results demonstrated that PLGA microspheres coloaded with TGF-β1 and anti-miR-141 can enhance the viability of NPCs and facilitate the protein-secreting capacity of the extracellular matrix of NPCs, and that combination of TGF-β1 and anti-miR-141 have the strongest effect. In vivo repair effect of PLGA microspheres on IDD Whether PLGA microspheres coloaded with TGF-β1 and anti-miR-141 could promote the repair of IDD by enhancing the biological properties of NPCs was tested in the rat model of IDD established by tail puncture (Fig. 3A). MRI at 2 months after modeling showed that the intervertebral disc signal was normal in the Ctrl group, while it became significantly weakened in the Mock group due to the puncture damage, with an obvious loss of intervertebral height and the most serious degeneration among the groups. Compared with those in the Mock group, the intervertebral disc signal was only slightly decreased, with a smaller loss of intervertebral height and significantly mild degeneration, in the TMS, AMS, and ATMS groups (Fig. 3B, C). Safranin O-fast green staining showed that the NP tissues of the operative segment were evenly stained red and had an intact structure in the Ctrl group. The NP tissues of the operative segment were dehydrated and disappeared in the Mock group. In the TMS, AMS, and ATMS groups, more NPCs could still be seen, and there were still many red-stained extracellular matrix components in the operative segment, especially in the ATMS group (Fig. 4A, B). Immunohistochemistry and Western blotting showed that the levels of ACAN and COL2A1 in NPCs were significantly lower in the Mock group than in the Ctrl group, while they were significantly higher in the TMS, AMS, and ATMS groups than the Mock group, especially in the ATMS group (Fig. 4C-F). Discussion Targeted delivery via microparticle systems is highly promising for the treatment of IDD [22]. First, the microparticle system can raise the chemical stability of drugs. For example, some drugs can be degraded in vivo by enzymes, such as siRNAs by RNases and proteins in the stomach by pepsin or pancreatin. This degradation can be ameliorated if the drug or the corresponding biodegradable component is loaded into a microparticle system, thereby prolonging its lifetime and increasing its efficiency. Second, the properly designed microparticle system can help improve the penetration and biodistribution of biomolecules targeting the diseased tissue. Highly biocompatible PLGA-based microspheres have been widely studied as carriers [23]. PLGA microspheres of small-molecule drugs, proteins, and plasmid DNA for in situ therapy have advantages over direct injection of drugs or gene transfection agents at the lesion site, according to previous studies on IDD [24]. Moreover, PLGA randomly aggregated by two monomers (lactic acid and glycolic acid) is a degradable functional polymer organic compound, characterized by good biocompatibility, nontoxicity, and good capsule-forming and film-forming properties [25]. The occurrence and progression of IDD result from the dysregulation of gene expression and biological factors related to various signaling pathways in NPCs. Therefore, it is necessary to promote the development of therapies combining genes and biological factors [26,27]. Research in this field used to focus on the separate delivery of drugs or genes by different carriers, but it is difficult to deliver the biological factors and genes to the same target to exert a synergistic effect in this way. Codelivery using a properly loaded and efficiently releasing carrier has significant advantages over separate delivery [28][29][30]. The biological factors and genes coloaded can be simultaneously taken up by target cells, so that the two can exert a synergistic effect. Further work on novel codelivery systems is necessary for targeted treatment of IDD [31,32]. TGF-β1 is a pleiotropic cytokine that can regulate proliferation, apoptosis, differentiation, inflammation, extracellular matrix synthesis, and developmental processes of NPCs [33,34]. TGF-β1 is thought to reduce catabolism and inflammation by promoting matrix synthesis and inhibiting apoptosis, thereby facilitating chondrogenesis and protecting the intervertebral disc [35]. Mounting evidence has shown that miR-141 is a key player in the pathogenesis of IDD [36,37]. Mechanistically, the NPC apoptosis induced by the crosstalk between miR-141 and SIRT1/NF-κB pathway is a key determinant of IDD, indicating that miR-141 is an important target for therapeutic intervention in IDD [38]. In this study, PLGA-based microspheres coloaded with TGF-β1 and anti-miR-141 were synthesized as carriers to repair the degenerated intervertebral disc by inhibiting NPC apoptosis and promoting NPC extracellular matrix synthesis. Finally, PLGA microspheres capable of codelivering TGF-β1 and anti-miR-141 were successfully prepared. The microspheres possessed good loading efficiency and a good release effect, so they stably and sustainably released TGF-β1 and anti-miR-141. To test whether the microsphere release solution could enhance the biological functions of NPCs, the effects of single-loaded and coloaded microspheres on NPCs were compared between groups. Compared with that in the MS group, the release solution of single-loaded and coloaded microspheres enhanced the proliferation of NPCs as well as the extracellular secretion function to varying degrees, without inducing apoptosis. What is noteworthy is that the NPCs in the ATMS group displayed stronger proliferation capacity and extracellular secretion function than those in the TMS and AMS groups. This confirmed the synergistic effect of TGF-β1 and anti-miR-141 in vivo. Consistent results were obtained in the animal experiment: IDD was relieved better in the TMS, AMS, and ATMS groups than the Mock group, and the intervertebral disc in the ATMS group had the best MRI signal intensity and histological structure. Conclusion PLGA microspheres coloaded with TGF-β1 and anti-miR-141 can effectively suppress NPC apoptosis and facilitate NPC proliferation and matrix secretion by sustainably and stably releasing prochondrogenic growth factors and therapeutic genes, thereby relieving IDD.	2023-01-08T05:09:05.172Z	2023-01-06T00:00:00.000	{ "year": 2023, "sha1": "9b8e7c2749d2ddc1aebc739e1d338e1fb89251e1", "oa_license": "CCBY", "oa_url": null, "oa_status": null, "pdf_src": "PubMedCentral", "pdf_hash": "9b8e7c2749d2ddc1aebc739e1d338e1fb89251e1", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
190427039	pes2o/s2orc	v3-fos-license	COPD: misuse of inhaler devices in clinical practice Background and objectives: Inhalers mishandling remain an important clinical issue worldwide. The aim of this study was to evaluate inhalation technique in stable COPD out-patients. The variables under study were type of inhaler device (ID), patients’ preference for an inhaler, number of IDs used by each patient, beliefs about inhaler medication and some demographic, clinical and functional patients' characteristics. We aim to assess how they are related to inhalation technique. Methods: A cross-sectional study was conducted in a hospital outpatient respiratory care. COPD patients over 40 years old, diagnosed according to GOLD criteria, and using IDs were included consecutively. The Beliefs about Medicines Questionnaire (BMQ), a demographic and a clinical survey were applied. The number of IDs used by each patient and the patients’ preference for some IDs were recorded. Patients were asked to demonstrate the use of their prescribed inhalation devices, and inhaler technique was assessed by using previously defined checklists, including essential steps and critical errors. A statistics analysis was then performed. Results: We studied 300 subjects performing a total of 521 inhalation manoeuvers with 10 different IDs. At least one step incorrectly performed was found in 48.2% of demonstrations and in 29.9% critical errors were observed. Misuse was related to priming/loading in 6.9%, to inhalation manoeuver in 13.1% and to both in 10%. There was a statistically significant association between critical errors and type of ID (P<0.001). No significant relationship was found between correct performance of key manoeuvers and patients’ preference or number of inhalers used per patient. Misuse due to critical errors was observed in 39.3% of patients and was positively related to female gender, age ≥65, lower education level and lower socioeconomic status (higher Graffar classification score), but not to patients’ clinical or functional characteristics. In the sub-group of patients presenting critical errors when using IDs, there was a statistically significant inverse association between BMQ Necessity score and number of critical errors. Conclusions: Inhalers mishandling remains disappointingly common. A good inhalation technique depends on the type of ID, and failure of inhalation manoeuver was the main cause of ID misuse. It was not associated to multiple inhalers’ use nor to patient’s preference, but to the patient’s beliefs about the necessity to use them. Elderly patients, women and those with lower education level or lower socioeconomic status demonstrate a worse inhalation technique. Background and objectives: Inhalers mishandling remain an important clinical issue worldwide. The aim of this study was to evaluate inhalation technique in stable COPD out-patients. The variables under study were type of inhaler device (ID), patients' preference for an inhaler, number of IDs used by each patient, beliefs about inhaler medication and some demographic, clinical and functional patients' characteristics. We aim to assess how they are related to inhalation technique. Methods: A cross-sectional study was conducted in a hospital outpatient respiratory care. COPD patients over 40 years old, diagnosed according to GOLD criteria, and using IDs were included consecutively. The Beliefs about Medicines Questionnaire (BMQ), a demographic and a clinical survey were applied. The number of IDs used by each patient and the patients' preference for some IDs were recorded. Patients were asked to demonstrate the use of their prescribed inhalation devices, and inhaler technique was assessed by using previously defined checklists, including essential steps and critical errors. A statistics analysis was then performed. Results: We studied 300 subjects performing a total of 521 inhalation manoeuvers with 10 different IDs. At least one step incorrectly performed was found in 48.2% of demonstrations and in 29.9% critical errors were observed. Misuse was related to priming/loading in 6.9%, to inhalation manoeuver in 13.1% and to both in 10%. There was a statistically significant association between critical errors and type of ID (P<0.001). No significant relationship was found between correct performance of key manoeuvers and patients' preference or number of inhalers used per patient. Misuse due to critical errors was observed in 39.3% of patients and was positively related to female gender, age ≥65, lower education level and lower socioeconomic status (higher Graffar classification score), but not to patients' clinical or functional characteristics. In the sub-group of patients presenting critical errors when using IDs, there was a statistically significant inverse association between BMQ Necessity score and number of critical errors. Conclusions: Inhalers mishandling remains disappointingly common. A good inhalation technique depends on the type of ID, and failure of inhalation manoeuver was the main cause of ID misuse. It was not associated to multiple inhalers' use nor to patient's preference, but to the patient's beliefs about the necessity to use them. Elderly patients, women and those with lower education level or lower socioeconomic status demonstrate a worse inhalation technique. Keywords: COPD; Inhalation technique; Inhaler devices Background and objectives COPD currently represents one of the most significant health problems at an international level. Inhaled medication is the mainstay of COPD management, and therapeutic success depends on the maintenance of a correct inhalation technique. There is a growing evidence concerning inhalers misuse as a common problem worldwide. 1,2 It can be associated with increased rate of severe COPD exacerbations (ECOPD), 3 but the impact of inhalers misuse on COPD outcomes remains currently unknown. In a country where patients have a good access to health-care services and to effective treatment, any development in treatment outcomes must address the improvement of inhalation technique. This can be one of the main cost/benefit measures improving healthcare in COPD patients. In 1965, Saunders published in the BMJ the first paper describing the misuse of inhaled medication. 4 In fact, misuse of inhaler devices (IDs) in obstructive airway diseases is an old problem, and has not improved over the past 40 years, despite the progressive technical improvement of IDs. 5 Currently, up to 94% of patients have demonstrated inhalers' mishandling in clinical studies. 6 Teaching and maintaining a correct inhalation technique has a positive impact on disease and patient outcomes. 7 It should remain a constant concern of any health professional involved in the management of COPD patients. 8,9 The knowledge of difficulties and barriers that hinder a correct inhalation technique is of paramount importance to develop any educational intervention regarding the correct use of IDs. However, this can be a very difficult task. Assessment of inhalation technique is complex and always somewhat subjective, and consensus is lacking between researchers regarding the definition of critical errors and standardization of inhaler technique checklists. 7 The aim of this study was to evaluate the inhalation technique in stable COPD outpatients, because there was a gap of information in Portuguese population. We intended assessing whether the type of ID, the preference or number of IDs used by each patient, the demographic, clinical or functional characteristics of patients, and the patients' beliefs about inhaled medication are associated with a correct inhalation technique. This last aspect was never studied, to the best of our knowledge. Materials and methods A cross-sectional study was conducted in the outpatient respiratory care of Guimarães hospital, between March 2016 and May 2017. COPD patients over 40 years old diagnosed according to GOLD criteria, without acute exacerbations for >4 weeks and using inhalation devices were consecutively included. Exclusion criteria were refusal to participate and an inability to understand simple questionnaires. No participants were at the same time enrolled in another different studies, and all gave their written informed consent. The study was approved by the Guimarães Hospital Ethics Committee, the Research Ethics Committee of Minho's University and by the Portuguese Data Protection Agency. We followed the STROBE guidelines for reporting observational studies. 10 A survey of demographic and clinical data, the Graffar Social Classification, 11 validated for use in Portuguese population and the cross-cultural adaptation of the Beliefs about Medicines Questionnaire (BMQ-specific) into Portuguese were applied. 12 Patients' beliefs about inhaled medication can influence adherence to medication but we suspect they can also motivate patients to learn the correct use of inhalers. BMQ is an 11-item questionnaire with a five-item Necessity scale and a six-item Concern scale. The Necessity and the Concern scales assess, respectively, the beliefs about the necessity and the beliefs about concerns related to side-effects, dependence and toxicity of the medication. Answered on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree), the points are summed, and total scores range from 5 to 25 in the Necessity scale and 6 to 30 in the Concern scale. It was clearly detailed that questions referred to the inhaled medication, and not to the device itself. The number of ECOPD referred in the last year was evaluated. We defined ECOPD according to GOLD as an acute worsening of respiratory symptoms that results in additional therapy, 13 but also requiring an unplanned medical visit, because patients have difficulty in remembering unreported exacerbations. Evaluation of symptoms was done using the Portuguese versions of the COPD assessment Test (CAT) and the Medical Research Council Dyspnoea Questionnaire (mMRC). All subjects performed spirometries according to ERS/ATS criteria, and referenced according to the Global Lung Function Initiative predict equations (GLI 2012). 14,15 Inhalation technique was assessed by using previous defined checklists, as presented in Tables 1 and 2. They were developed according to the instructions provided by the manufactures and to previous literature, 16 and included essential steps and critical errors. Errors are considered critical when they can substantially affect drug delivery to the lungs, and are related to priming/loading or inhalation manoeuver. The definition of critical errors when using inhalers is of great importance. However, there is currently a lack of consensus on their definition, deserving discussion. 7 Participants were asked to demonstrate the use of their prescribed ID just as they do it at home, but demonstrations were done with inhalers containing placebo medications. Assessment of patients handling of IDs, by recording the correct steps and critical errors, were done by a single trained senior pulmonologist, to avoid interobserver variability. The number of IDs simultaneously in use by each patient was recorded. Patients using two or more inhalers were inquired for device preference, and invited to justify their answer, clearly stated that the question is related only to inhalers' aspects. This was an open question. Answers were then collected in 5 groups: more practical to use, easier to use, ID characteristics, accustomed to using, and others. Because of the difficulty to distinguish between "being more practical" and "easier", these two answers were later analyzed together. The variables under study, evaluated for potential association with incorrect inhalation technique, were the type of ID, patients' preference, use of multiple devices, beliefs about inhaled medication and some patients' characteristics: age, gender, monthly income, social classification, education level, CAT score, mMRC grade, number of ECOPD referred in the last year, FEV 1 %, and GOLD 2017 stage and classification. Because patients were using 1-4 IDs at the same time, to assess inhalation technique three different outcomes, exhibiting different aspects of the same reality were defined. Correct use rate (CUR) was defined as the ratio between the sum of the number of correct steps in all IDs used by each patient and the total possible number of steps. Critical errors rate (CER) was defined as the ratio between the sum of the number of critical errors done with all the IDs and the total number of possible critical errors. The patients' ability (ABL) to use inhalers was defined as the ratio between the number of IDs without any critical errors and the total number of IDs used. Data were compiled in Microsoft Office Excel 2013 (Microsoft Corporation, Redmond, WA, USA). Statistical analysis Statistical analysis was performed with IBM SPSS Statistics for Windows, Version 23.0. (IBM Corp., Armonk, NY, USA). Group differences in the sample were analyzed with Student's T test and Chi-square independence test. Multivariate ANOVA modeling was used to identify differences among sample groups in: CUR, CER and the patients' ability to use inhalers. Spearman's correlation was used to explore the association between patients' beliefs (BMQ) and Table 2 Critical errors in different IDs 1 Aeroliser ® , Breezhaler ® , and Handihaler ® : failure to insert the capsule, failure to press and release buttons, powder remaining in the capsule after inhalation. 2 Diskus ® : failure to open the cover, to slide the lever until it clicks, or not keeping inhaler horizontally. 3 Ellipta ® : failure to slide cover down until a click is heard or block air vent with fingers. 4 Genuair ® : failure to remove the cap, to press and release the button until the control window has changed to green, not holding inhaler horizontally, and not changing control window to red after inhalation. 5 pMDI: failure to remove cap, not shaking the inhaler (suspensions only), not holding the inhaler in the upright position, poorly synchronized hand actuation and inhalation (except using a spacer), inhalation through the nose, actuation against teeth, lips or tongue. 8 Turbuhaler ® : failure to remove cover, to hold the inhaler upright when twisting the grip (tolerance ±45º) until a click is heard. Sample characteristics We studied 300 subjects performing a total of 521 inhalation manoeuvers with 10 different IDs in a total of 69 (13.2%) pMDI, 132 (25.3%) single-dose inhalers (sDPI), 239 (45.8%) multiple dose inhalers (mDPI) and 81 (15.5%) SMI-Respimat ® . Only 12 pMDIs (17.4%) were used together with a spacer. All participants referred using inhalers for over a month, regardless of having received prior instructions for the correct inhalers' use. The main demographic, clinical and functional characteristics of patients are described in Table 3. 89.7% of patients referred living with the family, but 23.2% of women and 5.6% of men (P<0.001) referred living alone. Tobacco smoking was the most common exposure identified and 16.4% of subjects were current smokers. Participants were currently using one (38.9%), two (49.5%), three (10.6%) or four inhalers (1%). The main reasons for an ID preference were the ease of use (65.9%), ID characteristics (24.6%) and being accustomed to using (2.9%). Devices' characteristics were frequently reported as pleasant because of the small size and feedback provided by some inhalers. Powder's bad taste and a significant effort needed during inhalation were frequently referred as unpleasant. Inhalation misuse by IDs At least one incorrect step was found in 48.2% of inhalations. In 156 (29.9%) demonstrations, critical errors were observed: 53.6% with pMDIs, 24.2% with sDPIs, 26.8% with mDPIs and 28.4% with the soft-mist inhaler. There was a statistically significant association between critical errors and the type of ID (P<0.001). Misuse was related to priming or loading in 6.9%, to inhalation manoeuver in 13.1% and to both in 10%. In mDPI group, critical errors ranged from 16.1% with Ellipta ® to 35.1% with Turbohaler ® . No significant relationship was found between correct performance of key maneuvers and patient's preference: 26.3% of preferred and 28.1% of non-preferred IDs presented incorrect use (P=0.120). No significant relationship was found between the correct performance of key manoeuvers and the number of inhalers currently used by patients (the incorrect use was 31.6% with one inhaler, 24.8% with two, 33.4% with three and 66.7% with four inhalers, P=0.739). The relationship between IDs and critical errors/inhalation technique is presented in Table 4. Participants showed more difficulty in inhalation maneuver with pMDI, in priming when using a Turbuhaler ® and in both loading and inhalation maneuver when using the Handihaler ® . Inhalation misuse by patients' characteristics Misuse due to critical errors was observed in 52.1% of women and 35.4% of men, in a total of 39.3% patients. The statistically significant relationship between the studied outcomes, as referred in the section on materials and methods, and patients' characteristics are presented in Table 5. A statistically significant association was also found between CUR and age <65 (<65 years: mean score=0.8264, ≥65 years: mean score=0.7581, P=0.026). No statistically significant association was found between the studied outcomes and ECOPD, CAT score, mMRC grade, FEV 1 % and GOLD stage or classification. Discussion In our study, the majority of patients was treated by pulmonologists for COPD, even if they were simultaneously being cared by their family physician for other co-morbidities. Even so, inhalers' misuse was disappointingly common. This is consistent with other published studies. In a recent study on 2,935 patients, handling errors were observed in over 50% of demonstrations. 3 Different studies reported different rates of misuse, using different methods and studying different populations. 17 A good inhalation technique depends on the type of ID. Critical errors were observed regardless of the inhaler used, but their proportion is different according to ID. Failure of inhalation maneuver was the main cause of ID misuse. However, evaluation of inhalation maneuver was the most difficult and subjective step to evaluate, especially when using a DPI, because we do not have measure inhalation parameters. This could favor some DPIs, mainly the Turbuhaler ® , Diskus ® , Ellipta ® and Spiromax ® , for which poor inhalation flow cannot be evidenced by lack of changing of control window's color or by powder remaining in the device. Therefore, there is a possibly a greater rate of misuse than we actually described. In patients needing >1 ID, whenever possible, we suggest the prescription of inhalers of the same type. Although differences were found between all types of inhalers, misuse related to inhalation manoeuver when using a pMDI was the most common reason for any inhaler misuse. This can be related to ID characteristics or to insufficient teaching or training. This group of inhalers is usually considered the most difficult to use, despite requiring a minimum inspiratory flow for correct airway deposition. 18 Poor coordination and failure to inhale slowly and deeply are well-known causes of pMDI misuse. Their use together with a spacer, although somewhat unpopular in practice, has been recommended in certain circumstances, and may overcome some difficulties. 19 In our study, only a small number of patients used pMDIs together with a spacer, which is insufficient to draw any conclusions. The softmist inhaler represents a more recent category of a liquid ID that can lead to high lung depositions in patients with low inspiratory flow. In our study, they represent 15.5% of the IDs evaluated, with a rate of misuse significantly lower than pMDI and comparable to DPIs. They can be a good therapeutic option, limited by the reduced number of drugs available on a SMI. DPIs were the most popular inhalers used in our study, probably because they deliver a large range of different drugs. The sub-group of sDPI presented the better rates of correct use, and Ellipta ® was the mDPI easier to use. In our survey, a good inhaler technique was not associated with patient's preference nor to multiple inhalers' use, unlike in previous studies. 20,21 A previous study referred the importance of patients' confidence on the use of their inhalers to improve treatment adherence. 22 In our study, in the group of patients presenting critical errors when using their inhalers, the patients that believe less in the need for medication were more prone to make a higher number of critical errors. This is a new information and needs to be interpreted with caution. This is a cross-sectional study and there is an interaction effect between education level and response to BMQ. Some patient's characteristics are significantly related to misuse of IDs. Being older, having lower education level or lower socio-economic status was significantly related to an incorrect inhalation technique, as in other published studies. 6 Therefore, the large rate of inhalers misuse in our study is not surprising, especially after considering the overall education level and socioeconomic background of the studied population. As in previous published studies, 6,23 females are also more prone to inhalers misuse. However, there is an interaction effect between gender and education level in the sample. Possibly, the effect observed in gender difference (ie, higher proportion of female with critical errors) may be associated with lower education level in the women subgroup. Women also exhibit a lower socioeconomic status and are more prone to live alone, without any help from family members in the use of IDs. All of this can justify the higher proportion of IDs misuse in female gender. We suggest that educational interventions could be reinforced in such patients. In our study, patients' clinical or functional characteristics were not related to inhalers' misuse. Again, this is consistent with other published studies. 18 Although it would be expectable, the impact of inhalers misuse on ECOPD is currently unknown, and probably difficult to be proved, given the small number of studies reporting significant association between critical errors and COPD outcomes. 7 We also failed to demonstrate a significant association between inhalers' misuse and COPD acute exacerbations. Both COPD and aging process reduces inspiratory muscle function and the ability to generate sufficient inspiratory flow to allow significantly lung deposition. This is usually the limiting factor for the proper use of an inhaler, but it was not objectively measured in the present study. We understand that it could, by some extent, explain the lack of association found between inhalers' misuse and COPD acute exacerbations. However, the aim of this study was the patient's ability and knowledge to use IDs, and not the patient's capacity to generate sufficient inspiratory flow. Our study was conducted in a 'real word' setting, but in a single institution. Patients were mostly treated by pulmonologists and were recruited sequentially, so we cannot exclude selection bias. This may limit the generalization of results to other populations. 24 However, we have studied a significant number of patients using a wide range of IDs. Subjects were recruited without prior notification and were invited to demonstrate the use of their prescribed ID by using placebo inhalers, to facilitate the evaluation of the inhalation technique. Assessing inhalation technique by using checklists is somewhat subjective, but currently, there is no 'best' method. Some authors suggest the need of studies based on generally accepted checklists of maneuvers affecting drug delivery, to facilitate comparisons of results. 5 Previous published studies measured inhaler technique by counting the number of correct steps, counting the number of critical errors or essential steps, or classifying the quality of the inhaler technique. 6,20,[25][26][27] Our checklists were designed with full of steps and critical errors, to minimize subjective evaluation, and inhalation technique was evaluated by a single trained pulmonologist, to avoid inter-observer variability. The definition of critical errors is of great importance, because they are likely to significantly decrease delivery of medication to the lungs, impairing health-related outcomes. Nonetheless, there is currently a wide variation and a lack of consensus on their definition. 7 Therefore, our defined checklist of steps for a correct inhalation technique and the choice of critical errors, although based on previous literature, deserves discussion. Very few original studies have been developed in Portuguese populations of COPD patients, regarding misuse of IDs, 23,28 and to the best of our knowledge, this is the first study relating inhalers misuse to the patients' beliefs about inhaled medication. A correct understanding of the relationship between inhalers and some specific patient-related characteristics can be useful in clinical practice, adding value in a resource-limited community. Nevertheless, matching the adequate medications with the adequate devices for each patient will always be a challenging issue. Conclusions Despite significant developments in device engineering, inhalers mishandling remains an important clinical problem. A good inhalation technique depends on the type of ID. Some inhalers are more prone to critical errors, and different inhalers are susceptible to different types of critical errors. Elderly patients, women, patients with lower education level, lower socioeconomic status or less believers in the need of inhaled medication demonstrate a worse inhalation technique. Therefore, any educational intervention should be reinforced in patients with these characteristics. This knowledge can be important in clinical practice by helping the choice of IDs, in predicting difficulties and in planning educational interventions regarding the correct use of inhalers.	2019-06-14T13:46:42.611Z	2019-05-30T00:00:00.000	{ "year": 2019, "sha1": "f4e56c0716ad2f9180c2cb15db5b711c59bb601e", "oa_license": "CCBYNC", "oa_url": "https://www.dovepress.com/getfile.php?fileID=50245", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "7ecc83d4d5013accc3c4d482d3e36bb93d1e1d92", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
184485585	pes2o/s2orc	v3-fos-license	Arbuscular Mycorrhizal Symbiosis Affects Plant Immunity to Viral Infection and Accumulation Arbuscular mycorrhizal (AM) fungi, as root symbionts of most terrestrial plants, improve plant growth and fitness. In addition to the improved plant nutritional status, the physiological changes that trigger metabolic changes in the root via AM fungi can also increase the host ability to overcome biotic and abiotic stresses. Plant viruses are one of the important limiting factors for the commercial cultivation of various crops. The effect of AM fungi on viral infection is variable, and considerable attention is focused on shoot virus infection. This review provides an overview of the potential of AM fungi as bioprotection agents against viral diseases and emphasizes the complex nature of plant–fungus–virus interactions. Several mechanisms, including modulated plant tolerance, manipulation of induced systemic resistance (ISR), and altered vector pressure are involved in such interactions. We propose that using “omics” tools will provide detailed insights into the complex mechanisms underlying mycorrhizal-mediated plant immunity. Introduction Arbuscular mycorrhiza (AM), a mutualistic association between the roots of most land plants and fungi from the phylum Glomeromycota [1], confers a series of benefits to host plants [2]. AM fungi improve plant growth and fitness in exchange for carbohydrates from their host to complete their life cycle [3,4]. The mycorrhizal extracellular hyphal spreads widely into the soil, thereby acquiring water and nutrients, especially phosphate [5]; AM fungus also develops in roots where the AM fungus colonizes the cortex and forms highly branched intracellular structures called arbuscules, which facilitate the transfer of mineral nutrients to the root cells. In addition to the improved plant nutritional status, the physiological changes that trigger metabolic changes in the root via AM fungi can also increase the host ability to overcome biotic and abiotic stresses [6][7][8][9][10]. A mycorrhizal bioprotection effect has been observed against soil-borne fungal pathogens that cause wilting or root rot [11,12], and AM symbiosis induces host plant resistance against below-ground and shoot pathogens, nematodes, or chewing insects [13][14][15][16]. In this review, we will focus on the contribution of AM fungi in plant and virus interactions, which have been less investigated. Viral diseases in plants are a major threat to food security worldwide, and this problem is exacerbated by crop management practice and climate changes [17]. The control of plant viruses is mainly based on prevention by using genetically resistant plants and through vector eradication [18]. However, resistance sources are lacking in many cases, and genetic resistance achieved by genetic engineering can be overcome by viruses as it is usually based on a gene-for-gene interaction [19]. Common strategies to control plant virus infection that target vectors via agrochemicals are unacceptable because of their high cost and potential adverse environmental effects. The possibility of priming plant immunity against viruses by using beneficial microorganisms, such as AM fungi, deserves considerable interest. Therefore, this review aims to provide an overview of the impact of AM symbiosis on viral pathogen infection and the mechanisms involved in such interactions. Impact of AM Symbiosis on Viral Development The studies related to the interactions between AM symbiosis and viral pathogens are summarized in Table 1. The effect of AM fungi on viral infection is variable, and considerable attention is focused on shoot virus infection. Focusing on disease severity, Shaul et al. [20] analyzed the interactions between Rhizophagus intraradices and Tobacco mosaic virus (TMV) in tobacco leaves and observed that the disease symptoms are more enhanced in mycorrhizal plants than in non-mycorrhizal plants. Similarly, Nemec and Myhre [21] demonstrated that the leaf shock symptom development is severe in mycorrhizal Citrus rootstocks. The increase in virus accumulation in leaves of Potato virus Y-infected mycorrhizal strawberry and potato plants was observed [22,23]. Mycorrhizal plants could become increasingly sensitive to viral presence over time as virus concentration continuously increases compared with that in non-mycorrhizal controls [24]. During the early stages, decreased or no difference in symptom severity or virus infection is observed in mycorrhizal plants compared with non-mycorrhizal plants [22]. A study demonstrates a clearly protective effect of AM fungi against viral infections in roots and shoots and in disease symptoms [25]. Recently, Hao et al. [26] showed that mycorrhizal colonization significantly decreases nematode vector Xiphinema index reproduction in soil and gall formation on roots and protects grapevine against grapevine fanleaf virus (GFLV). These inconsistent findings related to mycorrhizal plants and viral pathogens that vary with the plant-AM fungus-virus interactions involved have been reported. AM Fungi The reviewed experiments are limited to the single inoculation of one of the following species: Funneliformis mosseae, Funneliformis geosporum, Rhizophagus intraradices, and Glomus sp. (Table 1). These species can be easily propagated and are the most common symbionts which were geographically distributed at a global scale [27]. Monospecies inoculum is used in all these studies. Berruti et al. [3] used a meta-analysis to show that plant growth promotion effects are more successful in single-species mycorrhizal inoculation experiments than in experiments with more than one AM fungi species. This phenomenon can be explained by the fact that the functional redundancy in AM fungi resulted in few fungal species that can alleviate stress and benefit a plant [28]. Furthermore, different isolates within the same AM fungi species can increase the variations in plant response [29], suggesting that functional heterogeneity exists in these species. Though the current general trend that uses single species in these reviewed studies is reasonable, the selection of remarkably effective AM fungi strains for AM fungus-virus interactions is also required. Viruses Virus identity appears to play an important role in the impact of AM symbiosis on virus infections given that viral infection of potato plants resulted in a variety of symptoms depending on the viral strain. The interactions between AM fungi and ssRNA(+) viral plant pathogens, the single largest group of RNA viruses, have been extensively studied. These studies cover 6 out of 30 families in this group of RNA viruses. Among ssRNA(+) viruses, including Bromoviridae, Closteroviridae, Potyviridae, Secoviridae, and Virgaviridae. Only one strain of the Bromoviridae family shows an insignificant difference in cucumber mosaic virus (CMV-Y, yellow strain) accumulation in the cucumber leaves of mycorrhizal and control cucumber plant determined using indirect enzyme-linked immunosorbent assay (ELISA) at 1, 2 and 3 weeks after inoculation [30]. Once inside the plant, this virus can inhibit the plant's ability to signal for gene silencing in other tissues, thereby furthering CMV-Y infection. Similar to other viruses, this kind of virus replicates in the cytoplasm, and moves through the plasmodesmata via cell-to-cell transfers, but the phloem can be utilized for long-distance movement in the plant. Other studies demonstrate a negative effect of AM fungi against ssRNA(+) virus in the shoot and root, leading to an increased virus concentration or symptom severity in mycorrhizal plants. No change in virus accumulation or symptom severity is observed between mycorrhizal and non-mycorrhizal plants 14 days post-virus inoculation (dpi) of tomato spotted wilt virus (TSWV), an RNA virus of ambisense polarity belonging to the family Peribunyaviridae. However, a prolonged increase in virus titer is observed in mycorrhizal plants in a longer period [24]. "Recovery" is a phenomenon defined by the reduction or disappearance of symptoms in virus-infected plants that initially exhibit severe disease and by being protected from reinoculation with the same virus [31]. Results show that 25% of mycorrhizal plants and 65% of non-mycorrhizal plants recovered at 34 dpi, and the decreased of recovery in mycorrhizal plants indicates that the plant's response to TSWV infection is attenuated by mycorrhizal colonization [24]. Mycorrhizal colonization has a beneficial effect in attenuating the disease caused by tomato yellow leaf curl Sardinia virus (TYLCSV), an ssDNA virus belonging to the Geminiviridae family, and causes one of the most serious viral diseases of tomatoes [25]. Considering that geminiviruses colonize the nucleus of cells and are phloem-limited [32], these studies suggest that the different results observed with virus is likely due to the particular nature of the viruses. Plants A plant is another related aspect that should be considered. Nine plant species are reviewed, while, four studies focus on tomato, including different varieties with a range of genetic differences (Table 1). AM fungi protect the same variety of tomato ("Moneymaker") against the TYLCSV, but not from the TSWV [24,25]. Different plant species show a significantly different responsiveness to mycorrhizal inoculation [5,33], and the effectiveness of mycorrhizal bioprotection depends on the plant species involved. Though major crops, such as wheat and maize, have relatively high mycorrhizal dependence on plant growth and nutrient uptake, negligible attention has been given to identifying the bioprotection effects against viral disease on these plants. Environment Conditions Many factors also affect the success of bioprotection, including AM fungi species compatibility with the target environment and degree of competition with other soil organisms in the timing of inoculation. All experiments are conducted under controlled (greenhouses) conditions (Table 1), and pots are usually filled with sterilized substrates with low mineral nutrients, leading to optimized requirements for viruses and AM fungi to infect or colonize compared with that in open-field conditions. Given that high nutrient levels in substrates can reduce AM fungi colonization [5], virus infection is difficult in highly sophisticated management cultivation measures [18]. The positive or negative impacts of AM fungi on plant-virus interactions can be eased or reversed via agricultural practices and in multiple stressed environmental conditions. Additional insights into the variability in tripartite performance in a range of different environments can help increase the efficacy of AM fungi. Mechanisms of Interactions between AM Fungi and Plant Viruses The underlying mechanisms of the impact of AM symbiosis on infection by viral pathogens remain poorly understood. Several mechanisms, including modulated plant tolerance, the manipulation of induced systemic resistance (ISR), and altered vector pressure are involved in such interactions (Figure 1). Mycorrhizal effects likely result from a combination of several mechanisms [13,34]. The mechanisms responsible for the specific plant-AM fungus-virus interaction highlight the need to consider mycorrhizal symbiosis in the context of plant immunity to exploit potential benefits for plant bioprotection. Viruses 2019, 11, x FOR PEER REVIEW 6 of 12 practices and in multiple stressed environmental conditions. Additional insights into the variability in tripartite performance in a range of different environments can help increase the efficacy of AM fungi. Mechanisms of Interactions between AM Fungi and Plant Viruses The underlying mechanisms of the impact of AM symbiosis on infection by viral pathogens remain poorly understood. Several mechanisms, including modulated plant tolerance, the manipulation of induced systemic resistance (ISR), and altered vector pressure are involved in such interactions (Figure 1). Mycorrhizal effects likely result from a combination of several mechanisms [13,34]. The mechanisms responsible for the specific plant-AM fungus-virus interaction highlight the need to consider mycorrhizal symbiosis in the context of plant immunity to exploit potential benefits for plant bioprotection. Modulated Plant Tolerance Mycorrhizal hyphae are considerably thinner than roots and can penetrate smaller pores; they emerge from the root surface and acquire macro and micronutrients from soil volumes that are inaccessible to roots [35]. The increased uptake of nutrients in the host plant may affect the susceptibility of the plant to viral infection. Maffei et al. [25] showed that TYLCSV-infected mycorrhizal plants exhibit less severe symptoms than infected non-mycorrhizal plants, but nonmycorrhizal plants do not suffer from phosphate starvation when they are watered with a modified nutrient solution with optimized phosphate content. This study indicates that the improved nutritional status of mycorrhizal plant alone could not explain its bioprotection against viruses. Daft and Okusanya [22] demonstrated that enhanced virus production can be achieved in nonmycorrhizal plants under higher phosphate levels, and the amount of Arabis mosaic virus from strawberry plants is greater in mycorrhizal than in non-mycorrhizal plants, thus, these results could be attributed to the high phosphate levels in mycorrhizal plants. Borer et al. [36] also showed that increased phosphorous content is associated with an increase in barley and cereal yellow dwarf virus infection. Thus, AM fungi can benefit virus development rather than provide bioprotection. Given the high-affinity phosphate transporter in an AM fungus [37], the nutritional aspects of AM symbiosis have been studied extensively from the molecular perspective [38]. The AM marker gene LePT4, a Modulated Plant Tolerance Mycorrhizal hyphae are considerably thinner than roots and can penetrate smaller pores; they emerge from the root surface and acquire macro and micronutrients from soil volumes that are inaccessible to roots [35]. The increased uptake of nutrients in the host plant may affect the susceptibility of the plant to viral infection. Maffei et al. [25] showed that TYLCSV-infected mycorrhizal plants exhibit less severe symptoms than infected non-mycorrhizal plants, but non-mycorrhizal plants do not suffer from phosphate starvation when they are watered with a modified nutrient solution with optimized phosphate content. This study indicates that the improved nutritional status of mycorrhizal plant alone could not explain its bioprotection against viruses. Daft and Okusanya [22] demonstrated that enhanced virus production can be achieved in non-mycorrhizal plants under higher phosphate levels, and the amount of Arabis mosaic virus from strawberry plants is greater in mycorrhizal than in non-mycorrhizal plants, thus, these results could be attributed to the high phosphate levels in mycorrhizal plants. Borer et al. [36] also showed that increased phosphorous content is associated with an increase in barley and cereal yellow dwarf virus infection. Thus, AM fungi can benefit virus development rather than provide bioprotection. Given the high-affinity phosphate transporter in an AM fungus [37], the nutritional aspects of AM symbiosis have been studied extensively from the molecular perspective [38]. The AM marker gene LePT4, a mycorrhiza-specific phosphate transporter, is preferentially expressed in arbuscule-containing cells of mycorrhizal tomato roots [39,40]. However, the expression profiles of LePT4 in mycorrhizal plants are not modified by TYLCSV infection [25]. Furthermore, the improved nutrient status of mycorrhizal plants, especially under nutrient deficient conditions, leads to vigorous plant growth, which can compensate for viral damage. However, improved plant growth may impact mycorrhizal plants as it increases potential virus multiplication [21]. Though TSWV infection significantly suppresses the AM-induced growth increase, mycorrhizal tomato plants showed a growth increase compared with that of mock-inoculated ones under viral stress [24]. Therefore, enhanced growth in mycorrhizal plants could also compensate for damage caused by viruses, and mycorrhizal symbiosis still benefits plants because they are able to tolerate increased viral pressure under certain conditions. Manipulation of Induced Systemic Resistance The modulation of plant physiology and signal transduction pathways during mycorrhizal symbiosis formation and function has received considerable attention [41][42][43]. A transient and weak activation of plant immune system takes place in response to mycorrhizal colonization, with the elicitation of specific defense reactions [44]. Microbe-associated molecular patterns (MAMPs) from AM fungi are recognized by the innate immune system of a host plant, while MAMP-triggered immunity (MTI) could prime the salicylic acid-dependent defense responses [45]. The enhanced production of phytoalexins and phenolic compounds; accumulation of hydrolytic enzymes, such as chitinases and β-1,3-glucanases; and activation of phenylpropanoid metabolism in plant roots are involved in MTI on AM fungal colonization [34]. The pre-conditioning of the host plant elicited by AM fungi strengthens and speeds up systemic defense responses against subsequent plant pathogens [12]. Phytoalexin synthesis and cell wall fortification, which are effective against bacterial or fungal pathogens [46], usually could not prevent virus spread or replication [47]. β-1,3-glucanases, which inhibit callose deposition degradation in plasmodesmata, could be accumulated in mycorrhizal cucumber plants [48], and delay cell-to-cell virus spread and loading into phloem [49]. Such priming of plant defense conferred by mycorrhizal symbiosis may be involved in AM fungi-mediated bioprotection against plant viruses. Mycorrhiza-induced resistance (MIR) against many plant fungal and bacterial pathogens shares common characteristics with the systemic acquired resistance (SAR) after pathogen attack and is associated with the SAR-like priming of defense response, such as the accumulation of pathogenesis-related (PR) proteins [34]. Gallou et al. [50] observed a strong induction of PR protein genes (PR1 and PR2) in mycorrhizal plants challenged with Phytophthora infestans in vitro. However, the accumulation and mRNA steady-state levels of PR proteins, PR-1 and PR-3, are low in the leaves of mycorrhizal plants infected with the tobacco mosaic virus [20]. The expression of PR proteins is tightly correlated with the SA signal transduction pathway during necrotic lesion formation in tobacco-virus interactions [51]. SA levels are enhanced in TSWV-infected mycorrhizal and non-mycorrhizal plants [24], while Shaul et al. [20] further indicated that the delayed PR-1 and PR-3 gene expression in mycorrhizal plants infected by TMV is not involved in SA-dependent defense, because the exogenous application of SA to the foliage does not abolish the mycorrhizal plant response. The priming of jasmonic acid (JA)-dependent defense responses is demonstrated in mycorrhizal plants under a pathogen infection unlike in the non-mycorrhizal control [52]. AM fungi reduce the development of plant pathogens through ISR, a resistance phenomenon usually induced by non-pathogenic microorganisms [10]. Reports of decreased pathogen development in shoot or in non-mycorrhizal parts of mycorrhizal root systems using a split-root system is confirmed as a mycorrhizal-mediated ISR [53][54][55]. ISR is predominantly regulated by the JA-mediated and ethylene-mediated signaling pathways [34]. Li et al. [56] showed that mycorrhizal plants have a higher JA content compared with that of non-mycorrhizal plants during Phytophthora sojae infection, and Pozo et al. [57] demonstrated that the expression of marker genes for JA responses are significantly increased in mycorrhizal tomato plants. JA has a limited effect in early local defenses of potatoes infected with potato virus Y-NTN [58], and Moizzi et al. [24] indicated that increased levels of JA in mycorrhizal plants are detected with or without TSWV infection. Considering that the SA and JA pathways are usually mutually antagonistic, these signaling pathways may not act independently but influence each other through a complex network in virus-AM fungus-plant interactions [59]. A high-throughput transcriptional profiling analysis via microarrays is conducted to monitor transcriptional changes in the roots and shoots of mycorrhizal plant infection with TSWV, and this transcriptome study may shed light on tripartite interaction [24]. The number of differentially expressed (DE) genes in the roots of TSWV-infected mycorrhizal plants is higher compared with that measured in the single-inoculation treatments. In shoots, the impact of combined TSWV and AM fungus appears intermediate between that observed for the mycorrhizal (lowest) and the virus (highest) interaction separately. A total of 215 genes modified the regulation in the shoots TSWV-infected mycorrhizal plants, while 579 DE genes were found in the roots. This transcriptome data indicates that the expression levels of several candidate virus-responsive upregulated genes related to sugar metabolism, defense, and response to hormones are reduced in mycorrhizal plants compared with that in non-mycorrhizal plants after TSWV infection. For example, the expression of PR protein 10, which has antimicrobial activity, is downregulated in TSWV-infected mycorrhizal plants, but PR protein 10 could be linked to the reduction of plum pox virus infection in Nicotiana tabacum [60]. On the basis of the suppression subtractive hybridization study, Hao et al. [55] showed that glutathione S-transferase (GST) is upregulated during MIR. GST is involved in the detoxification of reactive oxygen species and is upregulated in response to TSWV [61] but is not activated in virus-infected mycorrhizal plants [24]. However, only a few transcriptomes of AM fungi-associated changes are available, and future "omics" studies of viral attackers might clarify whether the AM fungi priming of plant defenses is effective. Altered Vector Pressure The control of virus diseases could also be based on prevention by eradication of insect, nematode and fungal vectors [18]. These mycorrhizal protective effects range from enhanced plant tolerance to a reduction in pathogen infection [14,16,62]; therefore, the potential of AM symbiosis to restrict these vectors may contribute to diminishing viral disease severity. The soil-borne GFLV spreads mainly via the nematode vector X. index, which is suppressed by the AM fungus R. intraradices with induced local and systemic protection [55]. Therefore, the bioprotection effects of AM symbiosis to restrict a vector to biologically realistic thresholds may limit GFLV infection. The reduced viruliferous nematode development after R. intraradices inoculation does not exclude GFLV infection at an extremely high nematode pressure (100 nematodes per plant), but GFLV is absent from mycorrhizal grapevine roots 90 days after nematode inoculation at a low nematode pressure (10 nematodes per plant, approximately the levels of nematode abundance observed in vineyards [63][64][65]) but detectable in non-mycorrhizal roots [26]. Thus, management of the nematode vector by using AM fungi has a potential to diminish GFLV disease severity. However, the colonization of Plantago lanceolata by AM fungi improves the growth and reproduction in the shoot of the sucking insect Myzus persicae [66,67], which acts as a vector for the transport of various mosaic viruses, potato leafroll virus, potato virus Y, and Mikania micrantha wilt virus [68]. As a virus depends on a vector for its survival and transmission, the improved performance of this aphid on mycorrhizal plants may lead to enhanced plant infection by the viruses. For fungal vectors, root colonization by AM fungi is associated with symptomless root parasites, Olpidium species [69,70], which are also known as the vectors of viruses on cereals, tobacco and salad [71,72]. The potential of AM fungi to reduce fungal pathogen infections has been shown frequently [16,34], but no information is available concerning mycorrhizal protection against fungal vectors mediating virus transmission. Conclusions and Future Directions We provide an overview of the potential of AM fungi as bioprotection agents against viral diseases and emphasize the complex nature of plant-fungus-virus interactions. However, data are still limited to certain stages of virus symptoms, and the actual long-term processes attained by inoculating plants with AM fungi must be evaluated case-by-case in the field. These interactions depend on several biotic and abiotic factors, and practices such as the use of pesticides or fertilization (especially that of phosphorous), can be controversial for plant-fungus-virus interactions. The technological progress unraveled the mechanisms proposed for mycorrhizal-mediated bioprotection, and relevant strategies, such as next-generation sequencing, may further elucidate the mechanisms of induced resistance of AM symbiosis. The complex relationship between the systemic priming of plant defenses and the suppression of immunity, which are required for the establishment of AM symbiosis, must also be further studied involving "omics" tools [73]. In addition, the bioprotection efficiency of AM fungi may be improved if they are used in combination with other biological control agents. Elsharkawy et al. [30] showed that the co-inoculation of cucumber plants with AM fungi and a plant growth-promoting fungus Fusarium equiseti results in the effective control of CMV development, and the importance of microorganisms in rhizosphere and phyllosphere will be confirmed with microbiome studies. Though most studies on plant-AM fungus-virus interactions have been conducted under controlled conditions, the development of mycorrhizal inocula for large-scale field application is growing quickly [14,74]. In view of sustainable agriculture, unveiling the principles behind the functional interplay among the tripartite will be of major interest to the effective application of AM fungi in an integrated viral management program. Conflicts of Interest: The authors declare no conflict of interest.	2019-06-12T14:56:11.266Z	2019-06-01T00:00:00.000	{ "year": 2019, "sha1": "6739033c52aacd1b5b501ed8128756e30d76b1ec", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1999-4915/11/6/534/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "6739033c52aacd1b5b501ed8128756e30d76b1ec", "s2fieldsofstudy": [ "Environmental Science", "Biology" ], "extfieldsofstudy": [ "Medicine", "Biology" ] }
244027862	pes2o/s2orc	v3-fos-license	Role of alternatively spliced, pro‐survival Protein Kinase C delta VIII (PKCδVIII) in ovarian cancer Abstract Ovarian cancer is the deadliest malignant disease in women. Protein Kinase C delta (PRKCD; PKCδ) is serine/threonine kinase extensively linked to various cancers. In humans, PKCδ is alternatively spliced to PKCδI and PKCδVIII. However, the specific function of PKCδ splice variants in ovarian cancer has not been elucidated yet. Hence, we evaluated their expression in human ovarian cancer cell lines (OCC): SKOV3 and TOV112D, along with the normal T80 ovarian cells. Our results demonstrate a marked increase in PKCδVIII in OCC compared to normal ovarian cells. Therefore, we elucidated the role of PKCδVIII and the underlying mechanism of its expression in OCC. Using overexpression and knockdown studies, we demonstrate that PKCδVIII increases cellular survival and migration in OCC. Further, overexpression of PKCδVIII in T80 cells resulted in increased expression of Bcl2 and knockdown of PKCδVIII in OCC decreased Bcl2 expression. Using co‐immunoprecipitations and immunocytochemistry, we demonstrate nuclear localization of PKCδVIII in OCC and further show increased association of PKCδVIII with Bcl2 and Bcl‐xL in OCC. Using PKCδ splicing minigene, mutagenesis, siRNA and antisense oligonucleotides, we demonstrate that increased levels of alternatively spliced PKCδVIII in OCC is regulated by splice factor SRSF2. Finally, we verified that PKCδVIII levels are elevated in samples of human ovarian cancer tissue. The data presented here demonstrate that the alternatively spliced, signaling kinase PKCδVIII is a viable target to develop therapeutics to combat progression of ovarian cancer. \| INTRODUCTION Ovarian cancer remains the predominant cause of death from gynecologic malignancy in the United States. 1 This poor outcome is likely related to the fact that most patients are detected with advanced stage disease and widespread peritoneal metastasis. Dysregulation in signaling and metabolic pathways is seen with ovarian cancer where several genes are involved in cross talk resulting in a complex phenomenon. Often, these signaling pathways determine the response to drugs such as paclitaxel and platinum that are administered to treat the ovarian cancer. The serine/threonine kinase family of Protein Kinase C (PKC) isoforms regulate several signal transduction cascades and are shown to be extensively linked to various cancers. Alterations in expression levels of PKC isoforms are proposed as biomarkers in certain cancers. 2,3 Protein kinase C delta (PKCδ; PRKCD), a member of novel PKC subfamily, plays a central role in cell survival with dual effects: as a mediator of apoptosis and as a promoter of cell survival. The dual functions reported by studies are explained by PKCδ splice variants. Alternative splicing, wherein alternate patterns of exon inclusion/exclusion or choice of 3′ or 5′ splice sites during pre-mRNA splicing, generates more than one protein from the same gene. We demonstrated a novel splice variant in humans and showed that PKCδ alternative splicing produces PKCδI and PKCδVIII, which are switches that determine cell survival and fate. The characterization and function of PKCδVIII was published by our laboratory. 4 PKCδII is the mouse homolog of human PKCδVIII; both are generated by alternative 5′ splice site usage, and their transcripts share >94% sequence homology. We have shown that PKCδII and PKCδVIII function as pro-survival proteins in neurons and adipocytes. 4,5 PKCδ is shown to be involved in tumorigenesis and metastasis in several cancers such as breast, lung, and prostate cancer [6][7][8][9][10] ; however, specific function of PKCδ splice variants in any cancer and particularly in ovarian cancer has not been elucidated yet. Since PKCδ regulates apoptosis and survival, we evaluated the expression of PKCδ human splice variants PKCδI and PKCδVIII in human ovarian cancer cells (OCC) SKOV3 and TOV112D along with the normal T80 ovarian cells. Our results demonstrated a marked increase in PKCδVIII in OCC compared to normal ovarian cells which was not recognized thus far. Cellular pathways mediated by B-cell lymphoma 2 (BCL2) family of proteins determine the balance of apoptosis and survival in cancer. The BCL2 family has BCL2 homology (BH) domains that mediate their pro-survival or pro-apoptotic functions. The antiapoptotic proteins Bcl2 and Bcl-x have the BH4 domain. Bcl-x (also called BCL2like 1: BCL2L1) is alternatively spliced to generate two variants: Bcl-xL which is antiapoptotic and Bcl-xS which is pro-apoptotic. Bcl2 associated agonist of cell death (BAD), which is pro-apoptotic, contains the BH3 domain. The BCL2 family proteins associate in response to cellular cues and the association of specific partners promote either cellular survival or apoptosis. Bcl2 is held tightly by Bad in a complex. Pro-survival signals promote phosphorylation of Bad resulting in dissociation of this complex. Bcl2 can then associate with Bcl-xL to promote survival and inhibit cytochrome C-mediated apoptosis. Several studies [11][12][13] have shown that Bcl2 promotes increased survival in ovarian cancer cells. Resistance to drugs in cancer is sometimes conferred by an increase in the pro-survival Bcl2. The link between PKCδVIII and Bcl2 family of proteins is not yet elucidated. Since elevated expression of PKCδVIII, a pro-survival signaling kinase, may be a significant determinant of life quality and expectancy, we sought to understand the effects of PKCδVIII on cellular and metabolic functions in ovarian cancer and further elucidate the molecular mechanisms that promote its increased expression in ovarian cancer. Understanding the role of alternatively spliced genes contributing to increased survival and proliferation in ovarian cancer cells will lead to development of novel therapeutic approaches. Separately, 1 μl of cDNA was amplified by real-time quantitative PCR using Maxima SYBR Green/Rox qPCR master mix (Thermo Scientific) in an ABI ViiA7 sequence detection system (PE Applied Biosystems) to quantify the relative levels of the transcripts in the samples. Real-time PCR was then performed in triplicate on samples and standards. The plate setup included a standard series, no template control, no RNA control, no reverse transcriptase control, and no amplification control. After primer concentrations were optimized to give the desired standard curve and a single melt curve, relative quotient (RQ) was determined using the ∆∆C T method with βactin as the endogenous control and control samples as the calibrator sample. For absolute quantification (AQ), a standard curve was generated for PKCδVIII. To do so, PKCδVIII-pTracer plasmid was used to obtain a standard curve correlating the amounts (ng) with the threshold cycle number (C t values). A linear relationship (r 2 > 0.96) was obtained for PKCδVIII. Real-time qPCR was then performed on samples and standards in triplicates. Samples were normalized to βactin for absolute quantification of PKCδVIII expression levels. \| Transfection of PKCδVIII plasmid and antisense oligonucleotides The pTracer-PKCδVIII plasmid was cloned and verified by sequencing as described in our previous publications. 4,14 The 20-mer antisense oligonucleotide masking the SRSF2 binding site on PKCδ mRNA (ASO) is 2′-methoxyethyl-modified, RNase-H resistant and was synthesized by Ionis Pharmaceuticals, CA along with the scrambled control. Cells were transfected with 2 μg of pTracer-PKCδVIII plasmid or the ASO for 48 h using Lipofectamine 3000 reagent (ThermoFisher) per manufacturer's instructions. Control cells were transfected using Lipofectamine 3000 alone or control ASO as per experimental setup. Total RNA was harvested and PCR was performed as described above. \| Knockdown of PKCδVIII or SRSF2 PKCδVIII siRNA was custom designed (validated as described in Ref. [14]) to target PKCδVIII pre-mRNA, and purchased from ThermoFisher Scientific: PKCδVIII siRNA (ID: 10620110). SRSF2 siRNA (ThermoFisher ID:4392420, previously validated for specificity using three siRNAs and elimination of off-target effects 14 ) was purchased from ThermoFisher Scientific along with nonspecific siRNA used in experiments as control. Cells were transfected with 25 nM PKCδVIII siRNA or 50 nM SRSF2 siRNA along with control siRNA for 48 h using RNAiMax reagent (ThermoFisher) per manufacturer's instructions. Total RNA was harvested and PCR was performed as described above or whole cell lysates were analyzed using Western blot as described above to confirm the depletion. \| AOPI cellular survival assay Cells were grown in a 12-well plate and PKCδVIII was overexpressed as described above. Cells were treated with 25 μg/ml etoposide for 18 h. Cells were then trypsinized and washed once with PBS. The cell pellet (containing one million cells) was resuspended in 500 μl PBS and fixed by slow, drop-wise addition of 4.5mL ice-cold 70% ethanol while gently vortexing. Samples were incubated overnight at 4°C to complete fixation and then stored at −20°C until stained. A fresh solution of Propidium Iodide (1 mg/ml, ViaStain TM CS1-0109) and RNase A (2mg/ml, ThermoFisher EN0531) diluted in water. Fixed cells were centrifuged at 1000 rpm for 5 min. Cell pellet was washed twice with PBS and pellet was resuspended in 50 μl PI/RNase A solution and incubated at room temperature for 5 min. One milliliter PBS was added and samples were divided to create unstained negative control for analysis. Acridine Orange (ViaStain TM CS2-0106) was added (1:1) to samples for staining and incubated at 37°C for 30 min then analyzed on the Nexcelom K2 cellometer. \| Cell migration by scratch assay Cells were grown to confluency in 35-mm dishes with μ-Dish inserts (Ibidi solutions™, 81176) to make consistent and reproducible 500μm gaps. PKCδVIII was overexpressed in T80 cells or expression was knocked down in OCC for 48 h. The inserts were then removed and imaged on the Keyence BZx-810 microscope at 4x magnification. Images were taken at the same location saved into the Keyence BZx-810 microscope with images taken at 0, 24, and 48 h. Analysis of done measuring empty area between cells in µm 2 using the FastTrack A1 software (Ibidi solutions™). \| Cell migration assay Transwell cell migration assays were performed using BD Falcon Cell Culture Inserts in a 24-well plate. Ovarian cancer cells were plated in the upper insert in their respective serum-free medium and the outer, bottom chamber was filled with 300 μl of respective medium containing 10% FBS. The ovarian cancer cells with and without the PKCδVIII siRNA were allowed to migrate for 18 h. After 18 h, the cells that migrated to the bottom well were fixed and stained using crystal violet and imaged on Keyence BZx-810 microscope. The number of cells were counted using the automated Keyence software analysis module. \| Co-immunoprecipitation assay Cells were harvested and 200 μg of lysate was used for co-immunoprecipitation assay. Lysates were rocked with Protein A/G Agarose (Santa Cruzsc2003) for 30 min at 4°C to clear nonspecific binding and centrifuged at 2000 rpm for 1 min. The supernatant was rocked with 2 μg Bcl2agarose antibody (Santa Cruz sc7382AC), Bcl-xL-agarose antibody (Santa Cruz sc8392AC) or Bad-agarose antibody (Santa Cruz, sc8044AC) overnight at 4°C and pellet was washed and resuspended in Laemmli buffer followed by Western blot analysis as described above. \| Immunocytochemistry Ovarian cells were plated into a 12-well plate as described above. Cells were fixed with 4% paraformaldehyde for 30 min at room temperature and blocked with 1% bovine serum albumin in PBS for 1 h. Cells were incubated in primary antibodies against Bcl2 (1:100), Bad (1:100), Bcl-xL (1:100), and PKCδVIII (1:50). Cells were rocked overnight at 4°C. Cells were then washed with PBS and rocked in the dark for 1 hour in Alexafluor 488 and Alexafluor 577 secondary antibodies (1:1000, Invitrogen). Cells were briefly rinsed with PBS and stained with DAPI mounting media. Images were captured using Keyence BZx-810 microscope and analyzed using Keyence Analyzer software. \| WST cell viability assay Cells were grown to in 35-mm dishes and PKCδVIII siRNA was transfected as described above. WST-1 (Roche Molecular Biochemicals, IN) was diluted to a final concentration of 10% (v/v) in serum-free cell media. Cells were incubated for 2 h at 37°C. The formazon dye produced by viable cells was quantified using a spectrophotometer set at a wavelength of 440 nm, and absorbance was recorded for each well using reference wavelength of 690 nm. \| BrdU proliferation assay BrdU (Millipore, 2750) was diluted to 100 μM stock concentration in PBS, then diluted 1:10 in culture media. Cells were washed with PBS before incubated in BrdU media for 2 h. Cells were then fixed in 4% paraformaldehyde for 20 min at room temperature then blocked for 30 min at room temperature. One milliliter of antibody staining buffer with anti-BrdU primary antibody was added and incubated overnight at room temperature. Peroxidase-secondary antibody was added for 1 h at room temperature then Peroxidase substrate was added for 30 min at room temperature. Positive wells were visibly blue and intensities were measured by spectrophotometer readings at 450/540 nm. \| Construction of pSPL3_PKCδ and pSPL3_PKCδSRSF2 Minigenes The pSPL3_PKCδ splicing minigene was constructed as described in our previous publication. 14 Briefly, PKCδ exon 10 and its flanking 3ʹ intronic sequence (containing the branch point and 3ʹ splice site) and 5ʹ intronic sequences (containing flanking 5ʹ intronic sequence that includes 5′ splice site I and splice site II) were cloned into the multiple cloning site of the pSPL3 splicing vector between the splice donor (SD) exon and splice acceptor (SA) exon. The resulting splicing minigene was verified by restriction digestion and sequencing. To generate the pSPL3_PKCδSRSF2 minigene, the SRSF2 cis-element (sequence: ggccaaag) identified near 5′ splice site II of exon 10 of PKCδ pre-mRNA, was mutated in the pSPL3_ PKCδ minigene to "tagcccata" using QuikChange sitedirected mutagenesis kit (Stratagene), which allows for blue/white screening per the manufacturer's instructions. The mutated minigene, pSPL3_PKCδSRSF2, was verified by sequencing (published in Ref. [14]). Two micrograms of pSPL3_PKCδ or pSPL3_PKCδSRSF2 minigene was transfected into cells, total RNA was isolated and PCR was performed to visualize splicing of the minigene using primers SD 5'-TCTGAGTCACCTGGACAACC-3' and SA 5'-CACCTGAGGAGTGAATTGGTC-3'. \| Statistical analysis All experiments were repeated 3-5 times as biological replicates and experimental samples run in triplicate to ensure reproducibility of results. Analyses were performed using PRISM TM software and analyzed using two-tailed Student's t-test, one-way or two-way ANOVA as indicated in figure legends. p < 0.05, p < 0.01, and p < 0.001 were used as significant measures. \| PKCδVIII levels are increased in ovarian cancer cells In humans, PRKCD gene (i.e., PKCδ) is alternatively spliced to PKCδI and PKCδVIII mRNA 4 via utilization of alternate 5′ splice sites on exon 10 as shown in schematic of Figure 1A. We evaluated the expression of PKCδ splice variants PKCδI and PKCδVIII in normal immortalized human ovarian epithelial T80 cells and human ovarian cancer cells (OCC) SKOV3 and TOV112D. Total RNA was isolated for real-time SYBR Green quantitative PCR using primers specific for PKCδI and PKCδVIII. Simultaneously, whole cell lysates were analyzed by Western blot using antibody against PKCδI (Cell Signaling) and PKCδVIIIspecific antibody (raised by Patel lab as published in Ref. [4]). The levels of PKCδI were not significantly different between OCC and normal T80 cells. Results ( Figure 1B,C) demonstrate a 7-fold increased expression of PKCδVIII in SKOV3 compared to T80 while TOV112D had 12-fold increased expression of PKCδVIII compared to T80 cells. \| Increased expression of PKCδVIII in ovarian cells promotes cellular survival Since our results showed increased expression of PKCδVIII in ovarian cancer cells compared to normal ovarian cells, we sought to evaluate the impact of increased levels of PKCδVIII on cell survival in ovarian cells. Hence, we overexpressed PKCδVIII in T80 cells. Overexpression was confirmed by PCR using primers that detected PKCδI and PKCδVIII simultaneously ( Figure 2A). We evaluated cell survival in response to etoposide, an inducer of apoptosis that is also used in cancer treatment regimens, using Acridine Orange/Propidium Iodide (AOPI) staining. After 48 h of PKCδVIII overexpression, medium was changed and 25μg/ml etoposide was added for 18 h. The live and dead cells were measured on a Cellometer (Nexcelom). Briefly, PI stains cells with compromised membranes and hence fluorescence by AO is reduced in dead cells. Results ( Figure 2B) demonstrate that overexpression of PKCδVIII protects cells against etoposide-induced apoptosis. \| PKCδVIII increases expression of pro-survival proteins Bcl2 and Bcl-xL increased Bcl2 levels. Interestingly, we observe increased expression of Bcl-xL, with no significant change in Bcl-xS levels, in PKCδVIII overexpressing T80 cells. These results demonstrate that increased expression of PKCδVIII promotes genes in survival pathways in ovarian cells. \| Ovarian cells overexpressing PKCδVIII have increased cellular migration Next, we determined the migration of T80 cells overexpressing PKCδVIII compared to control T80 cells. Scratch assay is an established method to measure cell migration in vitro. 16 Ibidi culture inserts were used to create cell-free gaps which are reproducible. T80 cells were plated in 35-mm dishes with inserts and transfected with 2 μg of pTracer-PKCδVIII as described above. After 24 h insert is removed and images were taken at 24 h using Keyence BZx810. Quantification of migration is achieved by FastTrack A1 Image analysis software (Ibidi solutions™). Our results ( Figure 4A,B) demonstrate that PKCδVIII overexpression accelerated rate of migration in T80 cells. \| PKCδVIII associates with Bcl2 in ovarian cells Since PKCδVIII is a pro-survival kinase and our results above showed that its expression is increased in ovarian cancer cells, we sought to elucidate the survival pathway through which PKCδVIII mediated survival in ovarian cancer cells. We performed a co-immunoprecipitation in T80 cells along with T80 cells overexpressing PKCδVIII. Whole cell lysates were harvested, and the PKCδVIIIspecific antibody was used to immunoprecipitate the F I G U R E 1 PKCδVIII is increased in ovarian cancer cells. (A) Schematic depicts PKCδ pre-mRNA alternative splicing which produces PKCδI mRNA (via utilization of 5′ splice site I (SSI) of exon 10) and PKCδVIII mRNA (via utilization of 5′ splice site II (SSII) of exon 10). (B) Total RNA was isolated from T80, SKOV3, and TOV112D ovarian cells followed by real-time SYBR Green qPCR analysis using primers specific to PKCδVIII, PKCδI, and βactin (internal control). Graph shows relative quantification (RQ) for PKCδI and PKCδVIII with T80 set as reference. The experiments were independently repeated five times with similar results. Statistical analysis was performed using oneway ANOVA; * p < 0.001. (C) Western blot analysis was performed on whole cell lysates and probed using antibodies against PKCδVIII, PKCδI, and βactin as indicated in the figure. Densitometric analysis of individual bands for each protein were normalized to βactin and percent PKCδVIII (PKCδVIII expression in total PKCδ) was calculated using the following equation: percent PKCδVIII = [PKCδVIII / (PKCδVIII + PKCδI)] × 100. Statistical analysis was performed using one-way ANOVA; *p < 0.001 associated proteins. Western blot was then performed for proteins associated with survival in ovarian cancer. Our results ( Figure 5A,B) showed high levels of association of PKCδVIII with Bcl2, Bcl-xL, and Bad. Bax, Bak, Caspase-3, Caspase-9, PARP, and XIAP were not detected in the coimmunoprecipitation indicating that these proteins of the apoptosis pathway did not associate with PKCδVIII. Based on these results, we sought to validate that PKCδVIII was associated with the Bcl2, Bcl-xL, Bad protein complexes. Cell survival is mediated by Bcl2 and Bcl-xL which inhibit the activity of Bax-Bak as well as through inhibition of the cytochrome C-mediated caspase cascade. Bcl2 is sequestered by hypo-phosphorylated Bad which decreases the ability of Bcl2 to associate with Bclx-L to promote survival. Hence, whole cell lysates were harvested from T80 and T80 overexpressing PKCδVIII followed by immunoprecipitation using antibodies against Bcl2, Bcl-xL or Bad. Western blot was then performed to evaluate the associated proteins. Figure 5A shows the Western blot with the input lysate. Results from Bcl2 immunoprecipitation ( Figure 5C) showed PKCδVIII and Bcl-xL associated with Bcl2 in T80 cells. In PKCδVIIIoverexpressing T80 cells, the association between Bcl2 and PKCδVIII as well as Bcl2 and Bcl-xL is dramatically increased. A similar association was seen in Bcl-xL immunoprecipitated T80 cells ( Figure 5D) and increased levels of Bcl2 and PKCδVIII associated with Bcl-xL in PKCδVIII-overexpressing T80 cells. Interestingly in PKCδVIII-overexpressing T80 cells, results from Bad immunoprecipitation ( Figure 5E) showed a significant decrease in association of Bcl2 and Bcl-xL with Bad while PKCδVIII association with Bad did not change significantly. As seen in the Western blots of the lysates (panels in Figure 5A) the phosphorylation of Bad (S112) was significantly increased in PKCδVIII-overexpressing T80 cells. PKCδVIII did not associate with Bax or Bak (not shown). These results demonstrate that increased expression of PKCδVIII in ovarian cells promotes the association of survival complex Bcl2-Bcl-xL with simultaneous dissociation of Bcl2 and Bad complex that is concurrent with increased survival and migration. Bcl-xL To visualize the association and localization in situ, we performed immunocytochemistry in T80, SKOV3 and TOV112D cells. Using an antibody specific for PKCδVIII, results ( Figure 5E) demonstrate that PKCδVIII is expressed both in the cytoplasm and nucleus with higher levels of expression observed in SKOV3 and TOV112D cells compared F I G U R E 2 Overexpression of PKCδVIII in T80 cells promotes cellular survival. (A) 2μg of pTracer-PKCδVIII was transfected in T80 cells for 48 h. (A) RNA was isolated and PCR was performed using primers that simultaneously detect both PKCδ variants. Five percent of the products were separated by PAGE and silver stained for visualization. The experiments were independently repeated five times with similar results. Graph shows percent PKCδVIII (PKCδVIII expression in total PKCδ) that was calculated using the following equation: percent PKCδVIII = [PKCδVIII / (PKCδVIII + PKCδI)] × 100. Statistical analysis was performed using t-test; p < 0.001. (B) 48 h after pTracer-PKCδVIII transfection, T80 cells were treated with 25 μg/ml etoposide for 18 hours. Cells were fixed and stained with acridine orange (AO) and propidium iodide (PI) and viability of cells was analyzed using a cellometer. The experiments were independently repeated four times with similar results. Statistical analysis was performed using one-way ANOVA; *p < 0.001 Bcl2 in ovarian cancer cells Since our results above showed that PKCδVIII promoted survival pathways, we evaluated whether depletion of PKCδVIII in ovarian cancer cells affected cellular migration and survival genes. Previously, we validated a PKCδVIIIspecific siRNA that depleted the cells of PKCδVIII and did F I G U R E 3 Overexpression of PKCδVIII in T80 cells increases Bcl2 and Bcl-xL. (A) RNA was isolated and endogenous levels of PKCδVIII, Bcl2, and Bcl-xL in T80, SKOV3, and TOV112D cells were determined by real-time SYBR Green qPCR. Graph shows relative quantification (RQ) for Bcl2, Bcl-xL, and PKCδVIII with T80 set as reference. The experiments were independently repeated five times with similar results. Statistical analysis was performed using one-way ANOVA; p < 0.05, p < 0.01. (B) 2 μg of pTracer-PKCδVIII was transfected in T80 cells for 48 h. RNA was isolated and PCR was performed using primers for PKCδI, PKCδVIII, Bcl2, Bcl-x (primer pair simultaneously detects Bcl-xL and Bcl-xS) and βactin. PCR products were run on a 1% agarose gel and stained using ethidium bromide. Graphs show relative quantification of PKCδVIII, Bcl2, and Bcl-xL normalized to βactin with control set as reference. The experiments were independently repeated four times with similar results. Statistical analysis was performed using t-test; p < 0.01 and **p < 0.001. (C) 2 μg of pTracer-PKCδVIII was transfected in T80 cells for 48 h and whole cell lysates were analyzed using Western blot. Membranes were probed using antibodies as indicated in figure. Graphs show densitometric analysis of PKCδVIII, Bcl2 and Bcl-xL normalized to βactin. The experiments were independently repeated three times with similar results. Statistical analysis was performed using t-test; p < 0.05 and p < 0.01 Con PKCδVIIIoe Western Blot not affect PKCδI expression. 14 Hence, we transfected 25nM PKCδVIII siRNA (and a nonspecific siRNA (control) in SKOV3 and TOV112D cells for 48 h. Total RNA was isolated for real-time quantitative SYBR Green PCR using primers specific for PKCδI, PKCδVIII, and Bcl2. Simultaneously, whole cell lysates were harvested for Western blot analysis using antibodies against PKCδI (Cell Signaling), PKCδVIII (specific antibody as published in Ref. [14]) or Bcl2 (Cell Signaling). Results ( Figure 6A,B) demonstrate a significant decrease of PKCδVIII levels in PKCδVIII-depleted SKOV3 (SKOV3 PKCδVIIIsi ) compared to SKOV3 (control siRNA) as well as in PKCδVIII-depleted TOV112D (TOV112D PKCδVIIIsi ) compared to TOV112D (control siRNA). Further, our results demonstrate a concurrent decrease in Bcl2 expression with depletion of PKCδVIII. The levels of PKCδI remained unchanged with knockdown of PKCδVIII. \| Depletion of PKCδVIII decreases cellular migration in ovarian cancer cells Next, we determined cellular migration using the scratch assay (as described above in Figure 2) in SKOV3, SKOV3 PKCδVIIIsi , TOV112D, TOV112D PKCδVIIIsi . Results ( Figure 6C) demonstrate that SKOV3 PKCδVIIIsi and TOV112D PKCδVIIIsi have significantly lower migration rates compared to control SKOV3 and TOV112D, respectively, as determined by slower closure of wound gap. To further validate that PKCδVIII influenced cell migration, we performed the transwell cell migration assay for 18 h in SKOV3, SKOV3 PKCδVIIIsi , TOV112D, TOV112D PKCδVIIIsi . Results ( Figure 6D) demonstrate that SKOV3 PKCδVIIIsi and TOV112D PKCδVIIIsi have significantly lower cell migration rates compared to control SKOV3 and TOV112D, respectively, as determined by the lower number of cells migrating through the physical barrier toward the chemo-attractant (FBS). Results from these assays indicate that PKCδVIII affects the migration of ovarian cancer cells. \| Depletion of PKCδVIII decreases cell survival in ovarian cancer cells Separately, WST-1 assay was used to determine cell survival in PKCδVIII depleted SKOV3 and TOV112D cells as described above. After 48 hours of PKCδVIII-siRNA transfection, WST-1 assay was performed. The formazan dye produced by viable cells was quantified according to manufacturer's instructions. Results ( Figure 6E) demonstrate a decrease in cell survival in SKOV3 PKCδVIIIsi and TOV112D PKCδVIIIsi compared to SKOV3 and TOV112D, respectively. \| Depletion of PKCδVIII decreases cell proliferation in ovarian cancer cells Next, we determined the effect of depleting PKCδVIII on proliferation of ovarian cancer cells. SKOV3 and TOV112D were depleted of PKCδVIII using its specific siRNA as described above in a 48-well plate where each group was performed in triplicate. After 48 h, 100 μl BrdU F I G U R E 4 Overexpression of PKCδVIII in T80 increases cell migration. (A) T80 cells grown to confluency with μ-Dish inserts (Ibidi solutions, 81176) and were transfected with 2 μg pTracer-PKCδVIII plasmid for 48 h. Inserts were removed to make consistent gaps and cells were imaged at 0, 6, and 24 h. The red lines are drawn on the image to aid in visualization of the edges of the gaps. The cells were imaged using Keyence BZx-810 microscope and analysis of gap area between cells (μm 2 ) was quantified using FastTrack A1 software (Ibidi solutions™). (B) RNA was then isolated from the cells and PCR performed using PKCδVIII specific primers. The products were separated on a 1% agarose gel and stained using ethidium bromide. Graph shows relative quantification of PKCδVIII normalized to βactin with control set as reference. The experiments were independently repeated three times with similar results. Statistical analysis was performed using two-way ANOVA; * p < 0.001 was added to determine cellular proliferation. The BrdU incorporated into the cells was measured using peroxidase conjugate and represented the percent of proliferating cells. Results ( Figure 6F) demonstrate that depleting PKCδVIII from SKVO3 or TOV112D cells resulted in decreased proliferation compared to their respective controls. \| Alternative splicing of PRKCD promotes higher expression of PKCδVIII via SRSF2 in ovarian cancer cells We sought to evaluate the molecular events resulting in increased alternative splicing and expression of PKCδVIII in SKOV3 and TOV112D. Previously, 4 we demonstrated that alternative splicing of the PRKCD gene in humans results in two variants: PKCδI and PKCδVIII as shown in the schematic ( Figure 1A above). Using overexpression and knockdown studies, we had previously demonstrated that the splice factor SRSF2 (also known as SC35) regulated the expression of PKCδVIII in neuronal cells. 14 In this study using gel PCR, our results ( Figure 7A) demonstrate a marked increase in SRSF2 in SKOV3 and TOV112D cancer cells compared to T80 normal cells which were concurrent with increased expression of PKCδVIII. To evaluate whether increased expression of SRSF2 in the cancer environment resulted in increase in the pro-survival PKCδVIII expression, we performed knockdown studies. SRSF2 was depleted by transfecting SRSF2 siRNA along with its control siRNA into SKOV3 and Splicing minigenes are valuable molecular tools to identify splice factors that regulate alternative splicing and enable replication of the splicing event without influence of other endogenous factors. PKCδI mRNA is produced by utilization of 5′ splice site I (5′ SSI) and PKCδVIII mRNA is produced by utilization of 5′ splice site II (5′ SSII) on exon 10 of PKCδ pre-mRNA. We previously cloned a human PKCδ splicing minigene to elucidate the mechanisms governing PKCδVIII splicing. Briefly as depicted in schematic of Figure 7C, human PKCδ exon 10 (including both 5′ splice sites) and 200bp of its flanking 3′ and 5′ intronic sequences were cloned into the splicing vector pSPL3 between the splice donor (SD) and splice acceptor (SA) exons (minigene was verified by restriction digestion and sequencing as described in Ref. [14]). To determine that SRSF2 is critical for PKCδVIII expression, the SRSF2 binding site on PKCδ (close proximity to 5′ SSII; denoted by ∆ in schematic) was mutated in the minigene (as described in methods and Ref. [14]). The PKCδ splicing minigene (pSPL3_PKCδ) and SRSF2 binding site mutated PKCδ splicing minigene (pSPL3_PKCδSRSF2) was transfected into T80, SKOV3 and TOV112D for 24 h. PCR was performed using primers for SD and SA exons. The schematic in Figure 7C shows the primer positions (depicted by arrows on SD and SA exons) and the expected spliced products. Splicing of SD to SA exons serves as constitutive splicing control. Results ( Figure 7C) demonstrate that the ovarian cancer cell lines SKOV3 and TOV112D showed increased utilization of 5′ splice site II (which produces PKCδVIII mRNA) in the splicing minigene compared to T80 cells. Further, mutation of the SRSF2 binding site on the splicing minigene significantly decreased utilization of 5′ splice site II in T80, SKOV3, and TOV112D. Interestingly in T80 and TOV112D, the mutated pSPL3_PKCδSRSF2 minigene showed a concurrent increase in utilization of 5′ splice site I, showing that a change in balance of bound splice factors in these cells promoted the alternate splicing event. These results show that the increased expression of PKCδVIII in ovarian cancer cell lines SKOV3 and TOV112D is regulated by splice factor SRSF2. Finally, to validate that SRSF2 binding regulates PKCδVIII expression, we used 2′-methoxy-ethyl-modified (2′MOE), RNAse-H-resistant antisense oligonucleotides. We previously identified and validated an antisense oligonucleotide spanning the SRSF2 binding site 14 which inhibits binding of SRSF2 to exon 10 of PKCδVIII pre-mRNA. 50nM SRSF2 site antisense oligonucleotide (ASO) was transfected into SKOV3 (SKOV3 ASO ) and TOV112D (TOV112D ASO ) cells or nonspecific scrambled antisense oligonucleotide (control) was transfected for 24 h. RNA was isolated and real-time SYBR Green PCR was performed using primers specific for PKCδI and PKCδVIII. Results ( Figure 7D) demonstrate that ASO masking SRSF2 binding site decreased PKCδVIII levels in SKOV3 and TOV112D; the levels of PKCδI or SRSF2 did not change with ASO treatment. \| Ovarian cancer tissue explants show increased expression of PKCδVIII compared to normal ovarian tissue To validate our findings in human ovarian explants, we analyzed the expression profile of PKCδVIII using TissueScan ovarian cancer tissue array from Origene Technologies (catalog # HORT103). This array comprises F I G U R E 6 Depletion of PKCδVIII in ovarian cancer cells decreases Bcl2 and cellular migration, viability, and proliferation. 25 nM of PKCδVIII specific siRNA was transfected in SKOV3 (SKOV3 PKCδVIIIsi ) and TOV112D (TOV112D PKCδVIIIsi ) or 25nM control siRNA (control) was transfected into the cells for 48 h. (A) RNA was isolated and Real-time SYBR Green qPCR was performed for PKCδVIII, PKCδI, Bcl2, and βactin. Graphs show relative quantification of PKCδVIII, Bcl2, and Bcl-xL normalized to βactin with control set as reference. The experiments were independently repeated four times with similar results. Statistical analysis was performed using t-test; p < 0.01, p < 0.001 and ns = not significant. (B) Whole cell lysates were harvested and Western blot was performed using antibodies as indicated. The experiments were independently repeated four times with similar results. (C) SKOV3 and TOV112D were grown to confluency with μ-Dish inserts (Ibidi solutions, 81176), PKCδVIII was depleted by siRNA transfection for 48 h. Inserts were removed to make consistent gaps and cells were imaged at 0, 24, and 48 h. The red lines are drawn on the image to aid in visualization of the edges of the gaps. The experiments were independently repeated six times with similar results. Cells were imaged using Keyence BZx-810 microscope and analysis of gap area between cells (μm 2 ) was quantified using FastTrack A1 software (Ibidi solutions™). Statistical analysis was performed using two-way ANOVA; p < 0.01 and p < 0.001. (D) Cell migration assay using a transwell insert was performed as described in methods. After 18 h, cells in the bottom well were stained with crystal violet and imaged using Keyence BZx-810 microscope. Cells were counted using the Keyence Analyzer software. The experiments were independently repeated three times with similar results. Statistical analysis was performed using one-way ANOVA; p < 0.001. (E) Cell viability was determined using WST1 assay by measuring the formazon dye produced using a plate reader. The experiments were independently repeated four times with similar results. Statistical analysis was performed using t-test; p < 0.001. (F) Cell proliferation was determined using BrdU assay by incubating cells in BrdU for 2 h, then fixing and staining with anti-BrdU antibody. Cells were then stained with peroxidase secondary antibody and measured by plate reader. The experiments were independently repeated four times with similar results. Statistical analysis was performed using t-test; p < 0.001 of normal (Stage 0) and tumor samples (Stage I-IV) from individual patients. The cDNA was analyzed using SYBR Green qPCR in triplicate, normalized with βactin and absolute quantification (AQ) was calculated from the PKCδVIII standard curve. The results ( Figure 8A \| DISCUSSION An important mechanism of regulating gene expression is alternative splicing which expands the coding capacity of a single gene to produce different proteins with distinct functions. 17 Sequencing of the human genome predicts >90% of genes are alternatively spliced. Alternative splicing can occur through various mechanisms such as exon skipping, exon inclusion, alternative 3′ splice site usage, alternative 5′ splice site usage, or alternative polyadenylation site usage. Splicing of exons is modulated by transfactors (splice factors) recognizing the cis-elements on the pre-mRNA and this mechanism is tightly regulated. Under normal conditions, alternative splicing is tissue-specific, developmental-specific or hormone sensitive. However, mutations in a gene on the cis-elements that bind to splice factors can result in alternative splicing and production of proteins that causes diseases such as spinal muscular atrophy and frontotemporal dementia with parkinsonism linked to chromosome 17 (FTDP-17). 18,19 In other F I G U R E 7 Alternative splicing of PRKCD gene is regulated by SRSF2 in ovarian cells. (A) RNA was isolated from T80, SKOV3, and TOV112D ovarian cells and PCR was performed using primers for PKCδI, PKCδVIII or SRSF2. PCR products were separated on a 1% agarose gel and visualized using ethidium bromide. The experiments were independently repeated four times with similar results. Graphs show relative quantification of densitometric analysis of SRSF2 normalized to βactin or percent PKCδVIII (PKCδVIII expression in total PKCδ) was calculated using the following equation: percent PKCδVIII = [PKCδVIII / (PKCδVIII + PKCδI)] × 100. Statistical analysis was performed using t-test; p < 0.01 and p < 0.001. (B) 50 nM of SRSF2 specific siRNA was transfected in SKOV3 (SKOV3 SRSF2si ) and TOV112D (TOV112D SRSF2si ) or 50 nM control siRNA (control) was transfected into the cells for 48 h. RNA was isolated and Real-time SYBR Green qPCR analysis using primers specific to PKCδVIII, PKCδI, SRSF2, and βactin. Graphs show relative quantification of PKCδI, PKCδVIII, and SRSF2 normalized to βactin with control set as reference. The experiments were independently repeated four times with similar results. Statistical analysis was performed using one-way ANOVA; p < 0.01, *p < 0.001, and ns = not significant. (C) Schematic of PKCδ splicing minigene (pSPL3_PKCδ) with PKCδ exon10 containing 5′ splice site I (5′SSI) and 5′ splice site II (5′SSII) cloned between constitutive splicing donor (SD) and splice acceptor (SA) exons. The SRSF2 binding site on PKCδ exon 10 (close proximity to 5′ SSII; denoted by ∆ in schematic) was mutated to generate pSPL3_PKCδSRSF2. T80, SKOV3, and TOV112D were transfected with pSPL3_PKCδ or pSPL3_PKCδ**SRSF2 for 24 h. RNA was isolated and PCR was performed using primers specific to SD and SA exons (indicated by arrows in schematic of minigene). The expected products are shown as SD-SA (constitutive splicing), SSI (inclusion of exon 10 using 5′ splice site I which produces PKCδI), and SSII (inclusion of exon 10 using 5′ splice site II which produces PKCδVIII scenarios such as cancer or diabetes, the disease causes an increase in expression of splice factors or its phosphorylation states which then results in tighter binding of the splice factor to the cis-element thereby increasing the expression of specific splice variants. Examples of alternative splicing in cancer include p53 and estrogen receptor genes in breast cancer 20,21 and Bclx in nonsmall cell lung cancer (NSCLC). 22 In this study, we have demonstrated the novel presence of the pro-survival, alternatively spliced PKCδVIII in human ovarian cells. Further, we demonstrated that the expression of PKCδVIII is significantly increased in ovarian cancer, compared to normal ovarian cells. Using splicing minigenes and knockdown studies, we show that alternative splicing of PKCδVIII is regulated by splice factor SRSF2 in ovarian cells. SRSF2 is expressed at high levels in ovarian cancer cells SKOV3 and TOV112D resulting in the increased expression of PKCδVIII. We previously demonstrated that overexpression of PKCδVIII increases the human telomerase reverse transcriptase (hTERT) expression and increases telomerase activity. 23 Telomerase adds six nucleotide repeats at the end of telomeres to maintain chromosome length. We had demonstrated that high expression of PKCδVIII leads to lower senescence and indicated increased growth potential. In this study, our results demonstrate that PKCδVIII regulates the expression of Bcl2 and overexpression of PKCδVIII affects cell survival and proliferation in ovarian cells. Other groups [24][25][26] have previously shown that hTERT expression increased the expression of Bcl2 conferring resistance to apoptosis. The role of Bcl2 in promoting cell survival in tumors and hematological malignancies has been extensively studied. It is established that Bcl2 and its family of proteins are critical to tumor development, maintenance as well as resistance to cancer drugs. Our results show TOV112D cells have marked elevated levels of PKCδVIII and Bcl2 in concurrence with a previous study that showed high levels of Bcl2 in TOV112D. 27 Cisplatin is a first-line platinum-based drug used for the treatment of ovarian cancer. Wu et al have reported that TOV112D have acquired cisplatin resistance due to increased levels of Bcl2. 27 Our results suggest that PKCδVIII may contribute to increased cisplatin resistance in cancer. Protein kinase CδI, the ubiquitously expressed splice variant predominantly referred to as PKCδ in literature, is proteolytically cleaved at the V3 hinge domain in response to apoptotic stimuli by caspase 3. The cleaved catalytic domain promotes apoptosis in several cellular systems. [28][29][30] We have demonstrated that in humans, the caspase 3 recognition site in the hinge domain is disrupted by inclusion of 31 amino acids due to alternative splicing generating PKCδVIII. We have demonstrated that PKCδVIII independently functions to promote survival in neuronal cells. 4 Here, our results showed increased expression of PKCδVIII in ovarian cancer cells and thus, we sought to elucidate its function. Using robust cellular and molecular techniques, we have established that PKCδVIII promotes cellular survival and proliferation in ovarian cancer cells and additionally established the mechanism by which expression of PKCδVIII is increased in ovarian cancer cells. Our results showed that depletion of PKCδVIII in ovarian cancer cells also reduced their migration. There are several other assays that measure cancer cell invasion, adhesion and motility that determine cancer progression. We are pursuing this separately in conjunction with in vivo extrapolations while this project focused on the mechanisms underlying increased PKCδVIII expression. PKC isozymes have a significant role in signaling cascades in cancers as well as metabolic diseases. Earlier publications have showed the aberrant expression of PKC isoforms such as PKCα, PKCβ, PKCε, PKCδ, and PKCι in ovarian cancer cells [31][32][33][34][35][36][37][38][39] and demonstrated their roles either as tumor suppressers or as an oncogenic kinases. PKCδ has been reported as promoting apoptosis 32,40 as well as promoting tumor growth and metastasis. 6,41 A significant aspect of understanding opposing roles is the presence of alternatively spliced variants of PKCδ that were not evaluated in these previous studies. The tumor suppressor role of PKCδ in literature may be explained by the presence of the pro-apoptotic PKCδI splice variant. Other signaling kinases such as AKT, CLK are upregulated in cancer which also leads to hyper-phosphorylation of splice factors and changes in alternative splicing. It is also reported that AKT phosphorylates Bad to promote its dissociation from Bcl2 thereby inducing cell survival in cancer cells. [42][43][44] The regulation of expression of Bcl2 by PKCδVIII may be at the transcriptional level, posttranscriptional regulation via splice factors or rate of turnover. This avenue is being investigated in our lab. Our results show that overexpression of PKCδVIII causes dissociation of Bcl2-Bad complex. We detected phosphorylation on S112 on Bad in T80 which is increased with overexpression of PKCδVIII. However, in SKOV3 and TOV112D cells S112 phosphorylation was low; but we detected high levels of S136 phosphorylation of Bad and did not observe any direct effects of knockdown of PKCδVIII on phosphorylation of S136 on Bad indicating that Bad phosphorylation may not be a direct mechanism through which PKCδVIII promotes survival in ovarian cancer cells. In our co-immunoprecipitation assays, we could not get a robust band of Bad in the Western blot within Bcl2 or Bcl-xL immunoprecipitations. This suggests that the epitope of the Bcl2 and Bcl-xL antibodies used in the immunoprecipitations was in the proximity of Bad binding site. However, immunoprecipitation with Bad followed by western blot analysis showed association with Bcl2 and PKCδVIII. Our results demonstrate that PKCδVIII increases expression of Bcl2 and further PKCδVIII is a strong binding partner of Bcl2-Bcl-xL complex thereby promoting pro-survival pathways in ovarian cancer cells. Our results show localization of PKCδVIII is distributed between the cytoplasm and perinuclear region in T80 cells while PKCδVIII was detected predominantly in the perinuclear region and nucleus in the cancer cell lines SKOV3 and TOV112D. Previous study 45 has shown that a nuclear localization signal of six basic amino acids is present on the C terminus of PKCδ. These amino acids are conserved in PKCδVIII and likely contribute to the perinuclear and nuclear localization of PKCδVIII. Our results in human ovarian tissue samples validated that dramatically elevated levels of PKCδVIII are present in ovarian cancer compared to normal ovarian tissue. Several publications have discussed the role of PKCδ in cancer; however, this is the first report demonstrating the role of an alternatively spliced variant, PKCδVIII, in ovarian cancer. Our results underscore the importance of signaling kinases and their role in the cancer microenvironment and further demonstrates that targeting a pivotal signaling kinase is a robust approach to attenuate increased cancer cell survival. Our data using antisense oligonucleotides (ASO) demonstrate that the levels of PKCδVIII can be modulated by ASOs in ovarian cancer. This splice variant-specific modulation is highly significant as it enables translation of the therapeutic into clinical use where it could possibly be combined with other chemotherapeutic agents to combat complex cancers. ASOs have been successfully used 46 in treatment of diseases such as Duchenne muscular dystrophy, 47,48 spinal muscular atrophy, 49 familial hypercholesterolemia. 50 The data presented here demonstrate that the signaling kinase PKCδVIII is a viable target to develop therapeutics to combat progression of ovarian cancer. On a broader level, the splice variant PKCδVIII is specific to humans (the mouse homolog is PKCδII with distinct genetic sequence) and is also expressed in other tissues such as adipose and neuronal as demonstrated by us. 4,51 PKCδVIII promotes survival and its elevated expression may indicate a disease state such as cancer or obesity. We have identified and elucidated the function of the human specific PKCδVIII splice variant in ovarian cells which will significantly aid researchers to develop therapies that can be translated into clinical studies for treatment of ovarian cancer patients.	2021-11-12T16:19:43.231Z	2021-11-10T00:00:00.000	{ "year": 2021, "sha1": "581d538c8cb811228e70de9082e79f1dab4ed407", "oa_license": "CCBY", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1096/fba.2021-00090", "oa_status": "GOLD", "pdf_src": "Wiley", "pdf_hash": "a6d9f0362319f39d023b4c1eb27e38003bb1888c", "s2fieldsofstudy": [ "Medicine", "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
239844154	pes2o/s2orc	v3-fos-license	Structure in motion: visual motion perception as online hierarchical inference Identifying the structure of motion relations in the environment is critical for navigation, tracking, prediction, and pursuit. Yet, little is known about the mental and neural computations that allow the visual system to infer this structure online from a volatile stream of visual information. We propose online hierarchical Bayesian inference as a principled solution for how the brain might solve this complex perceptual task. We derive an online Expectation-Maximization algorithm that explains human percepts qualitatively and quantitatively for a diverse set of stimuli, covering classical psychophysics experiments, ambiguous motion scenes, and illusory motion displays. We thereby identify normative explanations for the origin of human motion structure perception and make testable predictions for new psychophysics experiments. The proposed online hierarchical inference model furthermore affords a neural network implementation which shares properties with motion-sensitive cortical areas and motivates a novel class of experiments to reveal the neural representations of latent structure. Introduction Efficient behavior requires identification of structure in a continuous stream of volatile and often ambiguous visual information. To identify this structure, the brain exploits statistical relations in velocities of observable features, such as the coherent motion of features composing an object (Fig. 1a). Motion structure thus carries essential information about the spatial and temporal evolution of the environment, and aids behaviors such as navigation, tracking, prediction, and pursuit [1][2][3][4][5][6][7][8]. It remains, however, unclear how the visual system identifies a scene's underlying motion structure and exploits it to turn noisy, unstructured, sensory impressions into meaningful motion percepts. In recent years, Bayesian inference has provided a successful normative perspective on many aspects of visual motion perception [9][10][11][12][13][14][15][16][17]. Human perception of motion stimuli spatially constrained by an aperture is well-explained by Bayesian statistical inference [9][10][11]14], and neural circuits that integrate local retinal input into neural representations of motion have been identified [18][19][20][21][22][23]. For the perception of structured motion spanning multiple objects and larger areas of the visual field, however, a comprehensive understanding is only beginning to emerge [15,[24][25][26][27]. While "common fate", that is, the use of motion coherence for grouping visual features into percepts of rigid objects, received some experimental support [24,28], the perception of natural scenes requires more flexible structure representations (e.g., nested motion relations and non-rigid deformations) than common fate alone. Recent theoretical work [15] has introduced a representation of tree structures for the mental organization of observed velocities into nested hierarchies. Theory-driven experiments subsequently demonstrated that the human visual system indeed makes use of hierarchical structure when solving visual tasks [16], and that salient aspects of human motion structure perception can be explained by normative models of Bayesian inference over tree structures [17]. Because these studies were restricted to modeling motion integration only with regard to the perceptual outcome-they analyzed presented visual scenes offline using ideal Bayesian observer models-it remained unclear how the visual system solves the chicken-and-egg problem of parsing (in real time) instantaneous motion in a scene while simultaneously inferring the scene's underlying structure. We address this question by formulating visual motion perception as online hierarchical inference in a generative model of structured motion. The resulting continuous-time model is able to explain human perception of motion stimuli covering classical psychophysics experiments, ambiguous motion scenes, and illusory motion displays. The model, which relies on online Expectation-Maximization [29][30][31], separates inference of instantaneous motion from identifying a scene's underlying structure by exploiting the fact that these evolve on different time-scales. The resulting set of interconnect differential equations decomposes a scene's velocities with the goal of minimizing prediction errors for subsequent observations. Beyond capturing human percepts in many psychophysics experiments qualitatively, the model explains human motion structure classification quantitatively with higher fidelity than a previous ideal observer-based model [17]. Furthermore, the model provides a normative explanation for the putative origin of human illusory motion perception, and yields testable predictions for new psychophysics experiments. Finally, we address how motion structure discovery could be supported by neural circuits in the brain. Studying the neural representations underlying motion structure perception is challenging, as the perceived structure often has no direct physical counterpart in the environment (e.g., the concept of a flock velocity in Fig. 1a). We derive a recurrent neural network model that not only implements the proposed online hierarchical inference model, but shares many properties with motion-sensitive middle temporal area (MT) [21] and dorsal medial superior temporal area (MSTd) [19,32]. The network model in turn allows us to propose a new class of stimuli for neuroscientific experiments that make concrete predictions for neural recordings. Results In what follows, we first present the online model for simultaneous hierarchical inference of instantaneous motion and of the scene's underlying structure. Next, we demonstrate the model's ability to explain human motion perception across a set of psychophysics experiments and discuss testable predictions for novel studies. Finally, we propose a biologically realistic neural implementation of online hierarchical inference and identify targeted experiments to reveal neural representations of latent structure. Online hierarchical inference in a generative model of structured motion A structural understanding of the scene in Fig. 1a requires the observer to decompose observed velocities of objects or their features into latent motion sources, s, that, together, compose the scene (Fig. 1b). These latent sources might or might not have a direct counterpart in the physical world. In Fig. 1b, for instance, each bird's velocity on the observer's retina can be decomposed into the observer's self-motion, s self , the flock's motion, s shared , plus a smaller, animal-specific component, s ind . Here, "flock motion" is an abstract mental concept that is introduced to organize perception, but doesn't have an immediate physical correlate. A correct decomposition leads to motion sources that aid interpretation of the visual scene, and thus supports behaviors such as navigation, tracking, prediction and pursuit. Such decomposition requires knowledge of the scene's structure, like the presence of a flock and which birds it encompasses (Fig. 1c). Wrong structural assumptions might lead to faulty inference of motion sources, like wrongly attributing the flock's motion in the sky to self-motion. Thus, the challenge for an observer is to simultaneously infer motion sources and structure online from a stream of noisy and ambiguous visual information. We formalized the intuition of structured motion in the generative model shown in Fig. 1d-g. The stochastic model, first introduced in [16], accommodates fundamental principles of physics (isotropy and inertia) and psychophysics (continuity of trajectories [33] and slow-velocity priors [9]), without making assumptions on specific object trajectories. For example, the motion of three flocking birds viewed by a stationary observer (motion tree in Fig. 1d) can be decomposed into four independent motion sources-one shared (magenta) and three individual (green, one per bird)-that evolve according to Ornstein-Uhlenbeck processes [34], generating smooth motion with changes typically occurring at time scale τ s (Fig. 1e). The resulting speed (absolute velocity) distribution of each motion source is governed by an associated motion strength, λ, such that the expected speed is proportional to λ. The observable velocities, v t , are in turn noise-perturbed (noise magnitude σ obs ; Fig. 1g) sums of the individual motion sources (collected in vector s t ), with the contribution of each individual motion source specified by a different column of the component matrix C (see Fig. 1f). This formalizes the intuition that observable velocities are the sum of their ancestral motion sources in the tree. In this model, the structure of a scene is fully characterized by the vector of motion strengths, λ = (λ 1 , .., λ m , .., λ M ), which describe the presence (λ m >0) or absence (λ m =0) of motion components, as well as their typical speed. In other words, given a reservoir of components, C, which might have been learned to occur in visual scenes in general, knowing λ is equivalent to knowing the motion structure of the scene. Inferring this structure in turn becomes equivalent to inferring the corresponding motion strengths. An agent faces two challenges when performing inference in this generative model (Fig. 1h). First, inference needs to be performed on the fly (i.e., online) while sensory information arrives as an ongoing stream of noisy velocity observations. Second, how observed motion is separated into latent motion sources, s, and motion structure, λ, is inherently ambiguous, such that inference needs to resolve the hierarchical inter-dependence between these two factors. We address both challenges by recognizing that motion structure, λ, typically changes more slowly than the often volatile values of motion sources, s, facilitating the use of an online Expectation-Maximization (EM) algorithm to infer both. This separation of time scales yields a system of mutually dependent equations for updating λ and s and furthermore affords a memory-efficient, continuous-time online formulation that is amenable to a neural implementation (see Methods for an outline of the derivation, and Supplemental Information, Section 3 for the full derivation). While the algorithm is approximate, it nonetheless performs adequate online hierarchical inference and closely resembles more accurate solutions, even for deeply nested motion structures (see Supplemental Figure S1). 2/24 Our online model computes, at any time, a posterior belief over the latent motion sources, s t , which is Gaussian with mean vector µ t and covariance matrix Σ t , as well as an estimate, λ t , of the underlying structure. The dynamics of µ t , Σ t , and λ 2 t (the inference is more elegantly formulated on the squared values) read: and The coupled equations (1)-(3) support the following intuition. Eqn. (1) calculates a running average of the motion strengths λ 2 t by use of a low-pass filter with time scale τ λ . Here, denotes element-wise multiplication and the function f Σ (λ 2 t ) (Fig. 1i) estimates the variance of the s-posterior distribution according to an adiabatic approximation (cf. eqn. (3), see Methods). The constants α and β contribute a sparsity-promoting prior, p(λ 2 ), for typical values of the motion strengths (see Methods for the their full expressions). By eqn. (2), the motion source means µ t are estimated by a slightly different low-pass filter that relies on a prediction error, t , between the model's expected velocities, C µ t , and those actually encountered in the input, v t (both normalized by observation noise variance to facilitate the later network implementation). This prediction error on observable velocities is transformed back to the space of latent motion sources via the transposed component matrix C T and then, importantly, gated by element-wise multiplication ( ) with the variance estimates f Σ (λ 2 t ). This gating implements a credit assignment as to which motion source was the likely cause of observed mismatches in t , and thus uses the scene's currently inferred motion structure to modulate the observed velocities' decomposition into motion sources. For flocking birds, for example, a simultaneous alignment in multiple birds' velocities would only be attributed to the shared flock velocity if such a flock had been detected in the past (λ shared large, and λ ind small). Otherwise it would be assigned to the birds' individual motions, s ind . Together, eqns. (1) and (2) implement a coupled process of structure discovery and motion decomposition, which distinguishes them through different time-scales. Notably, the proposed model is not a heuristic, but is derived directly from a normative theory of online hierarchical inference. Next, we explored if the model can explain prominent phenomena of human visual motion perception. Online inference replicates human perception of classical motion displays To explore if the proposed online model can qualitatively replicate human perception of established motion displays, we simulated two classical experiments from Gunnar Johansson [25] and Karl Duncker [35]. These experiments belong to a class of visual stimuli which we refer to as object-indexed experiments (Fig. 2a) because the observed velocities, v t , 4/24 belong to objects irrespective of their spatial locations. (Location-indexed experiments will be discussed below.) In Johansson's experiment, three dots oscillate about the screen with two of the dots moving horizontally and the third dot moving diagonally between them (see Fig. 2b and Supplemental Video S1). Humans perceive this stimulus as a shared horizontal oscillation of all three dots, plus a nested vertical oscillation of the central dot. Similar to previous offline algorithms [15], our online model identifies the presence of two motion components (Fig. 2c): a strong shared motion strength, λ shared (magenta) and weaker individual motion, λ ind , for the central dot (green). The individual strengths of the outer two dots (light and dark green), in contrast, decay to zero. Most motion sources within the structure are inferred to be small (dotted lines in Fig. 2d). Only two sources feature pronounced oscillations: the xdirection of the shared motion source, µ shared, x , (magenta, solid line) and the y-direction of the central dot's individual source, µ ind, y , (green, solid line), mirroring human perception. As observed velocities are noisy, they introduce noise in the inferred values of µ t , which fluctuate around the smooth sine-functions of the original, noise-free stimulus. As expected from well-calibrated Bayesian inference, the magnitude of these fluctuations is correctly mirrored in the model's uncertainty, as illustrated by the posteriors' standard deviation f Σ (λ 2 t ) (shaded areas in Fig. 2d). In the second experiment, known as the Duncker wheel, two dots follow the motion of a rolling wheel, one marking the hub, the other marking a point on the rim (Fig. 2e). The two dots describe an intricate trajectory pattern (see Fig. 2f and Supplemental Video S2), that, despite its impoverished nature, creates the impression of a rolling object for human observers, a percept that has been replicated by offline algorithms [15]. Likewise, our online model identifies a shared (magenta in Fig. 2g) plus one individual (dark green) component, and decomposes the observed velocities into shared rightward motion plus rotational motion for the dot on the rim (see Fig. 2h). Notably, the shared motion component is discovered before the revolving dot's individual motion, leading to a transient oscillation in the inferred shared motion source, µ shared (see light magenta trace in Fig. 2h) -an onset effect that could be tested experimentally. In summary, the online hierarchical inference model successfully identified the structure underlying the motion displays, provided Bayesian certainty estimates for the inferred motion, and replicated human perception in these classical psychophysics experiments. Online inference outperforms ideal observers in explaining human structure perception Having qualitatively replicated motion structure inference in common motion displays, we next asked if our online model could quantitatively explain human motion structure perception. To address this question, we reevaluated behavioral data from Yang et al. [17], where participants had to categorize the latent structure of short motion displays (see Fig. 3a). Motion scenes followed one of four structures (Fig. 3b) and were generated stochastically from the same generative model underlying our hierarchical inference model. Owing to their stochastic generation, scenes often were ambiguous with regard to their latent structure, prompting distinct error patterns in human responses (see confusion matrix in Fig. 3c). For instance, independently moving dots were more frequently misclassified as clustered motion (I-C element) than vice versa (C-I element), global motion was highly recognizable, and nested hierarchical motion was more frequently misperceived as clustered than as global. To test if human responses arise from normative, Bayesian motion structure inference, Yang et al. modeled these responses in two steps (blue branch in Fig. 3d): first, an offline Bayesian ideal observer, which was provided with the trajectories of all objects within a trial, calculated the likelihood for each of the four structures. Then, these four probabilities were fed into a choice model with a small set of participant-specific fitting parameters (see Methods). This model captured many aspects of human responses, including task performance, typical error patterns, singletrial responses, and participant-specific differences. Yet, the model arrived at these probabilities by comparing the likelihoods of the full sequences for all four candidate structures, and so had no notion of how a percept of structure could emerge over the course of the trial. Thus, we next asked if our online model, which gradually infers the structure during the stimulus presentation, was better able to account for the observed response pattern. As our model by design inferred real-valued motion strengths λ rather than only discriminating between the four structures used in the experiment, we added an additional stage that turned the inferred motion strengths into a likelihood for each of the four structures at trial end (red branch in Fig. 3d, see Methods). To do so, we computed five hand-designed features from the seven-dimensional vector λ t (besides one global and three individual strengths, there are three possible two-dot clusters), and trained a multinomial logistic regression classifier on the features to obtain likelihood values for each of the structures. The classifier was trained on the true structures of the trials, and thus contained no information about human responses. Finally, we fitted the same choice model as Yang et al. to the participants' responses. The confusion matrix predicted by our model shows an excellent agreement with human choices, both when averaged across participants (Fig. 3e), and on a per-participant basis (see Supplemental Figure S3 and S4). Indeed, our model beats the original computational model in terms of response log-likelihoods for all of the 12 participants (see Fig. 3f; p<0.001, two-sided paired Wilcoxon signed-rank test). Furthermore, the online model overcomes the systematic under-estimation of global motion (G-G matrix element) that previous, ideal observer-based approaches suffered from [16,17]. Importantly, in our model, any information connecting the stimulus to the eventual choice is conveyed through the motion strengths, λ t , as a bottleneck. The fact that the online hierarchical inference-based approach describes human responses better than the ideal observer-based model of Yang et al. indicates that our model may share mechanistic features with the human perceptual apparatus. Explaining motion illusions that rely on spatial receptive fields In contrast to the object-indexed experiments discussed above, another class of psychophysics experiments employs velocity stimuli that remain at stationary locations (see Fig. 4a), typically in the form of apertures of moving dots or drifting Gabors. This class, which we refer to as location-indexed experiments, is furthermore popular in neuroscience as it keeps the stimulus' local visual flow within an individual neuron's spatial receptive field throughout the trial [21]. We investigated our model's ability to explain illusory motion perception in two different types of locationindexed experiments: motion direction repulsion in random-dot kinematograms (RDKs) [36][37][38][39][40][41], see Fig. 4, and noise-dependent motion integration of spatially distributed stimuli [42,43], see Fig. 5. We modeled perception in these experiments by including a self-motion component and added a vestibular input signal to the observables (see Fig. 4b, and cf. Fig. 1a-c). The vestibular input, which we fixed to have zero mean plus observation noise, complemented the visual input, which is ambiguous with regard to self-motion and globally Figure 4. Hierarchical inference explains motion illusions in location-indexed experiments. (a) In location-indexed experiments, motion flow is presented at stationary spatial locations. (b) Considered latent motion components. Self-motion, which affects all retinal velocities in the opposite direction (-1) integrates both visual input and a vestibular signal (here: zero + noise). (c) Perceived object velocities, relative to the environment, are the sum of all inferred motion components excluding self-motion. (d) In motion direction repulsion experiments, two groups of dots move at constant velocity with opening angle γ. (e) The direction in which human perception of the opening angle is biased depends on the true opening angle. Black dots: human data, reproduced from [36]. Purple line: model precept. Insets: the model's inferred motion decomposition. (f) Varying the contrast of one dot group modulates the biased percept of the angle of the other group. Purple: model percept for γ=45°, qualitatively matching data from [37]. Blue: predicted inversion of the bias for smaller opening angles. (g) Same as panel f, but for varying the speed of the second group. Purple: model percept for γ=60°, qualitatively matching data from [38]. Dashed blue: model percept for γ=90°, qualitatively matching data from [36]. Solid blue: predicted biphasic function for smaller opening angles. (h)-(l) Extended experiment from [39] which surrounds the two central RDKs with additional RDKs in an annulus. The hierarchical inference model replicates human perception in various conditions. (h) A surround with dots moving vertically both up-and downwards ("bi-directional surround" in [39], indicated by orange arrows in the top-left sketch's annulus) causes no repulsion in the perceived directions of horizontally moving RDKs in the center (darker orange arrows in the top-left sketch's center). Our model replicates this perception as shown in the histogram of 200 trial repetitions. (i) Coherently moving annulus RDKs cause the perceived inner velocities to be biased away from the surround direction. (j) For diagonally moving inner RDKs, the same coherent downward surround has no noticeable effect. (k) Neither does a bi-directional surround bias the percept of diagonally moving inner RDKs. (l) An upward surround, in contrast, biases the percept of the inner RDKs to close-to-horizontal motion. shared object motion and can induce illusory self-motion ("vection") [44][45][46]. In turn, we model the subjectively perceived velocity of objects, relative to the stationary environment, as the sum of all inferred motion sources excluding self-motion (see Fig. 4c and Methods). In the RDK experiment, a participant fixates the center of an aperture in which two groups of randomly positioned dots move linearly with opening angle γ (see Fig. 4d) and subsequently reports the perceived opening angle. Motion direction repulsion occurs if the perceived angle is systematically biased relative to the true opening angle. As previously reported, the repulsion bias can change from an under-estimation of the opening angle for small angles to an overestimation for large angles (data from [36] reprinted as black dots in Fig. 4e). We replicated this effect by simulating two constant dot velocities with opening angles that varied across trials. Our model decomposed the stimulus into self-motion, shared motion and individual (group) motion. Across opening angles, it featured a triphasic psychometric function with angles smaller than~40°being under-estimated, angles between~40°and~110°b eing over-estimated, and even larger angels being unbiased (purple curve in Fig. 4e). The match with human biases 7/24 arose without systematic tuning of simulation parameters (the simulations presented in this manuscript were mostly performed with a set of default parameters, see Methods). Inspecting the model's inferred motion components revealed that, for small γ, the negative bias arose from integrating all dots into a single, coherent motion component while disregarding individual dot motions (left inset in Fig. 4e). Intermediate γ, in contrast, caused the shared component to be correctly broken up into two individual components-plus a small illusory self-motion component (right inset in Fig. 4e). This self-motion, which is ignored in the perceived velocities, widened the perceived opening angle between the two groups of dots. For even larger γ, the illusory self-motion vanished yielding unbiased percepts. For fixed opening angles, motion direction repulsion is furthermore modulated by relative contrast and speed difference between the two motion components. Specifically, for an opening angle of γ=45°, Chen et al. [37] has shown that increasing the contrast of one dot group inflates the perceived opening angle-here measured relative to horizontal to separate cause and effect-of the other, constant-contrast group (Fig. 4f, left). We replicated this effect in simulations that operationalized visual contrast as an (inverse) multiplicative factor on the observation noise variance, σ 2 obs . For an opening angle of γ=45°, our model featured a positive and monotonically increasing repulsion bias as the second group's contrast increases (purple line in Fig. 4f, right), similar to what has been previously reported. For smaller opening angles, in contrast, our model predicts an inversion of the repulsion bias, which first decreases at low contrast and then increases again for higher contrast (blue line in Fig. 4f, right)-a prediction that remains to be tested. Increasing the speed of one motion component for large opening angles also introduces a positive bias in the perceived opening angle of the other component in human participants [36,38]. We replicated this effect by increasing the second group's speed, which, for a γ=90°opening angle, yielded a relatively stable bias of~5°across different motion speeds (dashed line in Fig. 4g), in line with the aforementioned experimental data from Braddick et al. [36] and, for a γ=60°opening angle (purple line in Fig. 4g), qualitatively replicated the initial rise and then gradual decline in the bias, as reported for this opening angle by Benton and Curran [38]. Furthermore, our model predicts that the speed-dependent bias changes to a biphasic curve for smaller opening angles (blue line), providing another testable prediction. Extending the basic MDR experiment from Fig. 4d, Takemura et al. [39] investigated how motion in a surrounding annulus affects the perceived directions of "inner" RDKs, see sketch in the top left of Fig. 4h. Two inner RDKs move to the left and right, respectively, while two additional RDKs in the annulus move up and down, respectively. For this stimulus human observers show no direction repulsion [39]. We simulated this extended MDR experiment with our hierarchical inference model by extending the motion tree of Fig. 4b to include two group components for the outer and inner RDKs, respectively, on the third level, and four individual components (one per RDK) as leaves, on the forth level (cf. Supplemental Figure S5). Across 200 simulated trials (see Methods), the distribution of inner RDK directions perceived by the model at trial end (see histogram in Fig. 4h) match the reported unbiased perception of humans. Our model was further able to replicate human perceptual biases for various other combinations of dot motion in the inner and surrounding RDKs explored by Takemura et al. (see Fig. 4i-l, and Supplemental Figure S5 for example trials). The percepts to all combinations are qualitatively replicated by our model. When both surrounding RDKs move downward, as shown in Fig. 4i, the perceived motion of the inner RDKs is slightly biased upward. The reason for the bias in the model's percept is a small illusory self-motion component in upward direction which necessitates a slight diagonal upward tilt of the inner RDKs' individual motions for explaining their horizontal retinal velocities. When modifying the stimulus such that the inner RDKs move diagonally with a 90 degree opening angle (see Fig. 4j-l), human and model percepts remain unbiased in the case of downward (Fig. 4j) and bi-directional surrounding motion (Fig. 4k). In both cases, the directional contrast of the presented velocities obviates the illusory identification of self-motion, thereby implicating unbiased percepts of the model. If, however, the surrounding RDKs move upwards, strong direction repulsion on the inner dots was reported [39] leading to their perceived motion to become almost horizontal (Fig. 4l). In the model, this effect originates from illusory downward self-motion arising from the general alignment of the presented velocities. Overall, our hierarchical inference model replicated biased and unbiased perception across a variety of stimulus conditions. Turning to noise-dependent motion integration of spatially distributed stimuli, we investigated a motion illusion by Lorenceau [42] which has received little attention in the literature (see Fig. 5). Two groups of dots oscillate in vertical and horizontal orientation, respectively (see Fig. 5a and Supplemental Video S3). Both groups follow sine-waves with identical amplitude and frequency, but maintain a relative phase shift of π/2 that is consistent with an imaginary global clockwise (CW) rotation (indicated by a gray arrow in Fig. 5a). This stimulus can be considered location-indexed, as the small oscillation amplitude of less than 1 degree of visual angle caused the stimulus to conveniently fit into the receptive fields of individual neurons of the human homologue of area MT [47]. Interestingly, the stimulus' percept changes once disturbances orthogonal to the axes of oscillation are added (called "motion noise" in [42], see combined to a rigidly moving object according to "common fate", and both groups are perceived as moving separately. Their movement, however, is not perceived as strictly vertically and horizontally, but rather the stimulus induces an impression of slight counter-clockwise (CCW) rotation, that is, "opposite to veridical" [42]. With motion noise, in contrast, the percept switches in two ways: all dots appear to move coherently along a circle, and the perceived direction of movement becomes CW. These percepts are illustrated in Fig. 5b. Applied to this stimulus, our model replicates the perceived rotation direction reversal with increased motion noise, which we simulated through an increase in the observation noise σ 2 obs . Specifically, the model's perceived velocities for both groups of dots featured a slight global CCW rotation on top of two generally separated groups for the noise-free stimulus, and a single global CW rotation once observation noise is increased (Fig. 5c). Inspecting the model's motion decomposition provides a possible answer to how this flip in perceived rotation emerges, which is illustrated in Fig. 5d by the example of the horizontal group. On noise-free presentation, dot motion was decomposed into clockwise rotating self-motion (golden arrow) plus a chghorizontally elongated, yet slightly CCW rotating group motion (green arrow), leading to the transparent CCW motion percept. Once observation noise increased, the inferred motion structure discarded the separated groups in favor of a single global motion component (magenta), leading to the percept of coherent CW rotation for all dots (see Supplemental Figure S6 for trajectories of the motion strengths and sources under both conditions). Object recognition and perceptual switching of nested structure-from-motion displays Motion relations do not only aid dynamic tasks, such as tracking and prediction, but also provide essential cues for object recognition. Structure-from-motion (SfM), the perception of 3D objects from 2D visual displays, is well-studied in psychology [48][49][50][51][52][53][54] and neuroscience [55][56][57][58]. We asked whether our model can support SfM perception and replicate the salient phenomenon of perceptual switching when presented with ambiguous stimuli (see Fig. 6a). Furthermore, using the model, we identified novel SfM displays of nested objects which could inspire future psychophysics experiments studying how structure interacts with perceptual ambiguity. Typical SfM displays, like the point cloud-cylinder in Fig. 6a and Supplemental Video S4, involve rotational motion in three dimensions, contrasting with the translational motion in two dimensions considered so far. Our generative model supports such 3D rotation in location-indexed experiments: as illustrated in Fig. 6b, introducing a rotational motion source, s rot , which describes the cylinder's angular velocity around the y-axis, yields a linear dependence of the observed retinal velocities on s rot at every input location (dashed orange circles) owing to the locations' fixed coordinates. Thus, rotational motion is supported naturally by the component matrix, C, (cf. Fig. 1e) and integrates without any changes into our hierarchical inference model. Ambiguous SfM displays, such as the considered frontal view of a rotating cylinder, furthermore feature equivocal correspondences of spatially overlapping inputs to the cylinder's surface at the front and back. Mentally assigning the overlapping left-and rightward retinal velocities to their depth locations is key to forming a coherent percept of the 3D object. To support such percepts in our model, we added a basic assignment process: spatially overlapping velocities are assigned to their depth location on the cylinder (front or back) such that the assignment locally minimizes the model's prediction error, t , in eqn. (2). Furthermore, this assignment is independently re-evaluated at each input location with a uniform probability in time (see Methods). We tested the model's ability to perceive SfM by using a motion tree with self-motion, rotational motion and individual motion (see Fig. 6c, and Supplemental Figure S7 for a control simulation with more motion components). As shown in Fig. 6d, the model swiftly identifies rotational motion across all input locations at a constant angular speed, matching the human percept of a rotating cylinder. Subsequently, the percept switches randomly and abruptly between CW and CCW rotation, with inter-switch-intervals following a Gamma distribution (see Fig. 6e). The resulting stochastically switching percepts with typical durations of a few seconds match the reported bistable perception of humans [53,57]. To explore how more complex structures could interact with SfM perception, we asked how our model interprets the rotation of nested point-cloud cylinders (see Fig. 6f and Supplemental Video S4) Their rotation is easily identified by humans [49], and features a more complex structure than basic SfM displays that only require a single rotational motion source. To present this stimulus to the model, the extended graph in Fig. 6g features rotational sources not only for the inner and outer cylinders (light and dark blue, respectively), but also the possibility of shared motion (magenta) affecting both cylinders. Where both cylinders overlapped, the assignment now minimized the prediction error over 4 overlapping retinal velocities (24 possible combinations per location), but remained otherwise unchanged. When both cylinders rotated with the same angular velocity of 90°/s, the model inferred a single shared rotational component (see Fig. 6h) leading to the impression of rigid rotation in which perceptual switches occur simultaneously for both cylinders. Identifying a structure with a single component rather than separate rotations for both cylinders is the result of the model's preference of simple structures. Increasing the angular velocity of the inner cylinder by 50% to 135°/s (see Fig. 6i) did not change the model's percept of a rigidly shared rotation, but led to a slightly higher perceived speed of rotation. Inspecting the inference process revealed that the assignment process often assigned fast-moving dots of the inner cylinder to the outer cylinder and, vice versa, slower moving outer dots to the inner cylinder. This assignment yielded a sufficiently coherent interpretation of all retinal velocities as originating from a single rotation (within the bounds of perceptual acuity, σ obs ) for the model to prefer the simpler structure. Finally, a display in which the outer cylinder rotates faster than the inner cylinder (135°/s and 90°/s, respectively; see Fig. 6j) changed the model's inferred structure to perceiving different rotational speeds for both cylinders. Yet, even though each cylinder had its distinct perceived rotation, their rotational directions remained aligned and perceptual switches still occurred simultaneously, a perceptual linkage known from related experiments [54]. The nested SfM displays in Fig. 6f-j provide testable predictions for future psychophysics studies (see Supplemental Video S4 for a video of all conditions). The model's percepts across all conditions matched the percept of the authors. Experimental predictions from a biological network model of hierarchical inference Finally, we asked whether and how a biologically plausible neural network could implement our online hierarchical inference model. To this end, we devised a recurrent neural network model of rate-based neurons. Naturally, such modeling attempt relies on many assumptions. Nonetheless, we were able to identify several experimentally testable predictions that could help guide future neuroscientific experiments. Following Beck et al. [59], we assumed that task-relevant variables can be decoded linearly from neural activity ("linear population code") to support brain-internal readouts for further processing, actions and decision making. Furthermore, we employed a standard model for the dynamics of firing rates, r i (t), and assumed that neurons can perform linear and quadratic integration [59][60][61][62]: with time constant τ i , activation function f i (·), weight vector w i and matrix Q (i) for linear and quadratic integration, respectively. The rate vector, r(t), here comprises all presynaptic firing rates, including both input and recurrent populations. With these assumptions, we derived a network model with the architecture shown in Fig. 7a, which implements the online model, given by eqn. (1)-(3), via its recurrent interactions and supports linear readout of all task-relevant variables. That is, for every task-relevant variable, x, there exists a vector, a x , such that x = a T x r (see Supplemental Information, Section 5 for the derivation). The network consists of three populations. The input population (bottom in Fig. 7a) encodes the observed velocities, v t /σ 2 obs , and observation precision, 1/σ 2 obs , in a distributed code. While any code that supports linear readout of these variables could serve as valid neural input, we chose a specific model that, as shown below, captures many properties of motion-sensitive area MT. The distributed population (center in Fig. 7a) simultaneously represents the squared motion strengths, λ 2 t , mean of the sources, µ t , and prediction errors, t , in a distributed code with linear readout. For those, almost arbitrary readouts suffice, such that we chose randomly generated readout vectors, a. Notably, we propose the prediction errors, t , to be linearly decodable, which allowed eqn. (2) to be implemented with the neuron model in eqn. (4) (see Supplemental Information, Section 5.3 and 5.4). All neurons in the distributed population have simple activation functions, f i (·), that are linear around some baseline activity. The linear decodability of λ 2 t , µ t , and t are testable predictions. Finally, the 1-to-1 population (top in Fig. 7a) represents the uncertainty, Σ= f Σ (λ 2 ), in a one-to-one mapping, r m ∝ f Σ (λ 2 m ), with r m being the firing rate of either a single cell or, more likely, a small population. The theoretical motivation behind this representation is two-fold: on the one hand, the non-linear form of f Σ (·) prevents a distributed, linearly decodable representation (see Supplemental Information, Section 5.5); on the other hand, the particular shape of f Σ (λ 2 m ), shown in Fig. 1i, mirrors the typical activation function of Type-I neurons [63], such that 12/24 the proposed representation emerges naturally for the activation function, f Σ (a T λ 2 m r), in the 1-to-1 population (using the fact that λ 2 m can be read out neurally with weights w = a λ 2 m ). Overall, the network structure predicts λ 2 t , µ t , and t to be linear decodable, and the components of f Σ to be independently encoded in single neurons or small neural populations. Even though the network model supports both the object-indexed and location-indexed experiments from Fig. 2 -4, the retinotopic organization of the early visual system [21,64] brings a location-indexed perspective closer in line with our understanding of how the cortex encodes visual information. Furthermore, as we show in Supplemental Information, Section 2.5, our model can be extended to support motion sources in polar coordinates (see Fig. 7b), such that it supports salient real-world retinal input motifs, such as rotation and radial expansion/contraction about the fovea. (Note that rotation and expansion on the retina are conceptually distinct from the cylindrical rotation, s rot , in structure-from-motion, discussed earlier.) Representations of angular motion, s ϕ , and radial motion, s r , can also coexist with translational motion (i.e., linear motion in Cartesian coordinates) within the same population. Selective neural response to rotation, expansion/contraction and translation, as well as combinations thereof, such as spiraling, has been frequently reported in the dorsal medial superior temporal area (MSTd) [19,65]. Before demonstrating this capability in simulations, let us provide further information about the model's input population, and how it relates to known properties of area MT. To do so, consider the location-indexed stimulus in Fig. 7b. During fixation, each aperture stimulates a population in retinotopically organized, motion sensitive area MT [21]. Neurons in MT are tuned to respond preferentially to a certain direction and speed (Fig. 7c), such that the full population jointly covers all velocities in a polar grid [66,67]. The response of individual neurons to velocities within their spatial receptive field is commonly modeled by a log-normal function for speed [67] and a von Mises function for direction [68], leading to the bump-like response function shown in Fig. 7d. As a third factor, higher visual contrast (smaller σ 2 obs ) leads to higher firing rates [69]. As we derive in Supplemental Information, Section 5.6, a neural population with these response functions supports linear readout of input velocities, v t /σ 2 obs , and precision, 1/σ 2 obs , in Cartesian coordinates. This provided us with a biologically realistic and, at the same time, theoretically grounded input population model which we used in the following network simulations. We tested the network's ability to perform online hierarchical inference in the simulation shown in Fig. 7e-j. To challenge the network, we employed a stimulus that combined shared rotation and shared translation (motion tree in Fig. 7e). Six input populations with receptive fields shown in Fig. 7f projected to a distributed population of 100 neurons and a 1-to-1 population of size 8 (one per motion strength). After one second of retinal velocities of counterclockwise rotation (Fig. 7f, left), these velocities switched to clockwise rotation (center), followed by a superposition of clockwise rotation and rightward translation (right). As the network response for the three populations to this stimulus shows (Fig. 7h-j), input neurons fired sparsely and were only active if the stimulus matched their preferred direction or speed. Neurons in the distributed population, in contrast, showed fluctuating activity with little apparent structure, and exhibited population-wide transients upon changes of the input. Finally, the 1-to-1 population responded more graded and with a short delay, suggesting that every rate, r m , describes a small cortical population rather than individual neurons. Knowledge of the (randomly drawn) vectors, a x , of the simulated network, allowed us to read out the network's latent motion decomposition at each time point (solid lines in Fig. 7g). This revealed that the network correctly decomposed the input, including the overlaid rotational and translational motion, and closely matched the online model (dotted lines). In experiments with humans and animals, we have no access to these readout vectors, a x . We therefore simulated a possible experiment that tests our model and doesn't require this knowledge (see Fig. 7k-m), while benefiting from precise stimulus control. Several apertures, located at the receptive fields of recorded neurons in motion sensitive areas (e.g., area MT or MSTd), present a motion stimulus according to the generative model from Fig. 1. Velocities across the apertures are positively correlated owing to a shared motion source, but also maintain some individual motion (see Fig. 7k and Supplemental Video S5). A series of trials varies the fraction of shared motion in the stimulus, q := λ 2 shared /(λ 2 shared +λ 2 ind ), ranging from almost independent motion (Fig. 7l, left) to almost perfect correlation (right). According to the network model, λ 2 can be read out linearly. For the simulation in Fig. 7m, we presented the network with trials of seven values of q. We then trained a linear regression model to predict q from the neural activity for the two most extreme structures (blue dots in Fig. 7m), and decoded q for the intermediate structures using this regression model (red dots in Fig. 7m). Owing to the stochastic stimulus generation, the network's motion structure estimates, λ t , fluctuate around the true strength-yet, on average, the trained linear readout correctly identified the fraction of global motion in the stimulus. This is a strong prediction of the network model, which could be tested in a targeted neuroscientific experiment. Discussion We have proposed a comprehensive theory of online hierarchical inference for structured visual motion perception. The derived continuous-time model decomposes an incoming stream of retinal velocities into latent motion components which in turn are organized in a nested, tree-like structure. A scene's inferred structure provides the visual system with a temporally robust scaffold to organize its percepts and to resolve momentary ambiguities in the input stream. Applying the theory to human visual motion perception, we replicated diverse phenomena from psychophysics in both object-indexed and location-indexed experiment designs. Furthermore, inspection of the model's internal variables provided normative explanations for putative origins of human percepts and spawned concrete predictions for new psychophysics experiments. Finally, the online inference model afforded a recurrent neural network model with visual inputs reminiscent of cortical area MT and latent structure representations reminiscent of area MSTd. Our online model shares features with predictive coding [70,71], a theory positing that "higher" brain areas provide expectations to "earlier" areas in a hierarchical model of sensory input and that neural processing aims to minimize prediction errors between top-down expectations and bottom-up observations. Like predictive coding, the dynamics in eqn. (2) update the values of motion sources to minimize prediction errors, t , within the bounds imposed by the identified structure. Yet, structure identification according to eqn. (1) follows a different principle by computing a running average of motion source magnitudes. This contrasts with common theories of predictive coding in the brain [72,73], which assume that the same computational principle is repeated across cortical hierarchies, and demonstrates how hierarchical visual processing could combine multiple interacting algorithmic motifs. Moreover, the network model in Fig. 6a challenges the prevalent view [72,74] that error signals are necessarily represented by distinct neural populations (or alternatively distinct dendritic compartments [75]). While our network model supports the possibility of distinct error populations, we show that prediction errors could also be computed and conveyed by the same neurons representing other quantities, such as the motion sources, µ t , and even the structure, λ 2 t , using a distributed neural code. In the main text, we have for the sake of clarity limited the presentation of the theory to a basic version that nonetheless covers all essential concepts. In Supplemental Information, Section 4, we present several extensions that are naturally covered by our model: (i) observation noise, σ obs , can be time-and object-dependent, which is relevant for modeling temporary occlusion of a subset of stimuli; (ii) observation noise can be non-isotropic (different values in xand y-direction), which is relevant for angle-dependent edge velocities in apertures [76]; (iii) for optimal inference, different motion components can feature different time constants, since velocity is expected to change more slowly for heavy objects due to higher inertia; (iv) different motion components may tend to co-occur or exclude one another in real-world scenes, which can be modeled by an interaction prior of pairwise component compatibility; and (v) when motion components are not present for a long time, they will decay to zero, preventing their rediscovery, which can be mitigated by a prior on motion strengths. The current theory is limited to velocities as input, thereby ignoring the well-documented influence of spatial arrangement on visual motion perception, such as center-surround modulation [77,78], adjacency [26] or motion assimilation [79], as well as Gestalt properties [80]. Furthermore, the model does not solve the correspondence problem in object-indexed experiments, but simply assumes that velocities are correctly assigned to the input vector as objects move about the visual field. For location-indexed experiments, we have explored how structure inference in concert with a basic assignment process, which minimizes the observer's local prediction errors, could solve the correspondence problem during structure-from-motion perception. Our work focuses on the simultaneous inference of motion sources, s t , and motion strengths, λ t . Other quantities, such as time constants and, probably more importantly, the motion components, C, have been assumed to be given. It is worth noting, however, that gradient-based learning of C is, in principle, supported by the theory on long time scales (see Supplemental Information, Section 4.5). Finally, limited experimental evidence of the neural correlates of motion structure perception required the neural network model to rely on many modeling assumptions. The model's predictions should act as a starting point for further scientific inquiry of these neural correlates. Even though the sensory processes underlying object-indexed motion perception necessarily differ from those of location-indexed perception, our model describes human perception for both types of experiments. Thus, both types might share the same underlying neural mechanisms for structure inference. This raises the intriguing question whether there exist stable, object-bound neural representations of velocity. Furthermore, our work points towards a tight link between neural representations of latent structure and representations of uncertainty in that the estimated motion strengths, λ t , determine the credit assignment of prediction errors through the gating function, f Σ (λ 2 t )-a function that also computes the variance of motion components, e.g., the brain's uncertainty about flock velocity. Behaviorally, sensory noise directly impacts the perceived structure of a scene as demonstrated experimentally by the perceptual reversal in the motion illusion of Lorenceau [42] (cf. Fig. 5). More generally, our theory predicts that 14/24 the visual system will organize its percepts into simpler structures when sensory reliability decreases. Moreover, the reliability of visual cues plays a role in multisensory integration [81], with area MSTd [82,83], but not area MT [84], exhibiting tuning to vestibular signals. Thus, MSTd may be a candidate area for multisensory motion structure inference. Overall, we expect our theoretical results to guide targeted experiments in order to understand structured visual motion perception under a normative account of statistical information processing. Methods In what follows, we provide an overview of the generative model, the online hierarchical inference model, the computer simulations, and the data analysis. A more detailed presentation is found in the Supplemental Information. Generative model of structured motion. We consider K observable velocities, v k,d (t), in D spatial dimensions. For notational clarity, we will consider in this Online Methods section only the case D=1 and use the vector notation, v t =(v 1 (t), .., v K (t)) T . The extension to D>1 is covered in Supplemental Information, Section 2.4. Observable velocities, v t , are generated by M latent motion sources, s m,d (t), abbreviated (for D=1) by the vector s t =(s 1 (t), .., s M (t)) T . Velocities are noisy instantiations of their combined ancestral motion sources, v t ∼ N C s t , σ 2 obs /δt I , where C km = + 1, −1, and 0 in K × M component matrix, C, denote positive, negative and absent influence, respectively. For the formal definition, observations, v t , remain stable within a short time interval [t, t+δt), and the observation noise variance, σ 2 obs /δt, ensures a δt-independent information content of the input stream. In the online inference model, below, we will draw the continuous-time limit, which will become independent of δt. In computer simulations, δt is the inverse frame rate of the motion display (default value: 1/δt=60 Hz). Each motion source (in each spatial dimension) follows an Ornstein-Uhlenbeck process, ds m = −s m /τ s dt + λ m dW m , with time constant τ s , motion strength λ m (shared across dimensions), and Wiener process W m . The OU process's equilibrium distribution, N 0, τ s 2 λ 2 m , introduces a slowvelocity prior which, as we note, has recently been proposed to originate from the speed-contrast statistics of natural images [85]. The resulting marginal stationary velocity distribution of v k is v k ∼ N 0, σ 2 obs /δt + τ s 2 ∑ M m=1 C 2 km λ 2 m . Radial and rotational motion sources. In location-indexed experiments, the input's location (e.g., a neuron's receptive field) remains fixed. For D=2, the fixed input locations enable our model to support rotations and expansions around various axes. In this manuscript, we consider two cases: rotation around a vertical axis (SfM experiment in Fig. 6) and rotation/expansion around the fovea (network model in Fig. 7). For rotations around a vertical axis, each input v k has fixed polar coordinates (R k , ϕ k ) as sketched in Fig. 6b. When describing rotation by means of a rotational motion source, s rot t , we obtain for the noise-free part of the observed velocity in Cartesian coordinates: v k,x =−R k sin(ϕ k ) s rot t , v k,y =0, and v k,z =−R k cos(ϕ k ) s rot t . Owing to the linear dependence of v k on s rot , we can include the coefficients as a column in component matrix, C, and s rot as a motion source in the vector s t . Note that in the SfM experiments only the x-and y-directions are observed. Similarly, for rotation/expansion around the fovea, each input v k has fixed polar coordinates (R k , ϑ k ) with radial distance R k and angle ϑ k , relative to the pivot point (we use different symbols than for vertical rotation for notational clarity). Denoting radial and rotational motion sources by s r and s ϕ , we obtain for the noise-free part of v k in Cartesian coordinates: v k,x = s r cos ϑ k − s ϕ R k sin ϑ k , and v k,y = s r sin ϑ k + s ϕ R k cos ϑ k . Since R k and ϑ k are fixed coefficients, the mapping (s r , s ϕ ) → (v k,x , v k,y ) is linear and, thus, can be described by the component matrix C. The full derivation and an illustration of the velocity relations in polar coordinates are provided in Supplemental Information, Section 2.5. Online inference. The goal of motion structure inference is to simultaneously infer the value of motion sources, s t , and the underlying structure, λ, from a stream of velocity observations. The number of spatial dimensions, D, component matrix, C, time constant τ s , and observation noise σ obs are assumed to be known. The EM algorithm leverages that changes in s t and λ (if changing at all) occur on different time scales, τ s and τ λ , respectively. For τ λ τ s , the EM algorithm treats λ as a constant for inferring s t (E-step), and optimizes an estimate, λ t , online based on the inferred motion strengths (M-step). E-Step. For fixed λ, the posterior p( s t \| v 0:t ; λ ) is always a multivariate normal distribution, N (µ t , Σ t ), and can be calculated by a Kalman-Bucy filter [86,87]; see Supplemental Information, Section 3.1, 3.2, and 3.3.1 for the derivation. This yields coupled differential equations for the time evolution of µ t and Σ t . To reduce the computational complexity of the system, we perform an adiabatic approximation on the posterior covariance, Σ t , by assuming (a) that it has always converged to its stationary value, and (b) that off-diagonal values in Σ t are zero, that is, we ignore correlations in uncertainty about latent motion sources in the posterior distribution. As shown in the full derivation in Supplemental Information, Section 3.3, the first assumption is warranted because the stationary value of Σ t depends only on the current structure estimate, λ t ; then, because Σ t decays to stationarity at time scale τ s /2, it can always follow any changes in λ t which happen at time scale τ λ τ s . The second assumption is a modeling assumption: that biological 15/24 agents might disregard the subtle (and complicated) interactions between the uncertainties of different motion sources and rely on their individual uncertainties, instead. Using the two assumptions we derive a closed-form solution for the posterior variance, with c m 2 = ∑ K k=1 C 2 km denoting the vector-norm of the m-th column of C. This is eqn. (3) of the main text. The plot in Fig. 1i has parameters c m 2 =4, τ s =300 ms, and σ obs =0.05. By plugging the adiabatic approximation of the variance into the time evolution of µ t , we arrive at eqn. (2) of the main text (see Supplemental Information, Section 3.3.4 for the derivation). M-step. Using the posterior from the E-step, motion strengths, λ, are optimized to maximize the likelihood of the observed velocities. This optimization further incorporates prior distributions, p(λ 2 m ), most conveniently formulated over the squared motion strengths, for which we employ a scaled inverse chi-squared distribution, owing to its conjugacy to estimating the variance of s m (this is what λ 2 m controls). The prior features two hyperparameters, ν m and κ 2 m , which give rise to an intuitive interpretation as ν m pseudo-observations of average value κ 2 m . The partition function, A(ν m , κ 2 m ), only serves for normalization. By default, we employ a Jeffreys prior (ν m =κ 2 m =0), which is a typical choice as a non-informative prior in Bayesian statistics and promotes a preference for finding simple structures by assigning higher beliefs to small values of λ m (and highest to λ m =0). The only exception is the motion strength assigned to self motion, λ self , for which we employ a uniform prior distribution, formally by setting ν self = − 2 and κ 2 self = 0. These choices reflect the a-priori belief that motion components supported by C will usually be absent or small in any given scene-with the exception of self-motion-induced velocity on the retina, which occurs with every saccade and every turn of the agent's head (see Supplemental Information, Section 3.1.2 for the formal calculation of the M-step). In the online formulation of EM (see Supplemental Information, Section 3.2.3 and 3.3.4 for the derivation of the online EM algorithm and of the online adiabatic inference algorithm which constitutes our model, respectively), these priors give rise to the low-pass filtering dynamics in eqn. (1) for updating λ 2 m , with constants This completes the derivation of the online model for D=1 spatial dimensions. The extension to multiple dimensions is straightforward and provided in Supplemental Information, Section 3.1.3, 3.2.3 and 3.3.4 alongside the respective derivations. Preference for simple structures. The above Jeffreys prior on motion strengths, p(λ 2 m ), facilitates the discovery of sparse structures. This property is important when a large "reservoir" of possible motion components in C is considered: the model will recruit only a small number of components from the reservoir. In Supplemental Figure S2, we demonstrate this ability for the example of the Johansson experiment from Fig. 2b-d by duplicating the shared motion component, i.e., the first two columns in C are all 1's. As Supplemental Figure S2 shows, the model recruits only one of the two identical components and discards the other. This example of identical components in the reservoir represents the theoretically hardest scenario for maintaining a sparse structure. For the numerical simulation, input was drawn with observation noise variance σ 2 obs /δt, at the time points of input frames (every δt). The drawn input remained stable until the next frame. Between frames, the differential equations for online hierarchical inference were integrated with SciPy's explicit Runge-Kutta method "RK45" which adapts the step size. This integration method combines numerical accuracy with a parameterization that is almost invariant to the input frame rate. The default parameters that we used are listed in Table 1. Hierarchical motion experiments (Fig. 2). For the Johansson experiment, all K =3 dots followed sinusoidal velocities with frequency 0.5 Hz. Horizontal amplitudes were 2 √ τ s for all dots; vertical amplitudes were 0 for the outer dots and cos(45 • ) · 2 √ τ s for the inner dot. For the Duncker wheel, we set the wheel radius to R=1 and the rotation frequency to 1 Hz. This leads to the hub velocity v hub, y =0 and v hub, x =2π s −1 because the hub must travel 2πR during one period for slip-free rolling. For the rim velocities, being the derivatives of location, we thus find v rim, x =v hub, x + R ω cos(ω t) and v rim, y = − R ω sin(ω t), with ω =2π s −1 . For the simulation, we increased the observation noise to σ obs =0.15 and set λ m (t=0)=0.1 to highlight the gradual discovery of the motion components. 16/24 Structure classification (Fig. 3). The stimulus data and human responses were released by [17] on GitHub. The experiment is described in detail in [17]. There were 12 participants with each participant performing 200 trials. Each trial consisted of three dots moving on a circle for 4 s. Dots had different colors to prevent their confusion, but colors did not convey any information on the dots' roles within the structure. No data was excluded. Trials were generated stochastically from the same generative model that is considered in this work, with uniform probability for each of the four structures (Independent, Global, Clustered, Hierarchical) to underlie the trial. Motion strengths were chosen such that all dots had identical marginal velocity distributions, p(v k ), across all structures-leaving motion relations as the only distinguishing information (see [17], for detailed stimulus parameters and λ-values of all structures). Like [17], we treated the experiment as one-dimensional (D=1), operating directly on the angular velocities. Noise-free angular velocities were calculated from the circular distance of subsequent stimulus frames, and we set 1/δt=50 Hz to match the experiment's frame rate. For presenting the trials to our online inference model, we initialized each of the λ m at its average value (average taken across the ground truth of all structures). At trial end, the model yielded M=7-dimensional λ-vectors associated with 1 shared component, 3 cluster components (one per possible pair), and 3 individual components (see Supplemental Figure S3 for example trials). For logistic regression, we calculated 5 features, T i , from λ, namely: "Does shared motion stand out?" (9) λ m "Does one cluster dominate the others?" denotes the dot not being in cluster c. The features were hand-designed based on the reasoning that they may be useful for structure classification. Their most important property is that all information about a trial is conveyed through λ as a bottleneck. A multinomial logistic regression classifier was trained with L1-regularization on the feature vectors, (T 1 , .., T 5 ), to classify the ground truth structures of the trials. Then, we fitted the same choice model as [17] to the human choices, but replaced the ideal observer log-probability, log p( S \| v 0:T ), which was used in [17], with the class probability from the trained classifier, log p( S \| λ ): with lapse probability, π L , inverse temperature, β, and biases, b S , for all structures, S = G, C, H, relative to the independent structure (b I =0 by convention). Note that, in contrast to [17], we do not need to consider structure multiplicities here because the features are already symmetric with regard to the three possible cluster assignments. Like [17], we did not apply observation noise to the presented velocities, but maintained a non-zero observation noise parameter, σ obs , for the inference. Observation noise, σ obs , and lapse probability, π L , were shared parameters for all participants and were fitted jointly via a simple grid search. We obtained σ obs =0.04 and π L =4% (compared to 14% in [17]). The remaining 4 parameters, {β, b G , b C , b H }, were fitted via maximum likelihood for each participant. All reported confusion matrices and log-likelihoods were obtained by fitting the 4 per-participant parameters using leave-one-out cross-validation. The log-chance level in Fig. 3f is 200 · log(1/4) since each participant performed 200 trials. Location-indexed experiments (Fig. 4, Fig. 5, and Fig. 6). To support self-motion, we introduce a column of −1's in C as an additional component, which is connected to all visual velocity inputs and to a vestibular input v vst . In our simulations, the vestibular input is always stationary, but noisy: v vst ∼ N 0, σ 2 vst . The associated self-motion strength, λ self , uses a uniform prior (see discussion under eqn. (6)). Perceived velocities are the sum over all-except-self-motion: v perceived = ∑ m =self C * m µ m . (Fig. 4). In the MDR experiments with two RDKs, input was modeled as K =3 velocities: two for the two groups of dots, plus the vestibular input. Repulsion angles were estimated from 20 repetitions of 30 s long trials, with v perceived averaged over the last 10 s of each trial. Error bars from the simulations were too small to be shown in Fig. 4e-g. In Fig. 4e, the velocities for opening angle, γ, were given by (v x , v y )=v 0 · (cos(γ/2), sin(γ/2)) for the first group, with v 0 =2 √ τ s , and v 0 · (cos(γ/2), − sin(γ/2)) for the second group. As in Figure 3 of [36], the repulsion bias was measured with respect to the full opening angle. Motion-direction-repulsion (MDR) experiments In Fig. 4f, increasing contrast of the second group was modeled as dividing the observation noise variance by a factor, f , between 0.001 and 10, leading to variance σ 2 obs / f for this group's input. As in [37], the repulsion bias was measured only with respect to the first group's perceived direction. The expressed similarity to experimental data refers to the "2-motion condition" in Figure 7 of [37]. In Fig. 4g, the velocity of the second group was multiplied by a factor between 0 and 2, and the repulsion bias was measured only with respect to the first group's perceived direction. For a 60°opening angle, we qualitatively replicate the experimental data from Figure 2a&b in [38]. In order to maintain the simulation parameters from previous conditions, we did not attempt to quantitatively match the speed of targets and distractors in [38]. A direct quantitative comparison to the human data from Figure 4b in [36] is difficult because they had measured the point of subjective equality (PSE) to a 90°opening angle for this stimulus condition, finding a 10°bias for the full opening angle. For the Takemura experiment [39] in Fig. 4h-l, we used K =5 inputs: two inner RDKs, two outer RDKs, and the vestibular signal, which were organized in the motion tree shown in Supplemental Figure S5a. If not mentioned otherwise, the simulation parameters matched those from the basic MDR experiment in Fig. 4e. The inner stimuli had v x = ± v 0 , and v y =v 0 if non-zero. The outer stimuli had v x =0, and v y = ±v 0 . The standard deviation of the observation noise of the outer RDKs was divided by factor 6, reflecting that in [39] the outer RDKs' covered a threetimes larger area and had twice the dot density of the inner RDKs. Each histogram is based on 200 trial repetitions, which use identical initial conditions but different realizations of observation noise, with perceived velocities measured at trial end. The conditions in panels Fig. 4h-l correspond to figure panels 4a, 4b, 6 left, 6 center, 6 right, in [39]. Besides transparent motion (i.e., two perceived velocities), Takemura et al. reported also coherent motion (i.e., only one perceived velocity) for the inner RDKs in a fraction of trials. In our computer simulations, we focused only on the biased perception of two velocities. Lorenceau illusion (Fig. 5). For the Lorenceau illusion, we modeled each dot's velocity as a separate input owing to the spatially distributed nature of the stimulus. As in [42], the two groups of 10 dots each oscillated at a frequency of 18/24 0.83 Hz. For the oscillation amplitude, we chose R=1/2 (arbitrary units), leading to velocities v x (t)= R ω cos(ω t) for the horizontal group and v y (t)= − R ω sin(ω t) for the vertical group, with ω = 2π · 0.83 s −1 . As shown in Supplemental Figure S6, the model decomposes this stimulus into a deeply nested structure comprising self-, shared-, group-, and individual motion. For the noise-free stimulus condition, we used the default simulation parameters. For the condition with motion noise, the observation noise, σ obs , of the visual inputs (not the vestibular input) was multiplied by 25. (Fig. 6). We treat SfM as a location-indexed experiment owing to experimental findings [50,52]. For computer simulations, we model each cylinder as a ring in the x-z-plane, conflating its height into one receptive field (the simulations still run in 2D with x-and y-dimensions being modeled) . The outer cylinder had radius R=1.5, and the inner, if present, R=1.0. Normal rotation speed was 90°/s, and fast speed was 135°/s. Velocities were observed at seven equidistant receptive field locations along the x-axis, x RF ∈ {1.2, 0.8, 0.4, . . . , -1.2}. These correspond to angles, ϕ k , on the cylinders via x RF = R cos(ϕ k ) with the inner cylinder covering only five RF locations (cf. Fig. 6b). When presenting velocity observations, v k , each RF location x RF had multiple overlapping v k (2 for one cylinder, 4 for nested cylinders where they overlapped). The observation noise for velocity inputs, σ obs , was multiplied by 20 for the single cylinder-condition and by 30 for the nested cylinders-conditions, reflecting the high local ambiguity when measuring multiple overlapping speeds and directions [88]. For consistency with other simulations, we provided a vestibular input with the same parameters as in previous location-indexed experiments, although this signal plays no computational role in the SfM simulations. Structure-from-motion (SfM) experiments The model's component matrix, C, comprised translational self-motion, rotational motions for the outer cylinder, the inner cylinder (only in nested conditions), and shared for both cylinders (only in nested conditions), as well as translational individual components for each v k (see Fig. 6c and g). Rotational motion is naturally covered by our model as presented in "Radial and rotational motion sources" earlier in Methods and sketched in Fig. 6b. To solve the correspondence problem of overlapping v k , we devised the following assignment process. At every integration time step, δt, and for every RF location, x RF , keep the previous assignment with probability 0.7, and continue to the next x RF . Else, that is if the assignment is re-evaluated, calculate the Euclidean distance between the model's expected velocities, C µ t , and the observed velocities, P j v t , for all permutations, P j , of the overlapping inputs within this RF. Then choose the assignment, P j , that minimizes the Euclidean distance, i.e., the local prediction error, t , within the RF. Once all x RF were processed in this manner, perform the integration of ∂ t µ t according to eqn. (2) using the assigned permutations of v t . This completes the model for SfM perception. We note that, since the integration is performed only after all RF assignments have been made, the resulting global assignment process is independent of the order of iterating over the RFs and could, in principle, be performed in parallel and continuous time. The fact that all computations are spatially confined to information within each RF further improves the process's biological plausibility. For obtaining the switching distribution in Fig. 6e, we performed a 10, 000 s long simulation and followed ideas from [89]: first we identified a "perceptual threshold" as the mode of {\|µ rot t \| ∀ t} (the exact value is actually not important). Then we defined two possible percepts which correspond to positive (negative) values of µ rot t . A perceptual switch occurred whenever µ rot t crossed the negative (positive) threshold of the other percept. The Gamma distribution was fitted by maximum likelihood. Network implementation (Fig. 7). A detailed derivation of how to implement the online hierarchical inference model in a neural network model is provided in Supplemental Information, Section 5. In the following, we will focus on the specific model parameters used in the simulations of Fig. 7. For both simulations (the demonstration in Fig. 7e-j and the proposed experiment in Fig. 7k-m), there were K =6 location-indexed input variables in D=2 spatial dimensions. Input was encoded according to the model of area MT presented in Supplemental Information, Section 5.6. Each velocity, v k , was encoded by a population of 192 neurons, with tuning centers organized on a polar grid with N α =16 preferred directions, and N ρ =12 preferred speeds (sketched in Fig. 7c for smaller values of N α and N ρ ). Each neuron in each of the K populations thus has "coordinates" (n α , n ρ ) describing its preferred direction and speed. To account for the reported bias of MT tuning toward slow speed [67], the density of preferred speeds became sparser for higher speeds, which we modeled in Supplemental Information, eqn. (70) by µ ρ (n ρ ) = ρ min + d ρ n 1.25 ρ , with d ρ = (ρ max − ρ min )/(N ρ − 1) 1.25 , and ρ min =0.1, ρ max =8.0, for neurons n ρ =0, .., N ρ −1. Preferred directions covered the circle equidistantly. The remaining parameters in the tuning function were κ α =1/0.35 2 and σ 2 ρ =0.35 2 for the angular and radial tuning widths, respectively, and ψ=0.1 Hz for the overall firing rate scaling factor. For the network simulations, we increased the frame rate to δt=1/120 Hz for the sake of a higher sampling rate on the x-axis in Fig. 7h-j (the simulation software stores firing rates only at the time of frames). The distributed population comprised 100 neurons. Readout vectors, a x , for all variables represented by this population were drawn i.i.d. from a standard normal distribution, N (0, 1), for each vector element. Adjoint matrices 19/24	2021-10-26T15:08:03.830Z	2021-10-23T00:00:00.000	{ "year": 2022, "sha1": "c94d23768783fd5fc8b4284c7e22f8c596434291", "oa_license": "CCBYNC", "oa_url": "https://www.biorxiv.org/content/biorxiv/early/2021/10/23/2021.10.21.465346.full.pdf", "oa_status": "GREEN", "pdf_src": "BioRxiv", "pdf_hash": "22225ea6c9cf244b8b0724677cc460165beefc01", "s2fieldsofstudy": [ "Psychology", "Computer Science" ], "extfieldsofstudy": [ "Biology" ] }
73613058	pes2o/s2orc	v3-fos-license	Bäcklund Transformations between the KdV Equation and a New Nonlinear Evolution Equation We first give a Bäcklund transformation from the KdV equation to a new nonlinear evolution equation. We then derive two Bäcklund transformations with two pseudopotentials, one of which is from the KdV equation to the new equation and the other from the new equation to itself. As applications, by applying our Bäcklund transformations to known solutions, we construct some novel solutions to the new equation. Introduction In 1895 two Dutch mathematicians Korteweg and de Vries derived a nonlinear wave equation which is now called KdV equation and adopts the canonical form [1] = 6 + . ( Since the KdV equation (1) possesses the solitary wave solution where and 0 are constants, it provides a theoretical confirmation of the existence of the solitary wave observed in 1834 by the Scottish engineer John Scott Russell on the Union canal.In 1965 Zabusky and Kruskal [2] discovered that the interaction of two solitary wave solutions is elastic, and therefore they called this kind of solutions solitons.In 1967 Gardner et al. [3] related the solution of the Cauchy initial value problem for the KdV equation (1) to the inverse scattering problem for a one-dimensional linear Schrödinger equation and derived the analytical expressions of -soliton solutions.The -soliton solutions to the KdV equation ( 1) can also be derived from the Darboux [4] and Bäcklund transformations [5], which have been extended and applied to a large variety of nonlinear evolution equations (see [6][7][8][9][10][11][12][13]). It is well known that the KdV equation ( 1) is connected to the potential KdV equation via V = ∫ d.By use of this connection, Wahlquist and Estabrook [14] obtained the following Bäcklund transformation for the KdV equation (1), as well as for the potential KdV equation (3). Theorem 1 (see [14]).If V is a solution of the potential KdV equation ( 3), then the system on V is integrable, where is an arbitrary constant.Moreover, V also satisfies (3).So the integrable system (4) defines a Bäcklund transformation V → V for the potential KdV equation (3), and it also gives a Bäcklund transformation → for the KdV equation (1) which is defined by Note that (4) is a Lax pair of the potential KdV equation (3).In this paper, we first give another Lax pair of (3) and show that it defines a Bäcklund transformation from the KdV equation (1) to a new nonlinear evolution equation (7) (see Theorem 2).Then by combining the Bäcklund transformations given in Theorems 1 and 2, we derive two Bäcklund transformations defined by two pseudopotentials (see Theorem 8): one is from (1) to (7) (see ( 22)) and the other from (7) to itself (see ( 23)).As applications, by applying our Bäcklund transformations to known solutions of ( 1), (3), or (7), we construct some novel solutions to (7) (see Examples 6,7,11,and 12). Bäcklund Transformation with One Pseudopotential In this section, we prove the following result. Theorem 2. If V is a solution of ( 3), then the system on is integrable.Moreover, = ( + V) 2 satisfies the nonlinear evolution equation Proof.From ( 3) and ( 6) we have Therefore = ; that is, ( 6) is integrable. From the first equation of ( 6) we have Substituting the above equations into the second equation of ( 6) yields that satisfies and therefore So = ( + V) 2 = satisfies (7). To our knowledge, (7) has not been reported in literature previously. Remark 3. If is a solution of (7), then is a solution of (3). Remark 5.The integrable system (6) also defines a Bäcklund transformation with the pseudopotential : which takes a solution of the KdV equation ( 1) to a solution of (7). Substituting a known solution V of (3) into the integrable system (6), one can get a solution = ( + V) 2 of (7).In the next two examples, from stationary and kink solutions of (3), we generate solutions of (7).Example 6. Substituting the stationary solution of the potential KdV equation ( 3) into the integrable system (6) yields the following rational solution of (6), and therefore of ( 7) The profiles of the solution (17) at = −1, 0, 1 are shown in Figure 1.Note that, at ̸ = 0, the solution curve has three pieces; as approaches 0, the piece located in the middle becomes smaller and smaller and finally disappears. Bäcklund Transformations with Two Pseudopotentials By combining Theorems 1 and 2, we have the following result.The proof is obvious (see, for example, [16]). Theorem 8.If V is a solution of ( 3), then the system on and is integrable, where is an arbitrary constant.Moreover, = ( + ) 2 satisfies (7). Remark 10.The integrable system (21) also defines a Bäcklund transformation → with two pseudopotentials and for (7) as follows: Applying the Bäcklund transformation ( 22) or (23) to a known solution of (1) or of (7), it is usually not easy to get explicit formulas for and when solving the integrable system (21).But in this case, one can use an ordinary differential equation solver (see, for example, [13,16]) to obtain numerical approximations of and .In the next two examples, we use this method.	2018-12-29T05:41:14.643Z	2017-01-01T00:00:00.000	{ "year": 2017, "sha1": "7ce178959c6e8ad8561b6cddf3ddf52a5bf9f1f8", "oa_license": "CCBY", "oa_url": "http://downloads.hindawi.com/journals/mpe/2017/5157123.pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "7ce178959c6e8ad8561b6cddf3ddf52a5bf9f1f8", "s2fieldsofstudy": [ "Mathematics", "Engineering" ], "extfieldsofstudy": [ "Mathematics" ] }
18368660	pes2o/s2orc	v3-fos-license	Pseudo-Gaussian quantum models A new family of one-dimensional quantum models is proposed in terms of new potentials with a Gaussian asymptotic behavior but approaching to the potential of the harmonic o scillator when $x\to 0$. It is shown that, in the energy basis of the harmonic oscillator, the matrix elements of the Hamiltonian operators of these new models can be derived from generating functionals. The progress in algebraic and numerical computation offers one new possibilities of analyzing new classical or quantum systems that can not be analytically solved. The Schrödinger equation with an attractive radial Gaussian potential was analyzed with different perturbation methods leading to numerical results of convenient accuracy [1]. Other quantum models with Gaussian barriers were considered in the theory of electronic transport in mesoscopic systems [2,3]. Some time ago we studied a simple one-dimensional relativistic model of Gaussian well using generating functionals for finding the matrix elements of the Klein-Gordon operator in the energy basis of the non-relativistic harmonic oscillator (HO) [4]. In this framework computational methods allowed us to point out the structure of the discrete energy spectra of the Gaussian wells which indicates that inside the well the quantum motion is close to an harmonic one. Here we would like to continue this study in the case of the non-relativistic quantum system, looking for general one-dimensional models with pseudo-Gaussian potentials. Our purpose is to define a large family of models whose potentials have a Gaussian asymptotic behavior but approaching to the HO potential near x ∼ 0. The second objective is to show how the method of generating functionals works in this case, helping us to find the matrix elements of the Hamiltonian operators of our models in the energy basis of HO. The one-dimensional HO of mass m and frequency ω has the well-known Hamiltonian where V 0 is an arbitrary ground energy. The energy is measured in units of ǫ = ω so that we can write using the dimensionless coordinate ξ = x/a measured in units of a = / √ m ǫ, and denoting λ = 2V 0 /ǫ. The number operator N 0 has the eigenvalues 2n + λ + 1 where n = 0, 1, 2... is the quantum number of the discrete energy levels E n = ǫ [n + 1 2 (λ + 1)]. We propose a Gaussian generalization of HO, defining pseudo-Gaussian models (PGM) with new number operators in which the dimensionless potentials are supposed to exhibit Taylor expansions, but without terms proportional with ξ 4 , ξ 6 , ..., ξ 2r . It is not difficult to verify that this condition is accomplished if we take We constructed thus a large family of models, denoted from now by (λ, µ) r , depending on dimensionless parameters, λ ∈ R and µ > 0, and the integer number r = 1, 2, ... which is called the order of PGM, unlike the genuine Gaussian potential that is of the order r = 0. For λ ≥ 0 the potentials (4) are positively defined representing pseudo-Gaussian barriers but when λ < 0 we can speak about pseudo-Gaussian wells of different profiles (Fig. 1). The PGM we consider here have two principal virtues: their potentials (4) have a Gaussian asymptotic behavior (provided µ > 0) and, in the same time, the form of these potentials in a neighborhood of ξ = 0 is similar to that of the HO potential. Moreover, when the order r increases, the potential (4) closes to the HO potential since, according to Eq. (5), near ξ ∼ 0 these potentials differ among themselves only by terms of the order O(ξ 2r+2 ). An important consequence of this property is that In other respects, we observe that the PGM becomes HO in the limit of µ → 0 too, when we find again that W (ξ) → λ + ξ 2 . The Hamiltonian operators of PGM can be written in terms of physical quantities using the parameter ǫ as in the case of H 0 . Thus we have where the physical potential, depends on the physical coordinate x measured in units of a defined before. The structure of the energy spectra is determined by the concrete form of the potential (9). Since all the models (λ, µ) r have potentials that satisfy we can conclude that all of them have continuous energy spectra in the domain S c (H) = [0, ∞). However, the pseudo-Gaussian wells with λ < 0 give rise, in addition, to discrete energy spectra in domains S d (H) = [V min , 0) with V min = V 0 = 1 2 ǫ λ < 0. Obviously, the discrete spectra can have at most a finite number of energy levels. In general, the energy eigenfunctions of PGM as well as the discrete energy levels of the pseudo-Gaussian wells can not be calculated using analytical methods. Therefore, we must look for appropriate perturbation procedures leading to efficient symbolic calculations on computers combined with satisfactory convergent numerical methods. In the case of our PGM it is convenient to use perturbations in the energy basis of HO, {\|n \| n = 0, 1, 2...}, adopting the technique of generating functionals related to the generating functions [5] F τ (ξ) = 1 π which yield the HO eigenfunctions normalized in the ξ-scale as, The matrix elements of an operator X can be derived from the corresponding generating functional, according to the rule In general, the matrix X with the elements (14) is countable since the energy basis of HO is countable. The advantage of this method is that for HO or PGM the integral (13) reduces to some well-known Gaussian integrals. For example, in the simplest case of HO we obtain the functional giving rise to the diagonal matrix elements m\|N 0 \|n = (2n + λ + 1)δ nm . The generating functional of an arbitrary model (λ, µ) r , can be also calculated in terms of Gaussian integrals. Starting with two important terms, and the identity ξ 2k exp(−µξ 2 ) = (−1) k ∂ k µ exp(−µξ 2 ), we obtain the final result wherẽ depends on the coefficients (6). Now a nice exercise is to show directly that The method of generating functionals offers one the opportunity of using computers for finding the matrix elements of the operators N in the energy basis of HO. The first step is to derive the generating functional (19) for a PGM of a given order r calculating the function (20) under Maple or Mathematica. Furthermore, one must choose a concrete model (λ, µ) r , with fixed parameters, for which one has to calculate on computer the matrix elements of N according to the general rule (14). Thus one obtains a finitedimensional block˜ N representing a truncation of the matrix N which is countable. Finally, this block can be used for extracting physical results (e. g. the energy levels in pseudo-Gaussian wells). We note that in this approach the truncation of the matrix N takes over the role of the standard perturbation procedure. Therefore, increasing the dimension of the block˜ N one may improve the accuracy of results. Our preliminary tests on computers show that the numerical results obtained for our PGM are satisfactory convergent. Moreover, interesting behaviors can be observed even in the case of the models of lowest orders. Thus, it seems that the first discrete energy levels of the pseudo-Gaussian wells tend to respect the rules of HO while at the limit of separation between the discrete and continuous spectra we meet again the specific effect reported in [4], namely an inflexion point of the function giving the dependence of the eigenvalues of˜ N on the number of level. Finally, we conclude that the potentials proposed here realize a graduate balance (depending on r) between the short-range HO dynamics and the Gaussian asymptotic behavior. We hope that these models should be useful for designing new quantum electronic devices in semiconductor heterostructures.	2007-06-12T09:38:11.000Z	2007-06-12T00:00:00.000	{ "year": 2007, "sha1": "f48dd6deab34498b3569b6741130e44272774df1", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "f48dd6deab34498b3569b6741130e44272774df1", "s2fieldsofstudy": [ "Physics" ], "extfieldsofstudy": [ "Physics", "Mathematics" ] }
269849858	pes2o/s2orc	v3-fos-license	Spheroids Generated from Malignant Pleural Effusion as a Tool to Predict the Response of Non-Small Cell Lung Cancer to Treatment Background: Spheroids generated by tumor cells collected from malignant pleural effusion (MPE) were shown to retain the characteristics of the original tumors. This ex vivo model might be used to predict the response of non-small cell lung cancer (NSCLC) to anticancer treatments. Methods: The characteristics, epidermal growth factor receptor (EGFR) mutation status, and clinical response to EGFR-TKIs treatment of enrolled patients were recorded. The viability of the spheroids generated from MPE of enrolled patients were evaluated by visualization of the formazan product of the MTT assay. Results: Spheroids were generated from 14 patients with NSCLC-related MPE. Patients with EGFR L861Q, L858R, or Exon 19 deletion all received EGFR-TKIs, and five of these seven patients responded to treatment. The viability of the spheroids generated from MPE of these five patients who responded to EGFR-TKIs treatment was significantly reduced after gefitinib treatment. On the other hand, gefitinib treatment did not reduce the viability of the spheroids generated from MPE of patients with EGFR wild type, Exon 20 insertion, or patients with sensitive EGFR mutation but did not respond to EGFR-TKIs treatment. Conclusion: Multicellular spheroids generated from NSCLC-related MPE might be used to predict the response of NSCLC to treatment. Introduction Despite the significant progress in the management of lung cancer, it remains the leading cause of cancer-related death worldwide [1].A substantial proportion of lung cancer patients had unresectable disease at the time when they were diagnosed and were thus treated with chemotherapeutic agents, targeted therapies, or immunotherapies [2,3].The utilization of targeted therapies and immunotherapies for the treatment of non-small cell lung cancer (NSCLC) has significantly increased in the past few years [3].One of the reasons for this increased utilization of these treatments for NSCLC is the promising anticancer effect with manageable adverse effects shown by numerous clinical trials.Another reason is that the response of NSCLC to these treatments can be predicted by analyzing the gene or protein expression of the cancers [4,5].However, the tests that were used to predict the response of NSCLC to these therapies were not always accurate enough. Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) have significant anticancer effects in advanced-stage lung adenocarcinoma with sensitive EGFR mutations such as exon 19 deletion or L858R mutation, with response rates of around 70~80% [6,7].However, there is still a substantial proportion of patients whose tumor did not respond to EGFR-TKIs treatment despite the presence of these sensitive mutations.Utilization of programmed death-ligand 1 (PD-L1) expression or tumor mutation burden (TMB) of the cancer cells to predict the clinical response of NSCLC to anti-PD-1 or anti-PD-L1 treatment is also not as accurate as expected [8].In addition, there is currently no reliable test to predict the response of NSCLC to traditional chemotherapies. Chemotherapies showed significant anticancer effects in preclinical studies using cancer cell lines grown as monolayers and have been used to treat NSCLC for decades [9].However, the clinical benefit of chemotherapy was not as promising as the anticancer effects found in preclinical studies.One of the reasons for this difference in the anticancer effects of chemotherapies in preclinical studies and their efficacy in clinical practice is that the two-dimensional (2D) culture used in preclinical studies might not accurately represent the actual tumor in patients.The three-dimensional (3D) spheroid culture retains the architecture of the tumor and was shown to be a useful model to investigate the characteristics of the tumor [10][11][12].Primary cultures that used cells collected from malignant pleural effusion (MPE) have been shown to retain the heterogeneity of the tumor [13].Spheroids generated by patient-derived tumor cells (PDCs) collected from MPE have been shown to recapitulate the molecular properties of the original tumors, and implementation of PDCderived drug sensitivities could be used to predict the clinical response to treatment [14,15].These studies showed that spheroids generated from MPE might be a useful tool to predict the response of NSCLC to anticancer treatment.However, these models used multiple experimental procedures to generate spheroids from MPE and to evaluate the response of spheroids to anti-cancer treatment, which might reduce its applicability in clinical practice. In this study, we collected cells from NSCLC-related MPE to generate multicellular spheroids, and then treated these spheroids with EGFR-TKI gefitinib.After treatment, the viability of the cells in spheroids was evaluated.The purpose of this study was to explore the possibility of using multicellular spheroids generated from NSCLC-related MPE as a tool to predict the response of patients to anticancer treatments. Patients This study was approved by the Chang Gung Medical Foundation Institutional Review Board-approval: 100-2161B.Patients with NSCLC-related malignant pleural effusion were enrolled in this study after they signed the inform consent.The malignant pleural effusion was collected for the generation of spheroids.The characteristics of enrolled patients, including gender, age, the histology of cancer, and their response to treatment were recorded. Generation of Multicellular Spheroids (Colony Formation Assay) NSCLC-related MPE was placed in 50 mL centrifuge tubes for centrifugation.The supernatant of the MPE was sterile filtered and used as a component of the primary culture medium (70% DMEM +30% v/v sterilely filtered MPE-fluid + Penicillin-G/Streptomycin 1000 U/mL) [13].The cell pellets collected from NSCLC-related MPE were resuspended, and mononuclear cells were isolated by Ficoll-Hypaque (Histopaque-1077, Sigma-Aldrich, St. Louis, MO, USA) density gradient centrifugation (200× g, 20 min at room temperature). The primary culture medium was mixed with 0.8% Sea Plaque Low Melt Agarose (Lonza Cat # 50101, Lonza, Basel, Switzerland) and microwaved until the agarose completely dissolved.After then, 1.5 mL of this agarose mixture was added into the 6-well plate and solidified at room temperature to form the base agarose layer.To prepare the mixture for the upper agarose layer, the primary culture medium was mixed with 0.48% Sea Plaque Low Melt Agarose and microwaved until the agarose was completely dissolved.This upper agarose layer mixture was then placed in a water bath to keep the temperature at around 42 degree Celsius to avoid premature solidification.The mononuclear cell isolated by Ficoll-Hypaque density gradient centrifugation of the NSCLC-related MPE was resuspended in the upper agarose layer mixture and added into the 6-well plate (1.5 mL per well) pre-coated with the base agarose layer.After the upper agarose layer was solidified at room temperature, the culture well was filled with primary culture medium and placed into the incubator (37 degree Celsius, 5% CO 2 ).Incubation time for the formation of spheroids depended on the growth rate of the cells and was evaluated by direct observation under an inverted microscope. Immunohistochemistry Staining of TTF-1 After formation, the multicellular spheroids were fixed with formalin and embedded in paraffin.Sections of the specimens with a thickness of 5 µm were deparaffinized, and then the epitopes were retrieved by heating the deparaffinized tissue slides for 20 min in Envision Flex Target Retrieval Tris-EDTA pH 9.0 (Dako, Carpinteria, CA, USA).After then, the slides were cooled for 20 min and incubated with the thyroid transcription factor-1 (TTF-1) antibody (1:200; product code: NCL-L-TTF-1; Leica Biosystems, Newcastle Upon Tyne, UK).The Envision Flex Link HRP-DAB (Dako, Carpinteria, CA, USA) was used as a detection system following the manufacturer's instructions. Modified MTT Assay The MTT (M5655, Sigma-Aldrich, St. Louis, MO, USA) was diluted in DMEM without phenol red to a final concentration of 0.5 mg/mL.After treated with gefitinib for 7 days, 40 µL of the MTT solution was added into each well of the 6-well culture plates and then placed in the incubator (37 degree Celsius, 5% CO 2 ) for 2 h.The viability of the cells in spheroids was evaluated by direct visualization of the formazan product within the spheroids. Immunofluorescent Staining of Cleaved Caspase 3 After treatment, multicellular spheroids were fixed in 10% buffered formalin phosphate and then embedded in paraffin.Sections with a thickness of 8 µm of the paraffinembedded specimens were mounted on glass slides, deparaffinized, and then rehydrated.The antigen of the specimens was retrieved by boiling in a pressure cooker for 5 min with sodium citrate solution (pH 6.0) with 0.1% Tween 20.After cooling for 20 min at room temperature, the sections were blocked with 5% bovine serum albumin for 30 min and then incubated with rabbit polyclonal antibody to cleaved caspase 3 (1:100; AB3623, Chemicon, Temecula, CA, USA) at 4 degree Celsius overnight.The slides were washed and then incubated with goat anti-rabbit IgG conjugated with Alexa Fluor 546 (1:100; A11010, Molecular Probes, Invitrogen, Carlsbad, CA, USA) at room temperature for 1 h.The slides were then mounted with DAPI-containing antifade reagent (S36939, Invitrogen, Carlsbad, CA, USA) and photographed under the fluorescence microscope. Generation of Multicellular Spheroids by NSCLC Cell Lines and Evaluation of Their Response to Gefitinib After being cultured in the soft agar for 28 days, NSCLC cell lines A549, H1299, and PC9 formed multicellular spheroids with a size of around 300 µm~500 µm (Figure 1).The spheroids were kept in the agarose and were treated with gefitinib with concentrations of 0, 0.5, 5, and 25 µM, respectively, for 7 days. The slides were then mounted with DAPI-containing antifade reagent (S36939, Invitrogen, Carlsbad, CA, USA) and photographed under the fluorescence microscope. Generation of Multicellular Spheroids by NSCLC Cell Lines and Evaluation of Their Response to Gefitinib After being cultured in the soft agar for 28 days, NSCLC cell lines A549, H1299, and PC9 formed multicellular spheroids with a size of around 300 µm~500 µm (Figure 1).The spheroids were kept in the agarose and were treated with gefitinib with concentrations of 0, 0.5, 5, and 25 µM, respectively, for 7 days.After gefitinib treatment, the spheroids in the agarose were fixed with formalin and embedded in paraffin.Immunofluorescence staining for cleaved caspase 3 was used to evaluate the apoptotic response of the NSCLC cell line spheroids to gefitinib treatment.There was no significant increase in the cleaved caspase 3 staining in the gefitinib-resistant A549 and H1299 spheroids, indicating limited apoptotic response to gefitinib treatment (Figure 2).On the other hand, although there was only a slight increase in the cleaved caspase 3 staining in the gefitinib-sensitive PC9 spheroids, DAPI-staining showed a complete loss of the cellular nucleus after gefitinib treatment.These findings suggested that the nuclear structure of the cells in the PC9 spheroids were disrupted after gefitinib treatment, indicating a significant response to gefitinib treatment.After gefitinib treatment, the spheroids in the agarose were fixed with formalin and embedded in paraffin.Immunofluorescence staining for cleaved caspase 3 was used to evaluate the apoptotic response of the NSCLC cell line spheroids to gefitinib treatment.There was no significant increase in the cleaved caspase 3 staining in the gefitinib-resistant A549 and H1299 spheroids, indicating limited apoptotic response to gefitinib treatment (Figure 2).On the other hand, although there was only a slight increase in the cleaved caspase 3 staining in the gefitinib-sensitive PC9 spheroids, DAPI-staining showed a complete loss of the cellular nucleus after gefitinib treatment.These findings suggested that the nuclear structure of the cells in the PC9 spheroids were disrupted after gefitinib treatment, indicating a significant response to gefitinib treatment. In addition to cleaved caspase 3 immunofluorescence staining, a modified MTT assay was also used to evaluate the viability of the spheroids after gefitinib treatment.The spheroids were first photographed after being treated with gefitinib for 7 days.After then, 40 µL of MTT solution (0.5 mg/mL) was added to each well of the culture plate and placed in an incubator for 2 h.The formation of the purple formazan product was then photographed.After being treated with gefitinib for 7 days, there was significant formation of formazan product in the spheroids generated by gefitinib-resistant A549 and H1299, while the formation of formazan was significantly reduced in the spheroids generated by gefitinib-sensitive PC9 cells (Figure 3).The difference in the formation of formazan product in the spheroids generated by gefitinib-resistant or gefitinib-sensitive cell lines is consistent with the result of cleaved caspase 3 immunofluorescence staining.This finding suggested that direct visualization of the formation of formazan product in the spheroids might be used to evaluate the viability of spheroids after anti-cancer treatment.In addition to cleaved caspase 3 immunofluorescence staining, a modified MTT assay was also used to evaluate the viability of the spheroids after gefitinib treatment.The spheroids were first photographed after being treated with gefitinib for 7 days.After then, 40 µL of MTT solution (0.5 mg/mL) was added to each well of the culture plate and placed in an incubator for 2 h.The formation of the purple formazan product was then photographed.After being treated with gefitinib for 7 days, there was significant formation of formazan product in the spheroids generated by gefitinib-resistant A549 and H1299, while the formation of formazan was significantly reduced in the spheroids generated by gefitinib-sensitive PC9 cells (Figure 3).The difference in the formation of formazan product in the spheroids generated by gefitinib-resistant or gefitinib-sensitive cell lines is consistent with the result of cleaved caspase 3 immunofluorescence staining.This finding suggested that direct visualization of the formation of formazan product in the spheroids might be used to evaluate the viability of spheroids after anti-cancer treatment. Patient Characteristics MPE cells in 14 NSCLC patients were collected for the generation of spheroids (Table Patient Characteristics MPE cells in 14 NSCLC patients were collected for the generation of spheroids (Table 1).The gender, age, cell type, and the EGFR mutation status of these 14 patients were recorded.The EGFR mutation status of these enrolled patients were wild type (n = 6), exon 20 insertion (n = 1), L861Q (n = 1), L858R (n = 4), and exon 19 deletion (n = 2).Only one of six patients with EGFR wild type received erlotinib as a therapeutic trial before the result of the EGFR mutation test was available, and this patient did not respond to erlotinib treatment.The other five patients with EGFR wild type and the patient with EGFR exon 20 insertion did not receive EGFR TKIs treatment.All seven of the patients with L861Q mutation, L858R mutation, and exon 19 deletion in this study received EGFR-TKIs, and five of them responded to treatment. Modified MTT Assay of Spheroids General from NSCLC-Related Malignant Pleural Effusion After being cultured in the agarose for 28 days, the cells collected from MPE formed spheroids with a diameter of around 300 µm~500 µm (Figure 4A).Immunohistochemistry staining for the spheroids showed positive nuclear staining of TTF-1 (Figure 4B).The spheroids generated from MPE of patients were treated with gefitinib 25 µM for 7 days.After gefitinib treatment, the MTT solution was added into each well of the culture plate and then placed in the incubator (37 degree Celsius, 5% CO 2 ) for 2 h.The viability of the cells in spheroids treated with or without gefitinib was evaluated by direct visualization of the formazan products in the spheroids. The spheroids generated from MPE of patients with EGFR wild type (case 1~6) and exon 20 insertion mutation (case 7) were treated with or without gefitinib for 7 days and then incubated with MTT solution for 2 h in the incubator.The formation of formazan products can be found in the spheroids treated with or without gefitinib (Figure 5, case 1-7).This finding suggested that gefitinib treatment did not significantly reduce the cell viability of spheroids generated from MPE of patients with EGFR wild type or EGFR-TKI non-sensitive exon 20 insertion.On the other hand, the formation of formazan was significantly reduced after gefitinib treatment in the spheroids generated from MPE of those five patients with sensitive EGFR mutations and responded to EGFR-TKIs (Figure 5, case 8-12).In addition, the formation of formazan was similar after being treated with or without gefitinib in the spheroids generated from MPE of the two patients who had sensitive EGFR mutation but did not respond to EGFR-TKI treatment (Figure 5, case 13-14).These findings suggested that the response of MPE spheroids to gefitinib treatment was consistent with the clinical response of each corresponding patient to the EGFR-TKIs treatment. Modified MTT Assay of Spheroids General from NSCLC-Related Malignant Pleural Effusion After being cultured in the agarose for 28 days, the cells collected from MPE formed spheroids with a diameter of around 300 µm~500 µm (Figure 4A).Immunohistochemistry staining for the spheroids showed positive nuclear staining of TTF-1 (Figure 4B).The spheroids generated from MPE of patients were treated with gefitinib 25 µM for 7 days.After gefitinib treatment, the MTT solution was added into each well of the culture plate and then placed in the incubator (37 degree Celsius, 5% CO2) for 2 h.The viability of the cells in spheroids treated with or without gefitinib was evaluated by direct visualization of the formazan products in the spheroids.The spheroids generated from MPE of patients with EGFR wild type (case 1~6) and exon 20 insertion mutation (case 7) were treated with or without gefitinib for 7 days and then incubated with MTT solution for 2 h in the incubator.The formation of formazan products can be found in the spheroids treated with or without gefitinib (Figure 5, case 1-7).This finding suggested that gefitinib treatment did not significantly reduce the cell viability of spheroids generated from MPE of patients with EGFR wild type or EGFR-TKI non-sensitive exon 20 insertion.On the other hand, the formation of formazan was significantly reduced after gefitinib treatment in the spheroids generated from MPE of those five patients with sensitive EGFR mutations and responded to EGFR-TKIs (Figure 5, case 8-12).In addition, the formation of formazan was similar after being treated with or without gefitinib in the spheroids generated from MPE of the two patients who had sensitive EGFR mutation but did not respond to EGFR-TKI treatment (Figure 5, case [13][14].These findings suggested that the response of MPE spheroids to gefitinib treatment was consistent with the clinical response of each corresponding patient to the EGFR-TKIs treatment.12), the formation of formazan product in the spheroids treated with gefitinib was reduced as compared to the formation of formazan in spheroids without gefitinib treatment.For spheroids generated from MPE of patients with sensitive EGFR mutation but did not respond to EGFR-TKIs (case 13-14), there was formazan product formation in both the spheroids treated with or without gefitinib.The bar in this figure has a constant width of 250 µm. Discussion In this study, we showed that multicellular spheroids can be generated by cells collected from NSCLC-related malignant pleural effusion.In addition, the response of the spheroids to gefitinib treatment was consistent with the clinical response of the original tumors to EGFR-TKIs.These findings suggested that multicellular spheroids generated After being treated with or without 25 µm of gefitinib for 7 days, the MTT solution was added into the well of the culture plates and incubated for 2 h.For spheroids generated from MPE of patients with wild type or non-sensitive EGFR mutation (case 1-7), there was formazan product formation in both the spheroids treated with or without gefitinib.For spheroids generated from MPE of patients with sensitive EGFR mutation and had a clinical response to EGFR-TKIs (case 8-12), the formation of formazan product in the spheroids treated with gefitinib was reduced as compared to the formation of formazan in spheroids without gefitinib treatment.For spheroids generated from MPE of patients with sensitive EGFR mutation but did not respond to EGFR-TKIs (case 13-14), there was formazan product formation in both the spheroids treated with or without gefitinib.The bar in this figure has a constant width of 250 µm. Discussion In this study, we showed that multicellular spheroids can be generated by cells collected from NSCLC-related malignant pleural effusion.In addition, the response of the spheroids to gefitinib treatment was consistent with the clinical response of the original tumors to EGFR-TKIs.These findings suggested that multicellular spheroids generated from NSCLC-related MPE might be used as a tool to predict the response of the original tumors to anti-cancer treatment. The utilization of targeted therapies or immunotherapies has significantly increased because the response of NSCLC to these treatments can be predicted by testing the presence of a sensitive mutation or the expression of specific proteins in the cancers [8,[16][17][18].EGFR-TKIs was the most widely used targeted therapy for the treatment of NSCLC [16].The response of NSCLC harboring L858R mutation or exon 19 deletion to EGFR-TKIs gefitinib, erlotinib, afatinib, and osimertinib were around 70~80% in multiple clinical trials [7,19,20].In addition, osimertinib was also shown to be effective for around 70% of advanced-stage NSCLC patients with the T790M mutation [21].While the EGFR mutation status can be used to predict the response of advanced-stage lung adenocarcinoma to EGFR-TKIs, the accuracy of this prediction was not satisfying.Another commonly used treatable target for NSCLC was the fusion between echinoderm microtubule associated protein-like 4 and anaplastic lymphoma kinase (EML4-ALK) [22].The response rate of ALK inhibitors in patients with EML4-ALK fusion gene was around 73~83% [23][24][25][26][27].Despite the promising clinical benefit of these targeted therapies, there were still a substantial proportion of NSCLC patients harboring sensitive mutation but did not respond well to these treatments. Immunotherapy has become one of the major treatment options for patients with advanced-stage NSCLC in recent years.High expression of PD-L1 on tumor or higher tumor mutational burden (TMB) are commonly used as biomarkers to predict the response of cancer to anti-PD-L1 or anti-PD-1 therapies [28], although anti-PD-L1 or anti-PD-1 treatment were shown to provide clinical benefit in cancer patients with high expression of PD-L1.However, intra-tumoral heterogeneity and inter-tumoral heterogeneity of PD-L1 expression, the inter-observer variability in scoring PD-L1 staining, and reproducibility has reduced its accuracy as a biomarker to predict the response of NSCLC to anti-PD-L1 or anti-PD-1 therapies [29].In addition, TMB was found to have poor performance as a predictive and prognostic biomarker [30].These findings suggested that current biomarkers that were used to predict the response of NSCLC immunotherapies might not be as satisfying as expected. Multicellular spheroids retain the structure and some of the characteristics that mimic the actual tumor in patients, and thus might be a better model for the investigation of solid cancer treatment [12].A patient-derived lung tumor xenograft model was used to predict the response of NSCLC to anticancer treatment [15].However, this xenograft model involved an animal experiment, and may not be easily adopted in clinical practice.An ex vivo model of treatment response evaluation for NSCLC to chemotherapy and immunotherapy has been reported by using tumor fragment spheroids generated from surgical sample [31].While this tumor fragment spheroid model can be used to predict the response of NSCLC to anticancer treatments, surgical samples were not available in a significant proportion of NSCLC patients, which limited the utilization of this model in clinical practice. Spheroids generated from NSCLC-related MPE have been found to retain the characteristic of the original tumor and can be used as an ex vivo model for drug screening with chemotherapy, targeted therapy, and immunotherapy [32,33].These findings suggested that the clinical response of the actual cancer to treatment might be predicted by evaluating the response of spheroids generated from MPE to the same treatment.The presence of non-malignant cells in the MPE might interfere with the result of the laboratory test to evaluate the response of cancer cells to treatment when these cells were grown as a monolayer.One of the differences between non-malignant and malignant cells is the lack of contact inhibition in malignant cells [34].The colony formation assay is a widely used method to illustrate the loss of contact inhibition of malignant cells [35].In this study, we showed that the multicellular spheroids can be generated by culturing the cells collected from NSCLC-related MPE in soft agarose because of the loss of contact inhibition of the cancer cells. One of the obstacles of using spheroids generated by cells collected from MPE to predict the response of original cancer to treatment is to evaluate the viability of the cells within the spheroids grown in agarose because it does not allow for cell retrieval upon completion of treatment [35].Flow cytometric detection of annexin V-FITC/Propidium iodide-stained apoptotic cells was not applicable in this MPE spheroid model because the spheroids were grown in the soft agarose and were not able to be disaggregated for experiments easily.MTT assay and XTT assay were both been used to evaluate the cell viability in cancer studies [36,37].However, the presence of non-malignant cells in the NSCLC-related MPE might interfere with the result of both of the two assays and reduce their accuracy for the evaluation of cell viability.In this study, we added the MTT solution into the culture wells after NSCLC-related MPE spheroids were treated with or without gefitinib and evaluated the viability of spheroids by direct visualization of formazan product.Immunofluorescence staining of cleaved caspase 3 was also used to detect apoptotic cells in A549, H1299, and PC9 spheroids after gefitinib treatment (Figure 2).There was no significant apoptosis in gefitinib-resistant A549 and H1299 spheroids treated with or without gefitinib, while the cellular nucleus of the gefitinib-sensitive PC9 spheroids was completely disrupted after gefitinib treatment.These findings indicated that gefitinib treatment significantly reduced the viability of cells in PC9 spheroids but not in A549 spheroids and H1299 spheroids, which were consistent with the viability of spheroids evaluated by direct visualization of the formation of formazan product after adding MTT solution into the culture well.We then evaluated the viability of the spheroid generated from NSCLC-related MPE after gefitinib treatment by using this modified MTT assay as it can be more easily adopted in clinical practice than immunofluorescence staining of cleaved caspase 3. There were limitations in this study.First, this approach is only applicable when the spheroids can be generated from NSCLC-related MPE and thus may not be feasible if no spheroids are formed.Second, the incubation time for the formation of spheroids was 28 days in this study.This lengthy culture time might reduce its efficacy as a tool to predict treatment response.The incubation time for the formation of spheroids depended on the growth rate of the cells.In Figure 1, H1299 and PC9 formed multicellular spheroids with a diameter of around 250 µm at day 7. On the other hand, A549 formed multicellular spheroids in agarose layer slower than H1299 and PC9.It is possible to shorten the incubation interval for the generation of multicellular spheroids from NSCLC-related MPE as spheroids with a diameter of around 200 µm can be found after being incubated for 7 days (Figure 4).The aim of this study is to prove the concept to use multicellular spheroids generated from NSCLC-related MPE as a tool to predict the response of cancer to treatment.We grew the cells in the agarose layer for 28 days to form spheroids of a size that were easier for direct visualization of the formation of formazan product after adding the MTT solution into the culture plate, as well as to obtain sections of the spheroids for TTF-1 immunohistochemistry staining and cleaved caspase-3 immunofluorescence staining.Although treating smaller spheroids that were grown in a shorter incubation interval might be possible, it needs to be confirmed by further investigation.Third, the formation of formazan products in the spheroids can not to be quantified through direct visualization, and thus might not be able to detect a small difference in the viability of the cells in spheroids after treatment.This defect limited its efficacy to predict the response of MPE spheroids to treatments with less potent anticancer effects.Methods to measure the viability of spheroids have been introduced [38].Further experiments are needed to investigate the optimal quantitative assay for the multicellular model that we developed in this study.Fourth, the dose and duration of each anti-cancer treatment will need to be individually optimized.In this study, we treated the spheroids generated from NSCLC-related MPE with 25 µM of gefitinib for 7 days in contrast to the 0.5 µM of gefitinib for 3 days that were used to treat patient-derived xenograft models in a previous study [15].After being treated with or without gefitinib, there was no significant difference in the formation of formazan product in A549 and H1299 spheroids.These findings indicated that there was no significant difference in the viability of A549 spheroids and H1299 spheroid after treated with or without gefitinib.On the other hand, the formation of formazan product was absent in PC9 spheroids treated with 0.5 µM, 5 µM, and 25 µM of gefitinib for 7 days (Figure 3).This absence of formazan product after gefitinib treatment indicated that the viability of cells in the PC9 spheroids was significantly reduced.To significantly reduce the viability of spheroids generated from gefitinib-sensitive cells, while not reducing the viability of spheroids generated from gefitinib-resistant cells, we treated the spheroids generated from NSCLC-related MPE with 25 µM of gefitinib for 7 days.Using this approach of high concentration and lengthy duration of gefitinib treatment, we can detect the difference in the viability between gefitinib-sensitive spheroids and gefitinibresistant spheroids by visualization of formazan product after adding MTT solution into the culture well.However, this approach of high gefitinib concentration with lengthy treatment duration might not be suitable for other anti-cancer treatments such as chemotherapies, immunotherapies, or radiotherapy.Further study will be needed to determine the optimal dose and duration for other anti-cancer treatments in this NSCLC-related MPE spheroid model to predict the response of NSCLC to treatment. Conclusions In this study, we showed that multicellular spheroids can be generated by culturing the cells collected from NSCLC-related malignant pleural effusion in the soft agarose.We also showed that cell viability evaluated by direct visualization of the formation of formazan product of the MTT assay in the spheroids after treatment were consistent with the clinical response of the original tumor to treatment.These results suggested that spheroids generated from NSCLC-related MPE can be used as tools to predict the response of NSCLC to treatment. Figure 1 . Figure 1.Generation of multicellular spheroids by NSCLC cell lines A549, H1299, and PC9.After being cultured in the soft agarose for 28 days, NSCLC cell lines A549, H1299, and PC9 formed multicellular spheroids.The average size of the spheroids was around 300 µm~500 µm.The bar in this figure has a constant width of 250 µm. Figure 1 . Figure 1.Generation of multicellular spheroids by NSCLC cell lines A549, H1299, and PC9.After being cultured in the soft agarose for 28 days, NSCLC cell lines A549, H1299, and PC9 formed multicellular spheroids.The average size of the spheroids was around 300 µm~500 µm.The bar in this figure has a constant width of 250 µm. Figure 2 . Figure 2. Cleaved caspase 3 immunofluorescence staining of NSCLC spheroids after gefitinib treatment.After being treated with gefitinib for 7 days, there was no significant increase in the cleaved caspase 3 in the A549 (upper panel) and H1299 (middle panel) spheroids as compared to the spheroids without gefitinib treatment.DAPI nuclear staining showed an intact cellular nucleus of the A549 and H1299 spheroids treated with or without gefitinib.There was also little cleaved caspase 3 staining in the PC 9 spheroids (lower panel) treated with or without gefitinib.DAPI nuclear staining showed that the cellular nucleus was completely disrupted in the PC9 spheroid treated with 0.5, 5, and 25 µM of gefitinib for 7 days, indicating that gefitinib treatment significantly reduced the viability of the PC9 cells in the spheroids. Figure 2 . Figure 2. Cleaved caspase 3 immunofluorescence staining of NSCLC spheroids after gefitinib treatment.After being treated with gefitinib for 7 days, there was no significant increase in the cleaved caspase 3 in the A549 (upper panel) and H1299 (middle panel) spheroids as compared to the spheroids without gefitinib treatment.DAPI nuclear staining showed an intact cellular nucleus of the A549 and H1299 spheroids treated with or without gefitinib.There was also little cleaved caspase 3 staining in the PC 9 spheroids (lower panel) treated with or without gefitinib.DAPI nuclear staining showed that the cellular nucleus was completely disrupted in the PC9 spheroid treated with 0.5, 5, and 25 µM of gefitinib for 7 days, indicating that gefitinib treatment significantly reduced the viability of the PC9 cells in the spheroids.Diagnostics 2024, 14, x FOR PEER REVIEW 6 of 13 Figure 3 . Figure 3. Viability of A549, H1299 and PC9 spheroids after gefitinib treatment.After being treated with 0, 0.5, 5, and 25 µM of gefitinib for 7 days, 40 µL of MTT solution (0.5 mg/mL) was added to each well of the culture plates and they were incubated for 2 h.The formation of purple formazan product can be found in the A549 (upper panel) and H1299 (middle panel) spheroids treated with or without gefitinib.On the other hand, the formation of formazan product was significantly reduced in the PC9 spheroids that were treated with 0.5, 5, and 25 µM of gefitinib as compared to the formation of formazan in the PC9 spheroids without gefitinib treatment.The reduced formation of formazan in the PC9 spheroids suggested a significant reduction in the viability of cells after gefitinib treatment.The bar in this figure has a constant width of 200 µm. Figure 3 . Figure 3. Viability of A549, H1299 and PC9 spheroids after gefitinib treatment.After being treated with 0, 0.5, 5, and 25 µM of gefitinib for 7 days, 40 µL of MTT solution (0.5 mg/mL) was added to each well of the culture plates and they were incubated for 2 h.The formation of purple formazan product can be found in the A549 (upper panel) and H1299 (middle panel) spheroids treated with or Figure 4 . Figure 4. (A) Generation of spheroids by cells collected from the NSCLC-related malignant pleural effusion (MPE) in the soft agarose.After being cultured in the soft agarose for 28 days, the cells collected from NSCLC-related MPE formed spheroids with a diameter of around 300 µm~500 µm.(B) TTF-1 immunohistochemistry staining for the spheroids generated from NSCLC-related MPE.After formation, the spheroids were removed from the well and fixed by 10% formalin and then embedded in paraffin.TTF-1 immunohistochemistry staining showed positive nuclear staining in the spheroid. Figure 4 . Figure 4. (A) Generation of spheroids by cells collected from the NSCLC-related malignant pleural effusion (MPE) in the soft agarose.After being cultured in the soft agarose for 28 days, the cells collected from NSCLC-related MPE formed spheroids with a diameter of around 300 µm~500 µm.(B) TTF-1 immunohistochemistry staining for the spheroids generated from NSCLC-related MPE.After formation, the spheroids were removed from the well and fixed by 10% formalin and then embedded in paraffin.TTF-1 immunohistochemistry staining showed positive nuclear staining in the spheroid.Diagnostics 2024, 14, x FOR PEER REVIEW 8 of 13 Figure 5 . Figure 5.After being treated with or without 25 µm of gefitinib for 7 days, the MTT solution was added into the well of the culture plates and incubated for 2 h.For spheroids generated from MPE of patients with wild type or non-sensitive EGFR mutation (case 1-7), there was formazan product formation in both the spheroids treated with or without gefitinib.For spheroids generated from MPE of patients with sensitive EGFR mutation and had a clinical response to EGFR-TKIs (case 8-12), the formation of formazan product in the spheroids treated with gefitinib was reduced as compared to the formation of formazan in spheroids without gefitinib treatment.For spheroids generated from MPE of patients with sensitive EGFR mutation but did not respond to EGFR-TKIs (case 13-14), there was formazan product formation in both the spheroids treated with or without gefitinib.The bar in this figure has a constant width of 250 µm. Figure 5 . Figure 5.After being treated with or without 25 µm of gefitinib for 7 days, the MTT solution was added into the well of the culture plates and incubated for 2 h.For spheroids generated from MPE of patients with wild type or non-sensitive EGFR mutation (case 1-7), there was formazan product formation in both the spheroids treated with or without gefitinib.For spheroids generated from MPE of patients with sensitive EGFR mutation and had a clinical response to EGFR-TKIs (case 8-12), the formation of formazan product in the spheroids treated with gefitinib was reduced as compared to the formation of formazan in spheroids without gefitinib treatment.For spheroids generated from MPE of patients with sensitive EGFR mutation but did not respond to EGFR-TKIs (case 13-14), there was formazan product formation in both the spheroids treated with or without gefitinib.The bar in this figure has a constant width of 250 µm.	2024-05-19T15:59:23.762Z	2024-05-01T00:00:00.000	{ "year": 2024, "sha1": "3099388814414f1e837f7e120ff3c9d3f4ad0859", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/2075-4418/14/10/998/pdf?version=1715418639", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "55fcea8d72460ca3b485f824d3b6528bbe578722", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [] }
234590349	pes2o/s2orc	v3-fos-license	Knowledge About Dengue Fever Prevention Among People Visiting Benazir Bhutto Hospital Background: The higher incidence of dengue fever in Pakistan demands additional efforts in order to limit the disease. Despite active public health campaigns, low public awareness is one of the factors facilitating dengue virus transmission. For effective preventive measures, the assessment of the knowledge gap and then taking appropriate steps to fill the gap is required. The objective of this study is to assess knowledge about dengue fever prevention among people visiting Benazir Bhutto Hospital, Rawalpindi. Material and Methods: The descriptive cross-sectional study of 6 months duration was conducted with 280 participants selected via nonprobability convenience sampling. After informed consent, an interview was conducted based on a questionnaire that assessed socio-demographic parameters and knowledge about dengue virus transmission and prevention. Data were analyzed through SPSS v.22. The study was approved by the Ethical Review Board (ERB) of Rawalpindi Medical University and Allied hospitals. Results: Out of 280 respondents, 54.6% were males and 45.4% females and the mean age was 35.0 ± 13.1 years. The respondents having high knowledge scores were 66(23.6%) while those having moderate and low scores were 159 (56.8%) and 55 (19.6%) respectively. Educated respondents (p=0.03) and urban residents (p=0.05) had higher knowledge scores. Conclusion: Majority of the participants know about dengue fever. However, only one out of every four respondents have good knowledge scores for dengue fever prevention. Introduction Dengue fever is a mosquito borne tropical disease caused by dengue virus. WHO reports that globally there are 1.65 million cases annually and it is endemic in 112 countries. 1,2 Dengue hemorrhagic fever and dengue shock syndrome have a mortality of 26%. 3 Rainy season and humidity support dengue-vector breeding. 4 Factors facilitating dengue virus transmission include lack of proper preventive measures and lack of public awareness about the disease. 5 Since the 1970s, dengue epidemics have been affecting American population. 6 In South India, 86% of the participants had heard of dengue virus while 68% of respondents thought drains and garbage as breeding places of dengue vectors. Also, 25% of the participants were aware of stagnant water as a breeding habitat. 7 A similar study in Pakistan in 2014 reported, 58.6% of participants being aware of Aedes egypti mosquito as vector of dengue virus; 54.8% had knowledge about stagnant water as its breeding site; 47.6% and 44.7% were of the view that Aedes egypti bites during dawn and dusk respectively; 51%, 41.3% and 44.7% were knowledgeable of bleeding, headache and vomiting associated with dengue fever, respectively. 8 Another study from Karachi showed that only 35% people had adequate knowledge of dengue fever. 9 The most significant barriers to dengue virus control in Pakistan included high population density with slums having presence of mosquito breeding sites, and a passive role of government regarding community awareness regarding dengue fever. 10 Studies report that the only predictor of high practice score was obtaining a high knowledge score. 11 Dengue is a potentially preventable disease and the literature indicates that lack of knowledge about prevention is the key factor in disease susceptibility. Lack of adequate knowledge about dengue fever prevention is one of the main reasons for its prevalence in Pakistan. Despite the public awareness programs, there are outbreaks costing the lives of many every year. The objective of the study is to find the level of knowledge about dengue fever prevention in order to manifest any knowledge gap and ultimately to emphasize more on the public awareness efforts. Materials and Methods This descriptive cross-sectional study of 6 months duration was conducted at Benazir Bhutto Hospital, Rawalpindi. Using nonprobability convenience sampling, 280 participants of age 15 years or above were included in the study. Participants below the age of 15 years, those having any previous exposure to knowledge about dengue fever (e.g. those having family members admitted due to dengue fever in the department of infectious disease) and medical personnel including doctors, nurses and medical students were excluded from the study. Informed consent was taken via consent form. The study was approved by the Ethical Review Board of Institutional Research Forum, Rawalpindi Medical University. The participants were interviewed based on a referenced questionnaire that assessed the knowledge about source, vector, spread and prevention of dengue virus. Education status, demographic details and social parameters were also included in the questionnaire. The knowledge scores were categorized as low (up to 40%), moderate (from 41% to 70%) and high scores (above 70%) based on the number of questions correctly answered. Data were analyzed using SPSS version 22. Knowledge scores were correlated with different demographic parameters and a p-value of less than 0.05 was considered to be statistically significant. Results The study included 280 respondents out of which 153 (54.6%) were male while 127 (45.4%) were females. The mean age was 35.0 ± 13.1 years with 180 (64.3%) participants having age 15 to 39 years and 100 (35.7%) participants having age 40 years or above. Those who lived in urban areas were 186 in number (66.4%) while those living in rural areas were 94 in number (33.6%). Among the participants, 55 (20.0%) were uneducated, 159 (38.9%) had primary/secondary education and 66 (41.1%) had had higher secondary/higher education. The results of the study showed that 23.6% had high scores of knowledge while 56.8% and 19.6% had moderate and low scores respectively. Frequencies and percentages of participants with low, moderate and high scores were calculated (Table-1). 93.6% claimed that they know about dengue fever and the source of their knowledge is described in Table-2. The knowledge about prevention and transmission of dengue virus was evaluated by a series of questions, the results of which are shown in Table-3. Percentage of respondents with high scores of knowledge was higher in educated (77.2%) as compared to uneducated (22.7%) respondents. The relation between education status and knowledge level was significant (p=0.03). Urban residency was significantly associated with high knowledge scores (p=0.05). However, age and gender had no significant relation with knowledge. Discussion In this study, the percentage of respondents that had heard about dengue fever is 93.6% which is in accordance with other studies in Pakistan (89.9%-92.3%). 12,13 The percentages are lower in studies conducted in India and Thailand (85% and 67% respectively). 14,15 Majority of people in Pakistan have heard about the disease because of widespread media campaigns for dengue fever awareness. About four out of five respondents were aware that dengue virus is transmitted by mosquitoes, which is in accord with a past study. 14 However, only 56.1% respondents were aware that stagnant water is the breeding site. A similar percentage (40%) has been indicated previously.14 For 78.6% of the respondents, the media was the source of knowledge while in other countries, 43.9% participants had their friends or relatives as the main source of information about dengue fever. 16 About one out of four (23.6%) participants had good scores of knowledge about prevention and transmission of dengue virus which is less than previous studies conducted in Pakistan (35%) and Philippines (61.45%). 12,17 Majority of the respondents believed dengue fever can be prevented by mosquito nets and use of mosquito repellents (84.6% and 76.1%, respectively) which is contrary to the previous research results in Pakistan (49% and 38.6%, respectively) and Laos (31.58% and 59.91%, respectively) . 12,16 As the study employed non-probability sampling and the study setting was a tertiary care hospital, the results cannot be generalized to the population not in contact with medical care. Conclusion Majority of the participants have heard about dengue fever. However, only one-fourth of individuals have good knowledge for dengue fever prevention and transmission. Hence, knowledge about dengue fever prevention and transmission is generally inadequate.	2021-02-02T22:36:12.906Z	2020-12-12T00:00:00.000	{ "year": 2020, "sha1": "9dcd28ce8d3b61fb3da57cdb00c8ab0eec5feb47", "oa_license": "CCBYSA", "oa_url": "https://doi.org/10.37939/jrmc.v24i1.1536", "oa_status": "GOLD", "pdf_src": "MergedPDFExtraction", "pdf_hash": "9dcd28ce8d3b61fb3da57cdb00c8ab0eec5feb47", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
270188478	pes2o/s2orc	v3-fos-license	Kinetic Behavior of Glutathione Transferases: Understanding Cellular Protection from Reactive Intermediates Glutathione transferases (GSTs) are the primary catalysts protecting from reactive electrophile attack. In this review, the quantitative levels and distribution of glutathione transferases in relation to physiological function are discussed. The catalytic properties (random sequential) tell us that these enzymes have evolved to intercept reactive intermediates. High concentrations of enzymes (up to several hundred micromolar) ensure efficient protection. Individual enzyme molecules, however, turn over only rarely (estimated as low as once daily). The protection of intracellular protein and DNA targets is linearly proportional to enzyme levels. Any lowering of enzyme concentration, or inhibition, would thus result in diminished protection. It is well established that GSTs also function as binding proteins, potentially resulting in enzyme inhibition. Here the relevance of ligand inhibition and catalytic mechanisms, such as negative co-operativity, is discussed. There is a lack of knowledge pertaining to relevant ligand levels in vivo, be they exogenous or endogenous (e.g., bile acids and bilirubin). The stoichiometry of active sites in GSTs is well established, cytosolic enzyme dimers have two sites. It is puzzling that a third of the site’s reactivity is observed in trimeric microsomal glutathione transferases (MGSTs). From a physiological point of view, such sub-stoichiometric behavior would appear to be wasteful. Over the years, a substantial amount of detailed knowledge on the structure, distribution, and mechanism of purified GSTs has been gathered. We still lack knowledge on exact cell type distribution and levels in vivo however, especially in relation to ligand levels, which need to be determined. Such knowledge must be gathered in order to allow mathematical modeling to be employed in the future, to generate a holistic understanding of reactive intermediate protection. Introduction As one of the enzyme families in Phase 2 metabolism, glutathione transferases constitute the principal cellular protection from reactive lipophilic electrophiles [1,2].Exposure to reactive compounds can occur either directly (dietary components, environmental contaminants) or as a result of Phase 1 biotransformation of medical drugs and other xenobiotics.Several endogenous reactive molecules are also continuously formed including products from lipid peroxidation and catecholamines [3,4].In addition to their catalytic function, several GSTs have also been characterized as intracellular binding/transport proteins [5].Humans have seventeen soluble glutathione transferases and three that are membranebound [1].The enzymes are widely distributed in the body, and also in different subcellular compartments [6,7].Remarkably high concentrations of these enzymes testify to their important function. In this review, the kinetic mechanisms of glutathione transferases (GSTs) and their relevance to physiological function will be discussed.As pioneered by Professor Bengt Mannervik's laboratory [8] all GSTs investigated so far display a random sequential kinetic mechanism.The enzymes display incredibly broad specificity [9,10], both to conjugate and detoxify electrophilic lipophilic substrates and in binding lipophilic non-substrate ligands [11]. Glutathione transferases are subject to induction [12] and genetic polymorphism [13] and influence cellular signaling pathways [14].These aspects have been reviewed and together with the glutathione peroxidase function of GSTs [15,16] are outside the scope of the present article. Ligands and substrates can be of endogenous origin (e.g., bilirubin and hydroxyalkenals) or exogenous toxic compounds (and metabolites thereof).Several types of kinetic behavior, including negative co-operativity, cold adaptation, and ligand activation have been suggested to enhance the protective function of GSTs The concentration and location of GSTs in vivo, in relation to ligands and substrates, are thus paramount to understanding the function of this versatile detoxication system. Kinetic Mechanism Studies The first two laboratories to achieve purification and kinetic analysis of GSTs in the mid-seventies were led by Bill Jakoby and Bengt Mannervik.Jakoby's group first proposed a complex kinetic mechanism, ordered at high physiological GSH and ping-pong at lower GSH concentrations [17].One year later, studies in Mannervik's laboratory defined a steady-state random sequential mechanism [8,18] whereas the ping-pong mechanism could be ruled out.Over the years many studies on the kinetic mechanism of various GSTs have been performed.Invariably, a random sequential mechanism was found.Different studies favored steady-state [19][20][21][22][23][24][25][26] or rapid equilibrium mechanisms [27][28][29][30][31][32][33][34].The differences can either be attributed to the enzyme form investigated or the substrates used and also, these mechanisms can be difficult to discriminate.A convenient shortcut, to determine the kinetic mechanism of GSTs, involving alternate substrates [35], was used by Richard Armstrong's laboratory [36] and also for microsomal glutathione transferase 1 (MGST1) [37].Since there is an unlimited plethora of electrophilic substrates and several glutathione analogs [36,38,39] this approach can easily be utilized to determine that the mechanism is rapid equilibrium random sequential.It has been shown that product release can be rate-limiting in a few cases [40,41]. The question becomes, are these kinetic mechanistic characteristics relevant to GST function in vivo? Binding Function Glutathione transferases function as intracellular binding proteins [5].The binding of endogenous ligands such as organic anions in the liver has been unequivocally demonstrated [42].It follows that the binding of non-substrate ligands and the degree of saturation of the enzyme pool could limit detoxification capacity.In the case of MGST1, ligand-induced activation of the enzyme has also been observed [43,44].The potential for GST ligand interactions that affect detoxication capacity is thus of interest and needs to be determined. An important finding was that the dimeric soluble enzymes harbor two active sites [22] and therefore at least two binding sites.The interaction between these sites has been the subject of several investigations and data supporting either independence or co-operativity have been published [45][46][47].Thus, the stoichiometry and/or co-operativity of GST ligand/substrate interactions can potentially influence catalytic capacity.A central question thus becomes, what are the endogenous levels of enzymes and substrates/ligands in vivo? In MGST1 and 2, "a third of the sites reactivity" has been demonstrated [48][49][50].It is puzzling why all potential sites are not utilized. As GSTs are binding proteins it is not surprising that covalent modification by electrophilic reactive intermediates has been observed (e.g., [51,52]).Whether this sacrificial mechanism results in physiological protection is not known.In addition, covalent modification of MGST1 [53] can result in increased activity, raising the issue of whether such modification results in protection from toxicity. Random Sequential Kinetic Mechanism In the first analysis of a GST kinetic mechanism, Jakoby et al. [17] suggested that under physiological conditions (high GSH), the mechanism is ordered sequentially where GSH binds first and ping-pong at low GSH.However, later, a random sequential mechanism was established for all GSTs examined.Nevertheless, an ordered sequential mechanism where GSH binds first will prevail in practice (Figure 1).The very high levels of GSTs in vivo [6,54] and known levels of mercapturic acid excretion [55] indicate that the enzyme levels greatly exceed those of reactive intermediates substrates. Consequently, at physiological GSH levels, GSTs will be fully saturated regardless of GSH binding kinetics (which are quite slow for membrane-bound MGST1 [50]).Full GSH saturation of the GST pool fits well with their role as interception enzymes (that need to be fully catalytically competent to prevent reactive molecules/intermediates from reaching critical cellular targets).Enzyme forms without bound catalytically competent GSH thiolate would be detrimental (i.e., if the mechanisms were ordered for initial electrophile binding). At the prevailing high intracellular GSH concentrations, the deprotonated GSH thiolate fraction is also quite high, approaching those of the enzymes in some tissues.One can estimate that at physiological pH, up to 5 percent of GSH (pKa ≈ 8.7 [56]) is present in the thiolate form, resulting in a concentration of ≈250 µM in the liver.Quite a substantial amount that can undergo a non-enzymatic reaction with electrophiles.However, according to Ketterer et al. [57], the enzyme reaction takes precedence at low GSH concentration, even with very reactive electrophiles such as the acetaminophen reactive intermediate, N-acetyl-p-benzoquinonimine. Different kinetic studies found either rapid equilibrium or steady-state random sequential mechanisms.Given that the enzymes can be highly efficient [2] but turn over quite rarely [58], a rapid equilibrium mechanism formally describes catalysis.In practice, the enzyme will be ordered with respect to GSH as discussed below.Also, because catalytic events are rare in individual enzymes in vivo, product release cannot be rate limiting (i.e., product release can be rate limiting in the catalytic cycle of individual enzymes, but not for GST cellular protection). Glutathione levels and dynamics are quite robust [59][60][61].However, at extreme doses of some toxicants, (e.g., paracetamol) GSH can be depleted in the liver [62].Each liver GST (assuming 0.2 mM GST) has 25 molecules of GSH (5 mM GSH) at its disposal.The turnover number at saturating conditions can be quite high, theoretically emptying the liver of GSH in seconds.Glutathione transferases could, in theory, contribute to toxicity by the ensuing rapid GSH loss.However, there is no evidence for GSTs contributing to toxicity in this way other than after heterologous expression in E. coli [63].This type of toxicity is quite rare, and the high GST and GSH levels that have evolved underscore their vital function in reactive intermediate interception.In a rare case, GST gene duplication has been noted in a Saudi population [64].By way of pure speculation, during a severe famine, a higher level of GST could allow the consumption of otherwise toxic plants (resulting from either GST catalytic or binding function).Glutathione transferases are interception catalysts, and the total concentration of catalytically competent enzymes precisely determines protection capacity.GSTs also function as intracellular binding proteins (reviewed in [5]).It is therefore important to determine whether binding significantly lowers the enzyme concentration available for reactive intermediate protection.GST protein levels are in the 200 µM range in the liver [54], higher in the testes, and generally 2-to 30-fold lower in other organs [6].So, the question becomes, are ligand levels comparable to enzyme levels in some tissues?It has been stated that binding is normally far from saturated [67].For example, bilirubin, and its glucuronic acid conjugated forms, are present at a low micromolar level in the liver [68] as compared to GST, ≈200 µM.However, bile acids can be around 60 nmol/g in the human liver [69] corresponding to more than ≈60 µM intracellular, potentially lowering detoxication capacity.These compounds of course also partition into membranes so the concentration available to inhibit GSTs is not known.Common bile acids like chenodeoxycholic acid and cholic acid have logP values of 1-3 [70].Although fully cell-permeable, this partitioning predicts low cytosolic concentrations that would not cause GST inhibition. In toxicokinetic [62] and cell models [71], total GST levels could account for GSH conjugate formation, indicating that the binding of ligands does not impede catalytic function.There might of course be cell types and tissues and pathological conditions where endogenous/xenobiotic ligand concentrations approach those of GSTs.Clearly, the physiological level of ligands bound to GSTs in the liver and other tissues needs to be determined experimentally. Co-Operativity At ligand concentrations that approach those of GSTs, negative co-operativity has been observed [72].It has been suggested that in this case, negative co-operativity can preserve a significant fraction of the catalytic function of glutathione transferase activity.When incubating human placenta homogenate with a dinitrosyl-diglutathionyl-iron complex, negative co-operativity was observed at near-physiological conditions (2 µM GSTP), preserving half of the catalytic activity at a similar high ligand concentration, [72].In the same paper, negative co-operativity was also observed using rat liver homogenate.These experiments illustrate that negative co-operativity can indeed occur.Whether this binding mechanism has physiological significance in vivo, needs to be put into perspective of the physiological dinitrosyl-diglutathionyl-iron complex concentration.When studying the binding of the dinitrosyl-diglutathionyl-iron complex to purified glutathione transferases, negative co-operativity was also observed [45].Very tight binding of the dinitrosyl-diglutathionyl-iron complex to one subunit of GSTP1-1 (Kd = 1.5 nM) induced a more than 100- Glutathione transferases are interception catalysts, and the total concentration of catalytically competent enzymes precisely determines protection capacity.GSTs also function as intracellular binding proteins (reviewed in [5]).It is therefore important to determine whether binding significantly lowers the enzyme concentration available for reactive intermediate protection.GST protein levels are in the 200 µM range in the liver [54], higher in the testes, and generally 2-to 30-fold lower in other organs [6].So, the question becomes, are ligand levels comparable to enzyme levels in some tissues?It has been stated that binding is normally far from saturated [67].For example, bilirubin, and its glucuronic acid conjugated forms, are present at a low micromolar level in the liver [68] as compared to GST, ≈200 µM.However, bile acids can be around 60 nmol/g in the human liver [69] corresponding to more than ≈60 µM intracellular, potentially lowering detoxication capacity.These compounds of course also partition into membranes so the concentration available to inhibit GSTs is not known.Common bile acids like chenodeoxycholic acid and cholic acid have logP values of 1-3 [70].Although fully cell-permeable, this partitioning predicts low cytosolic concentrations that would not cause GST inhibition. In toxicokinetic [62] and cell models [71], total GST levels could account for GSH conjugate formation, indicating that the binding of ligands does not impede catalytic function.There might of course be cell types and tissues and pathological conditions where endogenous/xenobiotic ligand concentrations approach those of GSTs.Clearly, the physiological level of ligands bound to GSTs in the liver and other tissues needs to be determined experimentally. Co-Operativity At ligand concentrations that approach those of GSTs, negative co-operativity has been observed [72].It has been suggested that in this case, negative co-operativity can preserve a significant fraction of the catalytic function of glutathione transferase activity.When incubating human placenta homogenate with a dinitrosyl-diglutathionyl-iron complex, negative co-operativity was observed at near-physiological conditions (2 µM GSTP), preserving half of the catalytic activity at a similar high ligand concentration, [72].In the same paper, negative co-operativity was also observed using rat liver homogenate.These experiments illustrate that negative co-operativity can indeed occur.Whether this binding mechanism has physiological significance in vivo, needs to be put into perspective of the physiological dinitrosyl-diglutathionyl-iron complex concentration.When studying the binding of the dinitrosyl-diglutathionyl-iron complex to purified glutathione transferases, negative cooperativity was also observed [45].Very tight binding of the dinitrosyl-diglutathionyl-iron complex to one subunit of GSTP1-1 (K d = 1.5 nM) induced a more than 100-fold lower affinity in the second subunit (K d = 120 nM) in theory preserving at least half of the activity at low ligand concentrations.It is difficult to reconcile the experiments with purified enzymes to those with tissue homogenates since the tight binding to both sites observed in the former case should have saturated the enzyme and resulted in total inhibition in tissue homogenates.It is generally recognized that different binding affinities can be found depending on whether pure enzymes or more complex biological fractions (homogenates, whole blood and cells, etc.), are examined.Further studies are thus required to understand the physiological significance of negative co-operativity in the case of GSTs. Temperature-dependent positive and negative catalytic co-operativity has also been observed in human GSTP [73].This behavior would manifest at steady-state conditions but is less likely to occur when the enzyme turns over only occasionally. Partial Sites Reactivity There have been a few reports on sub-stoichiometric behavior in cytosolic GSTs [40,74], though this seems to be the exception.For membrane-bound GSTs, MGST1 and 2, a third of the site's reactivity is well established [49,50] and supported by structural data [75].Since a close relative to the MGSTs, leukotriene C 4 synthase displays all three sites' reactivity [76], such behavior is in principle possible.It is thus counterintuitive that MGSTs do not use all their active sites.The sacrifice of two-thirds of the potential sites must occur for good reason.Perhaps the behavior is an unavoidable consequence of subunit communication observed during GSH binding and thiolate anion stabilization [43].A third of the site's reactivity appears to be compensated for by MGST1 being the most highly concentrated GST in the liver [7]. Covalent Binding Early studies identified certain GSTs as targets for covalent binding [77].This is logical in view of their abundance and capacity for binding hydrophobic reactive intermediates.The covalent binding of reactive electrophiles to GSTs is presumably protective in itself.In addition, a self-preservation mechanism has been suggested where covalent modification of GSTP1-1 in one subunit induces protection of the second subunit [78]. Microsomal GST1 has also been observed to be a target for covalent modification [52].However, generally, modification of cysteine-49 results in activation of the enzyme [79].We have observed that native MGST1 is already partly activated [80].A fraction (≈10%) of the enzyme population can bind and activate GSH to thiolate approximately 30-fold faster than the remaining enzyme [66].Since activation is reversible [81] we proposed that the enzyme population exhibits a dynamic equilibrium between un-activated and activated forms.Upon binding of certain ligands or covalent modification of cysteine-49, the equilibrium is shifted favoring the activated enzyme form. Whether such activation increases the protective capacity of the cell is a conundrum.Activation of the enzyme increases the rate of GSH thiolate formation but not the chemical rate for electrophile conjugation.When measuring the steady-state activity with a reactive substrate such as 1-chloro-2,4-dinitro benzene (CDNB) under saturating conditions the activity is fully determined by GSH thiolate formation (and activation of the enzyme will result in an increased rate).However, at low CDNB concentrations when the chemical conjugation step becomes rate-limiting, more rapid GSH thiolate formation will not increase the enzyme rate.As turnover under physiological conditions [58] seldom occurs, the enzyme will always be fully loaded (at a third of the active sites) with GSH thiolate regardless of whether thiolate formation is slow or activated.Therefore, under normal conditions, the occasional covalent modification and activation of MGST1 will not increase detoxification capacity.At a high rate of formation of reactive electrophilic intermediates, it is conceivable that activated enzymes would contribute more efficiently, however, this could result also in more rapid GSH depletion.Whether activation, which has been observed in toxic conditions in vitro and in vivo [79], is a physiologically significant protective phenomenon therefore remains to be determined. Spatial Aspects In order to protect the cell from reactive intermediates, it makes perfect sense that GSTs must be present at very high concentrations.It also makes sense that the enzymes are widely distributed in the cell, consisting of cytosolic-, mitochondrial-, and membranebound forms [1,82].The soluble enzymes are also present in nuclei [83] and membranebound MGST1 can be found in most membranes reaching 3% and 5% in the rat liver endoplasmic reticulum and mitochondrial outer membrane, respectively [84].It has also been proposed that certain cytosolic GSTs can concentrate outside the nuclear membrane forming a "nuclear shield" [85].Whether the specific distribution of certain enzyme forms is advantageous for cell and DNA protection is a fascinating proposition and remains to be rigorously demonstrated. Substrates for GSTs are invariably hydrophobic and concentrate in membranes.It follows that membrane-bound GSTs could be of particular importance for the membranedissolved substrate fraction.The active site of MGST1 faces the inner phospholipid headgroup region [86,87], where hydrophobic substrates containing an amphipathic part (true for most GST substrates) are thought to accumulate.An investigation examining the conjugation of the very hydrophobic electrophile chlorotrifluoroethylene confirms this notion [88].Based on the stereochemistry of product formation in cells it could be demonstrated that conjugation is predominantly catalyzed by MGST1, even though the GST catalytic capacity of isolated cytosolic and membrane fractions were equal in vitro.In general, cytosolic GSTs cannot access membrane-embedded substrates [89], and act upon the fraction that equilibrates out from the membrane phase (governed by logP [71]).Membranebound MGST1, conversely, cannot access the cytosolic substrate fraction [89].It appears however that some fraction of cytosolic GSTs can adhere to membranes [7].It is not known if these can act directly on membrane-embedded substrates.Clearly, we still have much to learn about the role of membrane partitioning in GST function. Conclusions Glutathione transferases are unique in their capacity to protect from a huge variety of reactive molecules.We have detailed knowledge of GST structure, substrate preferences, and mechanisms.There are several interesting examples of kinetic behavior that could be adaptive and beneficial for organismal protection from reactive intermediates.It is time to put this data in a holistic perspective (Figure 2).For this, we need more detailed knowledge of the cell type and organ distribution of GSTs.In addition, the role of intracellular distribution of GSTs and its impact on protective capacity needs to be clarified. Certain individual GSTs were first described as binding proteins, like a kind of "intracellular albumin".The endogenous ligand levels could have a strong impact on cellular protection, but we simply do not know if this is the case.Ligand levels need to be ascertained. Based on the mercapturic acid levels in urine, it appears that GSTs turn over rather infrequently.This supposition needs to be evaluated quantitatively.The role of GSTs in healthy humans appears to require very high enzyme concentrations.Whereas during acute toxicity, such as paracetamol poisoning, GST catalysis results in GSH loss, which could be detrimental. Having obtained more detailed knowledge of in vivo levels of GSTs and their substrates and ligands, mathematical modeling of cellular, organ, and organismal glutathione conjugation is a logical path to understanding the physiology of reactive intermediate protection.Acknowledgments: I want to acknowledge Bengt Mannervik, who was one of my teachers when I studied Biochemistry at Stockholm University.His lectures inspired interest in enzymology and particularly enzyme kinetics.Later I started as a Ph.D. student in Joe DePierre's research group.There I was lucky to have Bengt's group as neighbors.His group shared knowledge, reagents, and expertise, which was absolutely essential for our research on microsomal glutathione transferase.Over the years I have had the advantage to collaborate with Bengt on many occasions.I see him as a tutor and curiosity-driven scientist with absolute and stringent scientific rigor.But most importantly he has become a close friend.I want to acknowledge my Ph.D. supervisor Joe DePierre who sadly passed this year.He displayed true scientific excellence and generosity.I also want to acknowledge all exam project students, Ph.D. students, post-docs, and collaborators who have made research in my group possible.Too many to mention, you can find their names in my publications. Conflicts of Interest: The author declares no conflicts of interest.Acknowledgments: I want to acknowledge Bengt Mannervik, who was one of my teachers when I studied Biochemistry at Stockholm University.His lectures inspired interest in enzymology and particularly enzyme kinetics.Later I started as a Ph.D. student in Joe DePierre's research group.There I was lucky to have Bengt's group as neighbors.His group shared knowledge, reagents, and expertise, which was absolutely essential for our research on microsomal glutathione transferase.Over the years I have had the advantage to collaborate with Bengt on many occasions.I see him as a tutor and curiosity-driven scientist with absolute and stringent scientific rigor.But most importantly he has become a close friend.I want to acknowledge my Ph.D. supervisor Joe DePierre who sadly passed this year.He displayed true scientific excellence and generosity.I also want to acknowledge all exam project students, Ph.D. students, post-docs, and collaborators who have made research in my group possible.Too many to mention, you can find their names in my publications. Conflicts of Interest: The author declares no conflicts of interest. Figure 1 . Figure 1.Although GSH transferases display a random sequential kinetic mechanism, in practice, as a result of rare turnover in vivo, GSH binds first forming the thiolate anion (highlighted in red).A large concentration of catalytically competent enzymes ensures efficient capture of reactive intermediates.Proton release upon thiolate formation has been demonstrated[65,66]. Figure 1 . Figure 1.Although GSH transferases display a random sequential kinetic mechanism, in practice, as a result of rare turnover in vivo, GSH binds first forming the thiolate anion (highlighted in red).A large concentration of catalytically competent enzymes ensures efficient capture of reactive intermediates.Proton release upon thiolate formation has been demonstrated [65,66]. Figure 2 . Figure 2.An impressive amount of data and knowledge on GST structure, mechanisms, and kinetic behavior has been gathered.However, we still need to determine cellular amounts and distribution of GSTs in vivo more precisely.Their ligand saturation, co-operative behavior, and relevance to protective capacity need to be understood.Just as global kinetic mechanisms are rigorously defined for individual enzymes, this knowledge needs to be expanded and used in the elucidation and description of the global role of GSTs in cells, whole organs, and organisms.Funding: The Swedish Cancer Society and the Swedish Research Council were the main financial contributors. Figure 2 . Figure 2.An impressive amount of data and knowledge on GST structure, mechanisms, and kinetic behavior has been gathered.However, we still need to determine cellular amounts and distribution of GSTs in vivo more precisely.Their ligand saturation, co-operative behavior, and relevance to protective capacity need to be understood.Just as global kinetic mechanisms are rigorously defined for individual enzymes, this knowledge needs to be expanded and used in the elucidation and description of the global role of GSTs in cells, whole organs, and organisms.Funding: The Swedish Cancer Society and the Swedish Research Council were the main financial contributors.	2024-06-02T15:16:46.342Z	2024-05-30T00:00:00.000	{ "year": 2024, "sha1": "087fc4f2a73f84753def13634c1ee342ca030ab4", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/2218-273X/14/6/641/pdf?version=1717051612", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "8e39dd06c1bf875d4081c3166bca7ac0d5f57bcc", "s2fieldsofstudy": [ "Chemistry", "Biology", "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
8930756	pes2o/s2orc	v3-fos-license	Comparison of the Control of Nausea and Vomiting among Several Moderately Emetic-Risk Chemotherapy Regimens Background: Different antiemetic medications with or without aprepitant are recommended for moderately emetic-risk chemotherapy (MEC) depending on the emetic potential of chemotherapy agents, although the criterion for the use of aprepitant is still unclear. The present study was designed to compare the control of chemotherapy-induced nausea and vomiting (CINV) among several MEC regimens used in the outpatient chemotherapy setting. Materials and Methods: A single center prospective observational study was carried out in 326 patients who received 2,061 chemotherapy cycles from January 2013 to December 2014. Antiemetic medication consisting of two-drug combination of granisetron (day 1) and dexamethasone (days 1-3) was carried out in 87.6% of patients receiving the first chemotherapy cycle. The checklist for CINV was provided to all patients, and the control of CINV was evaluated on the next visit based on the checklist. Complete inhibition of nausea and vomiting during acute and delayed periods were compared among MEC regimens. Results: Two hundred and one patients received the first cycle of chemotherapy, in which the rates of complete inhibition of nausea and vomiting were 87.6% and 95.5%, respectively, during acute period, and 68.2% and 92.0%, respectively, during delayed period. There were no significant differences in the control of CINV among oxaliplatin, carboplatin and irinotecan, except for the cyclophosphamide-base regimen. Conclusions: Two-drug antiemetic medication of 5-HT3 receptor antagonist and dexamethasone was sufficiently effective for prevention of CINV in most MEC regimens. Introduction Chemotherapy-induced nausea and vomiting (CINV) is one of most distressing adverse events during cancer chemotherapy [1]. Several clinical practice guidelines for prevention of CINV have been developed by the American Society of Clinical Oncology (ASCO) [2], Multinational Association of Supportive Care in Cancer (MASCC) [3], National Comprehensive Cancer Network (NCCN, 2015) [4], and Japanese Society of Clinical Oncology (JSCO) [5]. It has been reported that the implementation of the guideline-consistent antiemetic medication greatly improves the control of vomiting [6,7], although nausea that occurs during chemotherapy remains to be the least favorable adverse drug reaction [8,9]. On the other hand, the antiemetic medication for moderately emetic-risk chemotherapy (MEC) is confusing: According to the guidelines from ASCO (2012), NCCN (2015) and JSCO (2010), two-drug combination of palonosetron (day 1) and dexamethasone (days 1-3) is recommended as the standard antiemetic medication for MEC, but the addition of aprepitant to the standard regimen is an optional choice for MEC with relatively high emetic risk. However, the definition of relatively high emetic risk Ivyspring International Publisher anticancer drugs is vague. JSCO (2010) [5] indicated that aprepitant can be used for patients receiving carboplatin, ifosphamide, irinotecan, and methotrexate, while NCCN (2015) [4] suggested that aprepitant should be added to the standard regimen for select patients with additional risk factors or those who experienced CINV in previous therapy using two-drug regimen. ASCO (2010) [2] merely described that limited evidence supports the addition of aprepitant to the two-drug combination therapy by referring from the report of Rapoport et al. [10], in which aprepitant improves the rate of the complete inhibition of vomiting but not of the complete response in patients receiving oxaliplatin, carboplatin, epirubicin, idarubicin, ifosfamide, irinotecan, daunorubicin, doxorubicin, cyclophosphamide (<1,500 mg/m 2 ) or cytarabine (>1 g/m 2 ). To determine whether or not another antiemetic should be added to the standard two-drug antiemetic regimen in patients receiving MEC, the rates of complete inhibition of nausea and vomiting during acute and delayed periods were compared among several MEC regimens under the condition of two-drug antiemetic medication. Patients A total of 326 patients of >18-year-old in age received 2,061 cycles of moderately emetic-risk chemotherapy (MEC) regimens in our outpatient chemotherapy clinic during two years from January 2013 to December 2014, in which 201 patients received the first chemotherapy cycle. The exclusion criteria was age under 18-year-old, patients having nausea and/or vomiting due to organic causes such as brain metastasis and tumor infiltration of the bowel or other gastrointestinal abnormality. MEC used in the present study included oxaliplatin, carboplatin, irinotecan, cyclophosphamide, nedaplatin, and bendamustine. These anticancer drugs were used as the single-day regimen in combination with or without multiple treatment with low emetic risk or minimal emetic risk agents. The combination of cyclophosphamide and anthracyclines was regarded as highly emetic-risk chemotherapy (HEC) and excluded from the present study. Pharmacists were in charge of provision of drug information and safety precaution in daily life, and monitoring adverse drug reactions, including CINV, to all patients in our outpatient chemotherapy clinic. Antiemetic medication was carried out according to the Japanese Society of Clinical Oncology (JSCO) clinical practice guideline for antiemesis, in which the combination of 5-HT3 receptor antagonist such as intravenous granisetron (3mg) and intravenous dexamethasone (9.9mg) was prescribed before chemotherapy for prevention of acute CINV, and oral dexamethasone (4-8mg/day) was administered on days 2 and 3 of chemotherapy for prophylaxis of delayed CINV in the first cycle of chemotherapy. In patients who experienced CINV in the previous cycle, antiemetic drugs with different mode of action, including aprepitant [11], olanzapine [12] or benzodiazepines [13], were added to the standard antiemetic medication on the next course of chemotherapy. Evaluation of the control of CINV Patients were all provided with a checklist for daily check of CINV on the first visit of outpatient chemotherapy clinic. Using the checklist, patients checked daily the nausea by numeric scale (NRS: 0-10) and the number of vomiting episodes up to 7 days after chemotherapy. Pharmacists had an interview to all patients who visited to our outpatient chemotherapy clinic and asked the presence or absence of CINV, regardless of whether patients filled the checklist of adverse drug reactions, including CINV. The control of nausea and vomiting was recorded on the electric medical record after verifying them or hearing from patients on the next visit. Complete inhibition of nausea (NRS scale < 1) and vomiting (no episode) during acute (within 24 hours after chemotherapy) and delayed (during 2-7 days after chemotherapy) periods was assessed in patients receiving the first cycle of chemotherapy or in those with overall cycles of chemotherapy. In our outpatient chemotherapy setting, on-demand use of antiemetic medication was not prescribed in most cases, therefore, the present data were devoid of information about rescue treatment. Thus, the primary endpoint was complete inhibition of nausea/vomiting rather than complete response or complete protection. Statistical analyses Data were analyzed using IBM SPSS Statistics ver. 22 (IBM Japan Services Co., Ltd., Tokyo, Japan). Patients' demographics were compared among groups with different MEC regimens by one-way analysis variance (ANOVA) followed by Scheffe's test for parametric variables, and by Kruscal-Wallis test followed by Scheffe's test for non-parametric variables. P value of less than 0.05 was considered statistically significant. Ethical considerations The present study was carried out in accordance with the guidelines for the care for human study adopted by the Ethics Committee of the Gifu Graduate School of Medicine, and notified by the Japanese Government (approved no. 26-153 of the Institutional Review Board). Demographics of patients As shown in Table 1, 326 patients received 2,061 chemotherapy cycles from January 2013 to December 2014 in our outpatient chemotherapy clinic. The most common type of cancer was colorectal cancer (51.8%), followed by lung cancer, and gynecologic cancer. The most frequently used anticancer drugs was oxaliplatin (39.9%), followed by irinotecan, and carboplatin, while the most prevalent chemotherapy regimen for overall cycles was FOLFIRI (29.1%), followed by mFOLFOX6, XELOX, and paclitaxel + carboplatin. Control of CINV in the first cycle of MEC regimens As shown in Table 2, among 201 patients, all received single-day MEC in combination with (N=192, 95.5%) or without (N=9, 4.5%) multiple treatment with either low emetic risk (N=191) or minimal risk (N=1, 0.5%) chemotherapy agents. No multiple-day MEC regimens existed in the first cycle of the present study. Antiemetic premedication was carried out in 100% on day 1 and 87.6% on days 2-3. The rate of complete inhibition of nausea and vomiting was 87.6% and 95.5%, respectively, during acute period, while the rate was 68.2% and 92.0%, respectively, during delayed period. As shown in Figure 1A, there were no significant differences in the control of nausea and vomiting during acute and delayed periods among carboplatin, irinotecan, and oxaliplatin, except for cyclophosphamide. The rates of complete inhibition of acute and delayed vomiting were significantly (p=0.017) lower in cyclophosphamide-base regimens (n=14), including cyclophosphamide (600mg/m 2 ) + docetaxel (75mg/m 2 ) for breast cancer (n=11), cyclophosphamide (100mg/m 2 ) + methotrexate (40mg/m 2 ) + 5-fluorouracil (600mg/m 2 ) for breast cancer (n=1) and cyclophosphamide (600mg/m 2 ) + vincristine (1.5 mg/m 2 ) for malignant lymphoma (n=2). Figure 1B shows a comparison of the rates of complete inhibition of nausea and vomiting during acute and delayed periods among respective MEC regimens, including oxaliplatin + capecitabine (XELOX) or TS-1 (SOX), modified FOLFOX6 (mFOLFOX6), FOLFIRI, carboplatin in combination with pemetrexed, gemcitabine, paclitaxel or docetaxel, and cyclophosphamide + docetaxel. The rates of complete inhibition of nausea and vomiting were almost similar among these regimens except for cyclophosphamide + docetaxel. The rates of complete inhibition of vomiting during acute and delayed peri-ods were significantly lower in cyclophosphamide + docetaxel than in other regimens, although the rates of complete inhibition of nausea tended to be lower in cyclophosphamide + docetaxel. Table 3 shows the comparison of patients' demographics among MEC such as carboplatin, iri-notecan, oxaliplatin and cyclophosphamide. The percentage of female in irinotecan group was significantly lower than those in other groups. Moreover, patients receiving cyclophosphamide were significantly younger than those receiving other regimens. Figure 2 shows the rates of complete inhibition of nausea and vomiting during acute and delayed periods in multiple chemotherapy cycles from the first to the sixth or fourth cycle. No marked differences in the control of nausea or vomiting were observed among cycles of any MEC regimen. Discussion In our outpatient chemotherapy clinic, pharmacists met with all patients and were in charge of monitoring of adverse events, including CINV, planning of preventive measures against adverse events, and verification of the prescription order regarding cancer chemotherapy regimens. In the present study, the two-drug antiemetic medication of granisetron (day 1) and dexamethasone (days 1-3) was prescribed in 87.6% and 82.4% of the first cycle and overall cycles, respectively. Under such conditions, the rates of complete inhibition of nausea and vomiting were 87.6% and 95.5%, respectively, during acute period of the first chemotherapy cycle, and 68.2% and 92.0%, respectively, during delayed period of the first cycle. Our data on the control of CINV were generally consistent with those reported by Escobar et al. [14], who showed in 240 chemo-therapy-naïve patients receiving MEC that the rates of complete inhibition of nausea and vomiting are 76.7% and 90.8%, respectively, during acute period, and 61.5% and 83.5%, respectively, during delayed period. The slightly higher rates of the control of CINV observed in the present study as compared with those reported by Escobar et al. [14] may be due to the higher prevalence rate of antiemetic medication in our study as compared with their data (100% and 87.6% versus 94.9% and 43.8% during acute and delayed periods, respectively), since patients' demographics such as age, gender, proportion of colorectal cancer were similar between the two studies. It has been demonstrated that the control of CINV is significantly improved by adherence to the antiemetic guideline [6,7,15]. It was notable that the control of CINV in the first cycle were similar among various MEC regimens, including carboplatin, irinotecan, and oxaliplatin. However, the control of CINV was worst in cyclophosphamide-base regimens, including predominantly docetaxel + cyclophosphamide for breast cancer, as compared with other regimens. The demographics of patients showed that the proportion of female was highest in cyclophosphamide-base regimens. Moreover, patients receiving cyclophosphamide-base regimens were significantly younger than those of any other group. It has been demonstrated that female and younger age are significant risks for developing CINV. Sekine et al. [16] reported in 1,549 patients receiving HEC or MEC that female is more likely to show a failure in complete response than male. Hilarius et al. [17] also showed in 225 patients receiving HEC or MEC that the incidence rates of acute and delayed nausea as well as vomiting are significantly higher in female than male. Younger age is also linked with higher incidence of CINV, although the cut-off value for age varied depending on each study ranging from 40-year-old [18] to 65-year old [17,19]. Therefore, it is suggested that such risk factors rather than cyclophosphamide itself might be involved in higher incidence of CINV in patients receiving cyclophosphamide-base regimens observed in the present study, although we could not exclude a possibility that cyclophosphamide is a high risk for CINV [20]. The control of CINV was similar or rather improved after multiple chemotherapy cycles from the first cycle to the sixth or fourth cycle. This may be due at least in part to the following reasons: antiemetic drugs with different mode of action such as olanzapine and aprepitant were added to the two-drug antiemetic regimen in the subsequent cycle in patients who experienced significant CINV in previous cycle. Indeed, among 35 patients showing significant CINV in the first cycle, 29 patients received subsequent chemotherapy cycles. Among 29 patients, 18 patients (62.1%) were treated with additional antiemetic drugs, including aprepitant (20.7%), olanzapine (17.2%), and palonosetron (13.8%). The incidence rates of grade>2 nausea and vomiting in the first cycle tended to be higher in the additional antiemetic drugs-treated group than non-treated group (94.4% and 16.7% for nausea and vomiting, respectively, in additional antiemetic drugs-treated group versus 72.7% and 0%, respectively, in non-treated group). The rate of complete inhibition of nausea and vomiting during overall period was improved from 0% in the first cycle to 22.2% in the subsequent cycle with additional antiemetic drugs, although the rate was also elevated from 0% to 18.2% in the subsequent course without additional antiemetic medication. Aprepitant is known to be highly effective for prevention of vomiting [11] but to a lesser extent for the prophylaxis of nausea [21]. On the other hand, it has been reported that olanzapine, atypical antipsychotics agent, is more potent than aprepitant in preventing chemotherapy-induced nausea [12]. This may be due to the binding profile of olanzapine showing affinity for a variety of neurotransmitter receptors, including it only dopamine D2 receptors but also 5-HT 2B and 5-HT 2C receptors [22] that control the secretion of ghrelin [23], an appetite-promoting hormone. Therefore, aprepitant was added in patients who experienced vomiting, while olanzapine was included in case nausea was not controlled in the previous cycle. Taken together, it is suggested that two-drug antiemetic medication including 5-HT3 receptor antagonist and dexamethasone is enough to prevent CINV in patients receiving MEC other than docetaxel/cyclophosphamide regimen for breast cancer, in which three-drug antiemetic medication is required. However, aprepitant or olanzapine may be added to the two-drug regimen in patients with high risk of CINV, including younger age and female, or in those who did not control CINV in the previous cycle, as indicated by the NCCN 2015 guideline. There are several limitations in the present study. First, the present study was a non-randomized single-center study. Second, the small number of chemotherapy naïve patients were enrolled and the number of patients in each MEC regimen was so small as to clearly demonstrate the necessity of aprepitant in the specific MEC regimens. Third, gastrointestinal cancers such as colorectal cancer and stomach cancer were predominant (approximately 60 % of all patients), therefore, limited range of anticancer drugs were used in the present study. Fourth, complete response, one of most common indices of the control of CINV, was not included in the present study because of the difficulty in obtaining data on the rescue treatment from the outpatient setting. In conclusion, two-drug antiemetic regimen showed favorable control for most of the cases underwent MEC, however, more intensive supportive care might be necessary, especially for controlling nausea, in some of the patients or in some of the chemotherapy regimens.	2017-10-12T01:31:33.414Z	2016-03-18T00:00:00.000	{ "year": 2016, "sha1": "534ba3224ebfd52a699177c5d9ff1629bd8f5646", "oa_license": "CCBYNC", "oa_url": "https://www.jcancer.org/v07p0569.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "534ba3224ebfd52a699177c5d9ff1629bd8f5646", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
245514801	pes2o/s2orc	v3-fos-license	Off-label prescription of medicines: what do we know about the legislation in EU member states? Aim: Off-label prescription is not regulated on the European Union (EU) level and therefore not harmonised in the EU Member States (MS). Despite this, the use of medicines outside of the drug label occurs in clinical practice, and it can be included in treatment guidelines and/or reimbursed in some cases. It is, however, currently not clear to what extent off-label use national questionnaire EMACOLEX. Case The identified categorical elements and prerequisites in national categorised. a comparison made to the five Good Off-Label Use Practice (GOLUP) principles. Conclusion: The main contribution of this work is to flag that off-label prescription actually needs to be well defined and understood before it can be appropriately taken into consideration in regulatory discussions. There is a heterogeneity in legislation regarding off-label prescription in the investigated MS, which may lead to different perspectives. A common understanding of the concept and more alignment in off-label prescription practices and their regulation at MS level may contribute to further regulatory discussions. INTRODUCTION Off-label use and prescription are often referred to in the context of different discussions related to medicine development, approval and use. Prescription is part of European health policies, but it falls within the responsibility of the Member States (MS) [1] . Hence, how individual MS regulate off-label prescription is an internal matter of the MS [2] . Additionally, based on different sources, the prevalence of off-label prescription is reported as being extremely variable. Recent judgements of the Court of Justice of the European Union have triggered a discussion about off-label use of medicinal products and the extent to which these practices and data generated may be considered in the context of assessment of dossiers for approval of medicinal products [3] . Several categories of medicines, including orphan medicinal products, medicines for paediatric use and medicines in oncology and the central nervous system, should be approved via the centralised procedure. This centralised authorisation procedure is covered by European Union (EU) legislation [4] . The clinical experience with off-label prescription in the discussions related to approval of medicinal products can be different. For instance, according to the Orphan regulation, one criterion that needs to be fulfilled (if applicable) is "significant benefit". That means that, in cases where other satisfactory methods exist for the same indication, the medicinal product applied should be shown to be of significant benefit for the patients affected by the same condition. When defining which medicinal products to include as a comparator for assessing significant benefit, off-label prescribed medicinal products should not be considered a satisfactory method of treatment [5] . On the other hand, off-label use is widely considered when discussing medicinal products for paediatric patients [6][7][8] . Moreover, off-label use can be the basis for well-established use procedures [9] . Such a procedure can be started when an active substance of a medicinal product has been used for more than 10 years and its efficacy and safety have been well established, often based on results from the scientific literature [9] . Related to the above, the aim of this research was to provide an overview of the current legislation concerning off-label prescription and use of medicinal products in the distinct MS of the EU. In our study, we used the following definition for off-label prescription: "prescription of medicinal products not in accordance with the Summary of Product Characteristics (SmPC) for which the medicinal product is authorised". Since this study focused on the EU, only medicinal products that have been authorised for at least one indication by the European Commission or by a National Competent Authority were included. Consequently, the term (marketing) authorisation in this study refers to approval in the EU. This definition was also used in the study of Weda et al. [10] (2017). Off-label prescription occurs when an authorised medicinal product is prescribed for a different patient group, another indication, via a different administration route or with a different frequency or altered dose than included in the SmPC of that authorised medicinal product [11,12] . First, we identified the elements present in national legislations associated with off-label prescription. Subsequently, we made a classification based on the legislation in the MS, as well as assessed the content of the legislation in relation to the principles from the Declaration on Good Off-Label Use in Practice (GOLUP) [13] . SOURCES AND METHODS We prepared a questionnaire consisting of four open questions about national legislation regarding off-label prescription [Supplementary Material 1]. This questionnaire was distributed among legal experts of the European Medicines Agencies Cooperation on Legal Issues (EMACOLEX). EMACOLEX is mandated to enhance trust, knowledge and confidence between legal staff in order to legally assist the European Medicines Regulatory Network and the national medicines agencies [14] . At least three email reminders were sent. Based on the EMACOLEX questionnaire, the national legislation references of the primary sources of national legislation -the laws in their original language -for each MS were identified. Subsequently, the acts and regulations that were indirectly or directly associated with off-label prescription were examined in English with the use of a translation tool, with exception of the Dutch legislation which was studied in the original language. Case law was excluded, because the legal status varies between the countries and in most MS it is published in the original language only. Despite the existing possibility for translation from the original language, we decided not to follow this approach as it is known that certain nuances can be lost in translation and hence lead to incorrect interpretation. Identification of categorical elements in law from national legislation The method for comparative legal research, as executed by Elkins et al. [15] in 2009 and applied by Voermans [16] in 2019, was used for the identification of elements in law that were associated with off-label prescription. The identified regulated subjects from the national legislation of MS were categorised into categorical elements in law and the frequency of each of these elements was counted [ Figure 1]. Analysis of classification First, based on the nature of the legislation, a classification was made between explicit and implicit regulation regarding off-label prescription. MS that make in their national legislation a distinction between non-authorised medicinal products ("use or prescription without marketing authorisation") and authorised off-label medicinal products ("use or prescription outside the terms of the marketing authorisation") were categorised to have explicit regulation. MS that did not make such distinction were classified as having implicit regulation. Since this study focused on the EU, "marketing authorisation" refers to authorisation by the EC or any of the national competent authorities. Second, MS were classified based on the extent of regulation, based on two parameters: the number of regulations and the number of prerequisites. The number of regulations was defined by the number of distinct sections -an article of law in a certain act or code -that have a direct or indirect association with off-label prescription of medicinal products. The number of distinct subsections or sub-articles of regulations was not counted. The number of regulations which are related to off-label prescription shows to what extent a MS has made regulations about the topic. The number of prerequisites was defined by the number (between 0 and 14) of prerequisites for off-label prescription. Prior to the analysis of classification, prerequisites were identified from the categorical elements in legislation. The number of prerequisites is an indication of the level of granularity in which the topic of "off-label prescription" is addressed in a certain MS. The complete list of these prerequisites can be found in Supplementary Material 2. Analysis of content of legislation To assess the content of legislation, the prerequisites for off-label prescription identified in the national legislation of MS were related to the five GOLUP principles [Supplementary Material 3] [13] . The frequency of each prerequisite was counted. GOLUP aims to create a harmonised approach on how and when off-label prescription might take place in the EU [13] . The five GOLUP principles are as follows: 1. "Presence of a medical therapeutic need based on a current examination of the patient by a suitably qualified health care professional". 2. "Absence of authorised treatment and licensed alternatives tolerated by the patient or repeated treatment failure". 3. "A documented review and critical appraisal of available scientific evidence favours off-label use to respond to the unmet medical need of the individual patient". 4. "Patients (or their legal representative) must be given sufficient information about the medicines that are prescribed to allow them to make an informed decision". 5. "Presence of established reporting routes for outcomes and adverse events linked to off-label use" [13] . RESULTS Responses were received from 10 EU MS and 3 EEA countries [ Figure 2]. Since this research was initiated to inform EU regulatory discussions, the information from the EEA countries was appreciated but not included in the final analysis. In total, 24 categorical elements in legislation were identified. The full list of categorical elements can be found in Supplementary Material 4. Figure 3 displays the elements that were found in the legislation of two or more MS. In five out of 10 MS, there was no definition or reference to offlabel prescription in the national legislation, whereas, in the other five MS, a definition or reference to offlabel prescription was present. In five MS, "informed patient consent" was found. Categorical elements such as the "professional standards of physicians", "responsibility in case of off-label prescription" and "notification of a policy organ" were found in the lowest number of MS. In addition, "Distinction" (five MS) and "No distinction" (five MS), respectively, refer to whether there is a distinction in legislation between non-authorised medicinal products and off-label prescribed medicinal products or not. This in fact categorises a MS as having implicit or explicit legislation. Table 1 provides an overview of the extent to which off-label prescription is regulated. It includes the number of regulations, the number of prerequisites and the nature of regulation (explicit or implicit). The number of regulations that are associated to off-label prescription varies between one (Bulgaria, Czech Republic and Estonia) and five (Germany). Despite these differences in the extent of regulation, all MS have more general regulations, such as ethical or professional standards of the physician in general. Seven out of 10 MS have prerequisites for off-label prescription. Such prerequisites concern, e.g., "informed patient consent" or "notification of a policy organ". Bulgaria has the most prerequisites for off-label prescription (8) and Ireland, Romania and Slovenia the fewest (0). However, the MS without prerequisites (Romania, Ireland and Slovenia) do have regulations that are related to the topic. Classification of legislation associated with off-label prescription Lastly, five out of 10 MS were classified as having explicit regulation: Bulgaria, Czech Republic, Germany, The Netherlands and Slovenia. The other five MS were classified as having implicit regulation: Estonia, Finland, Ireland, Romania and Sweden. The legislation text of the MS with explicit regulation can be found in Supplementary Material 5, and the number of regulations and national legislation references per Member State is presented in Supplementary Material 6. Table 2 shows the 14 prerequisites for an off-label prescription of a medicinal product that have been identified in the law of the 10 MS. For each prerequisite, the relation to the GOLUP principles, the frequency and the MS in which the prerequisite has been identified are included. Except for "no payment of off-label treatment with public funds", all identified prerequisites are related to the five principles stated in GOLUP Declaration [13] . In fact, the prerequisite "no payment of off-label treatment with public funds" is related to the general aim of the GOLUP declaration. Content of legislation regarding off-label prescription The three most frequently applied GOLUP principles were: Principle 4 about informing the patient about the consequences of off-label prescription (nine times), Principle 5 regarding reporting outcomes and adverse events associated with off-label use (seven times) and Principle 3 concerning the scientific evidence for off-label use (five times). GOLUP Principle 2 about absence of alternative authorised treatments or repeated treatment failure was found only once. Bulgaria is the only MS that has included all five principles in their national legislation. However, all MS that have prerequisites for off-label prescription in their law (seven out of 10) have at least two prerequisites which are related to at least two different GOLUP principles, except for Germany that has two prerequisites that are related to the same GOLUP principle. A comparison of the prerequisites in the different countries has shown different levels of similarities. For instance, Bulgaria and Sweden share three similar prerequisites (the highest "n" of similarities): "informed patient consent", "scientific evidence" and "notification of a policy organ". The last two are also found in the Czech Republic. Germany, Estonia and the Netherlands have two similar prerequisites: requirement for "informed patient consent" and "informing the patient about the consequences of the treatment". DISCUSSION We studied the current legislation concerning off-label prescription and use of medicinal products in the distinct MS of the EU, which is of interest, as there is no EU legislation for off-label prescription. Hence, it is important to identify the exact nature of the national legislation in the different MS of the EU. Especially for marketing authorisation applications for medicinal products, it is currently not clear to what extent off-label use can be included in regulatory discussions at a European level at the different committees at the European Medicines Agency. Our study is the first to investigate in detail the legislation of off-label prescription in the EU. We found that the national regulation for off-label prescription is heterogeneous in the different MS of the EU. Seven out of the 10 MS that responded to the EMACOLEX questionnaire have prerequisites for off-label prescription in their national legislation, and the number of regulations related to off-label prescription varied between one and five. Furthermore, five MS had explicit and five MS had implicit regulation for off-label prescription. It is noteworthy that several countries (Finland, Estonia and Sweden) did not define or refer to off-label prescription in their national legislation, but nevertheless they all had prerequisites for off-label prescription. However, the opposite was also observed. Germany refers to off-label prescription in national legislation, while the legislation only contains general prerequisites (i.e., informed patient consent) that do not apply exclusively to prescription of off-label products but to all medicinal products. The lack of EU legislation and the current diversity of legislation regarding off-label prescription in the different MS leads to uncertainties. This issue was also flagged by Weda et al. [10] in their study from 2017. They investigated 24 MS and found that 10 MS have regulations concerning off-label prescription of medicinal products, while 14 MS do not have such regulations [10] . In our study, we further analysed to what extent national legislations reflect the principles formulated in the GOLUP document. This document was signed in December 2017 by various European associations to summarise the main principles and ensure high standards of patient care [13] . We identified that, in seven MS, the legislation and all 14 identified prerequisites are related to the GOLUP principles. The other three MS from the 10 responding countries do not have prerequisites at all and only in one MS (Bulgaria) the legislation contains all GOLUP principles. It seems that the GOLUP principles have been formulated based on the experience and elements from the legislation as reflected in the various MS. This study has some limitations. The first is that the analysis is based on 10 of the 27 MS which responded to the questionnaire. Consequently, it is unclear whether the regulation in these MS is representative for the whole EU. However, based on these 10 countries, it can already be concluded that off-label prescription is very heterogeneously regulated. The second limitation is that only the legislation of MS was taken into account. Case law was excluded, for reasons already explained. It could however contribute to a more complete picture of the off-label prescription regulation throughout MS in the EU. In conclusion, this review provides an overview of the current legislation regarding off-label prescription of medicinal products on MS level in the EU. The aim was to determine whether off-label prescription can be taken into consideration in regulatory discussions at EU level. We conclude that several steps are necessary before fully addressing this question. First, it is important to have one common understanding (a clear definition) of the concept of off-label prescription of medicinal products. Additionally, we checked to what extent national legislations' prerequisites have been reflected in the GOLUP principles of 2017. Despite the limitations of this study, we found that there is heterogeneity in legislation regarding off-label prescription in the investigated MS. Due to this, it may be expected that representatives of MS may have a different perspective regarding this concept. Hence, the main contribution of this work is to flag that off-label prescription actually needs to be well defined and understood before taken into consideration in regulatory discussions. A common understanding of the concept and more alignment in off-label prescription practices and their regulation at MS level may contribute to further regulatory discussions.	2021-12-29T16:16:26.717Z	2021-01-01T00:00:00.000	{ "year": 2021, "sha1": "27f7c0b36f9db91d894d0edd8dd6ca021390ede1", "oa_license": "CCBY", "oa_url": "https://rdodjournal.com/article/download/4472", "oa_status": "GOLD", "pdf_src": "MergedPDFExtraction", "pdf_hash": "7b650d0d484fb6a782b06428130b671bdf8358cc", "s2fieldsofstudy": [ "Medicine", "Political Science" ], "extfieldsofstudy": [] }
258958818	pes2o/s2orc	v3-fos-license	Oxidative Stress Inducers in Cancer Therapy: Preclinical and Clinical Evidence Reactive oxygen species (ROS) are metabolic byproducts that regulate various cellular processes. However, at high levels, ROS induce oxidative stress, which in turn can trigger cell death. Cancer cells alter the redox homeostasis to facilitate protumorigenic processes; however, this leaves them vulnerable to further increases in ROS levels. This paradox has been exploited as a cancer therapeutic strategy with the use of pro-oxidative drugs. Many chemotherapeutic drugs presently in clinical use, such as cisplatin and doxorubicin, induce ROS as one of their mechanisms of action. Further, various drugs, including phytochemicals and small molecules, that are presently being investigated in preclinical and clinical studies attribute their anticancer activity to ROS induction. Consistently, this review aims to highlight selected pro-oxidative drugs whose anticancer potential has been characterized with specific focus on phytochemicals, mechanisms of ROS induction, and anticancer effects downstream of ROS induction. Introduction Reactive oxygen species (ROS) are molecules that contain one or more unpaired electrons, which contribute to their high reactivity. ROS can be classified as free radicals and nonradical molecules [1]. Examples of free radicals include superoxide (O 2 •− ), hydroxyl (HO • ), and peroxyl (RO 2 •− ) radicals, and nonradical molecules include hydrogen peroxide (H 2 O 2 ) and organic peroxides (ROOH). Physiologically, ROS are produced as byproducts of metabolic reactions, namely electron leakage during oxidative phosphorylation in the mitochondria, through the activity of NADPH oxidases, and via the iron-dependent Fenton reaction [2,3]. In normal cells, intracellular ROS levels are regulated by antioxidants and maintained at levels that regulate various signaling pathways, including cell proliferation, metabolism, and differentiation, among others. At levels beyond the physiological threshold, ROS cause oxidative damage to various cellular components, namely nucleic acids, proteins, and lipids, which in turn can induce cell death [4]. In cancer cells, the redox balance is disrupted due to several factors, including activation of oncogenes, aerobic glycolysis, and hypoxia. This can result in the accumulation of ROS, which can be lethal to cancer cells [3,5]. Therefore, to combat this issue and prevent oxidative stress-induced cell death, cancer cells employ several mechanisms, such as enhanced expression and activity of components of the antioxidant defense system, for example, peroxide scavenging systems that reduce H 2 O 2 to H 2 O. Additionally, the availability of O 2 , mitochondrial localization in the cell, and the rates and concentrations of electrons and their carriers in the electron transport chain further regulate the production of ROS, specifically mitochondrial ROS [6]. Collectively, this maintains ROS at levels that facilitate activation of Figure 1. The paradoxical role of reactive oxygen species (ROS) in cancer. In normal cells, ROS levels are tightly regulated by a balance between pro-oxidative cellular processes and the antioxidant responses. Several factors can increase ROS production in cancer cells to levels which provide a survival advantage by activating tumorigenic pathways to drive tumor progression. This can be therapeutically exploited through ROS modulation with pro-oxidative drugs which tip the already skewed redox balance and further increase ROS production to levels that mediate oxidative stressinduced cell death. ROS-reactive oxygen species. Indeed, many chemotherapeutic drugs exhibit their anticancer effects through direct or indirect induction of oxidative stress. This review aims to highlight anticancer drugs that attribute their anticancer activity primarily to ROS induction, such that the use of ROS inhibitors/scavengers abrogates their anticancer effects. In addition to the above-mentioned rationale, we further narrowed down the drugs highlighted in this review to meet the following criteria: (1) they exhibit their anticancer activity in two or more cancer types; (2) the pathways that modulate their anticancer effects are induced downstream of ROS. Moreover, given the growing interest in phytochemical compounds as anticancer therapeutics, we specifically focused on pro-oxidative phytochemicals and further classified them based on whether the mechanism of ROS induction has been elucidated. In normal cells, ROS levels are tightly regulated by a balance between pro-oxidative cellular processes and the antioxidant responses. Several factors can increase ROS production in cancer cells to levels which provide a survival advantage by activating tumorigenic pathways to drive tumor progression. This can be therapeutically exploited through ROS modulation with pro-oxidative drugs which tip the already skewed redox balance and further increase ROS production to levels that mediate oxidative stressinduced cell death. ROS-reactive oxygen species. Indeed, many chemotherapeutic drugs exhibit their anticancer effects through direct or indirect induction of oxidative stress. This review aims to highlight anticancer drugs that attribute their anticancer activity primarily to ROS induction, such that the use of ROS inhibitors/scavengers abrogates their anticancer effects. In addition to the abovementioned rationale, we further narrowed down the drugs highlighted in this review to meet the following criteria: (1) they exhibit their anticancer activity in two or more cancer types; (2) the pathways that modulate their anticancer effects are induced downstream of ROS. Moreover, given the growing interest in phytochemical compounds as anticancer therapeutics, we specifically focused on pro-oxidative phytochemicals and further classified them based on whether the mechanism of ROS induction has been elucidated. Pro-Oxidative Drugs in Preclinical Study This section discusses pro-oxidative anticancer drugs, with special emphasis on phytochemicals, which are currently being investigated in preclinical studies. Phytochemicals with a Characterized Mechanism of of ROS Induction An overwhelming proportion of anticancer drugs that have been approved in the past decades are derivatives of natural products. For example, paclitaxel, a taxane diterpene that is widely used for the treatment of various cancers, including breast, endometrial, and ovarian cancers, was originally derived from the Pacific yew (Taxus brevifolia) [12]. This section highlights selected pro-oxidative phytochemical compounds that attribute their anticancer activity to ROS-mediated mechanisms, and whose mechanism of ROS induction has been characterized. Piperlongumine Piperlongumine (PPL) is an alkaloid/amide that was identified in root extracts of long pepper (Piper longum) and has been investigated for its anticancer activity in various cancer types, including hematological cancers, colorectal, gastric, lung, breast, prostate, and oral cancers, melanoma, and glioma [13,14]. Its in vitro anticancer activity can be attributed to induction of ROS through increased glutathione disulfide levels, decreased glutathione levels, and inhibition of thioredoxin reductase (TrxR), an enzyme which reduces thioredoxin, a redox protein that protects against oxidative stress [13,14]. PPL-mediated ROS accumulation further leads to ROS-mediated apoptosis [15], G1 or G2/M cell cycle arrest [16,17], ER stress [15], and oxidative DNA damage [17]. PPL was also found to modulate various signaling pathways involved in tumor progression, including the JAK/STAT3, ERK, NF-κB, and PI3K/AKT/mTOR pathways [13,14]. Consistent with the regulation of these important pathways, which are also involved in chemoresistance, PPL was reported to sensitize head and neck, gastric, and liver cancers to cisplatin [18], oxaliplatin [19], and sorafenib [20], respectively, through induction of ROS both in vitro and in vivo. However, despite the promising potential of PPL as both an anticancer therapeutic and a chemosensitizing agent, its pharmacokinetics have not been clarified. Additionally, its poor aqueous solubility and bioavailability limit its therapeutic potential [13,14]; however, efforts have been made on this front in relation to drug delivery systems, especially with nanoparticles and nanoemulsions, which have shown promising results [14,21,22]. Resveratrol Resveratrol (3,5,4 -trihydroxy-trans-stilbene) is a stilbenoid that is found in various plant species, including common grape vine (Vitis vinifera) and berries of the genus Vaccinium, and has been investigated for its biological activities including anticancer, antiinflammatory, and antimicrobial properties, among others. The anticancer activity of resveratrol is well characterized both in vitro and in vivo in various cancer types [23]. Several molecular mechanisms have been proposed for the anticancer activity of resveratrol, including ROS induction. However, with respect to ROS, the effect of resveratrol appears to be concentration dependent; at low concentrations, it exerts antioxidant effects, whereas at high concentrations (50-100 µM), resveratrol induces ROS production [24,25], which can be attributed to increased NADPH oxidase activity [25]. Resveratrol-induced ROS accumulation modulates autophagy and apoptosis in colon [25,26], pancreatic [27], and bladder cancer [28] cells. Cheng et al. [27] reported that resveratrol-induced ROS activate the Nrf2 signaling pathway, which subsequently suppresses NAF1 and induces apoptosis in pancreatic cancer cells. This also increased their sensitivity to gemcitabine. Additionally, resveratrol induces ROS-mediated DNA damage [24,29], which has been reported to mediate senescence through the DLC1-DYRK1A-EGFR axis in breast and liver cancer in vitro and in vivo [29]. Despite the promising potential of resveratrol, its unstable pharmacokinetics due to its high metabolism and poor bioavailability limit its clinical application. Consistently, resveratrol analogues, such as 3,4,4 -trihydroxy-trans-stilbene (a synthetic analogue) [30] and piceatannol (a natural analogue) [31], have been investigated to overcome the same and are more potent, bypassing the high dosage issue, and were also found to exert their anticancer activity through ROS induction. Moreover, various nanocarriers have been investigated to optimize drug delivery and improve bioavailability; although many have been studied in vitro, in vivo studies are lacking for the same and are hence warranted [32]. Plumbagin Plumbagin (5-hydroxy-2-methyl-1,4-napthoquinone) is a naphthoquinone found in the roots of Leadwort (Plumbago zeylanica L.), and its anticancer activity has been well characterized against various cancers, including breast cancer, melanoma, glioma, hepatocellular cancer, oral squamous cell cancer, and T-cell lymphoma, among others [45]. Various studies have shown that plumbagin is a potent inducer of ROS. The mechanism underlying ROS induction by plumbagin has predominantly been attributed to inhibition of the antioxidant enzymes TrxR [46,47] and glutathione reductase [46]. In addition to TrxR inhibition, Hwang et al. clarified that plumbagin is a substrate of TrxR; it is reduced by TrxR, which inhibits the interaction of active TrxR with its substrate oxidized thioredoxin [46]. Downstream of ROS induction, plumbagin induces mitochondria-dependent apoptosis in hepatocellular cancer [46], lung cancer [46,48], and cervical cancer [46,49,50], leukemia [47], pancreatic cancer [51], oral squamous cell cancer [52,53], and osteosarcoma [54] among others. It also mediates its anticancer effect by inducing ER stress-mediated apoptosis [52,54,55], S/G2 and G2/M cell cycle arrest [48,49,56] and mitochondrial membrane depolarization in an ROS-dependent manner [45]. With respect to signaling pathways, plumbagin was found to inhibit the NF-κB [57], PI3K/AKT/mTOR [58] and MKP1/2 [59] pathways in non-small cell lung cancer, bladder cancer, and lymphoma, respectively. Additionally, plumbagin has been studied as an adjuvant drug to improve the efficacy of existing chemotherapeutic strategies. Namely, plumbagin was reported to improve the efficacy of chemical-based androgen deprivation therapy for prostate cancer in vivo [60] and a synthetic version of plumbagin, PCUR-101, is presently being explored for the same in a Phase I clinical trial (NCT04677855) with patients with metastatic castration-resistant prostate cancer. It was also reported to improve the efficacy of cisplatin in oral squamous cell carcinoma [53]. Collectively, these findings highlight the potential of plumbagin as a pro-oxidative anticancer agent that warrants further research. Capsaicin Capsaicin (trans-8-methyl-N-vanillyl-6-nonenamide) is an alkaloid that is found naturally in plants of the genus Capsicum (chili peppers) and contributes to the burning sensation attributed to spice. Various biological properties have been attributed to capsaicin, including anti-inflammatory and analgesic activities. Intriguingly, although a link between chili pepper consumption and oral and gastrointestinal cancers has long been suggested, capsaicin has been reported as both a chemopreventive and as an anticancer agent [61,62]. Capsaicin has been reported to induce ROS-dependent cell death in various cancers, including colorectal [63], prostate [64,65], bladder [66][67][68], and pancreatic [69,70] cancers. It has also been reported to inhibit tumor growth in vivo in mouse xenograft models of prostate [64] and bladder [66] cancers. Mechanistically, capsaicin-mediated ROS accumulation leads to mitochondrial membrane depolarization [63,64,66], which further triggers mitochondria-dependent apoptosis, as well as G0/G1 cell cycle arrest [68]. Additionally, Sánchez et al. reported that in bladder cancer cells, capsaicin induces JNK activation in an ROS-dependent manner, which results in ceramide accumulation and contributes to apoptosis [65]. Several mechanisms underlying capsaicin-mediated ROS accumulation have been reported, including: (1) inhibition of the activity of antioxidant enzymes SOD, catalase (CAT), and glutathione peroxidase [70]; (2) inhibition of the activity of mitochondrial complex-I and complex-III in the electron transport chain [70]; (3) downregulation of the expression of sirtuin-1, a NAD-dependent deacetylase that regulates the expression of various antioxidant enzymes [69]; (4) upregulation of the expression of NADPH oxidase 4, which generates superoxide [69]; (5) increased expression of FOXO3a, which is a transcription factor that regulates the oxidative stress response [68]. Celastrol Celastrol (24,25,26-trinoroleana-1(10),3,5,7-tetraen-29-oic acid) is a pentacyclic triterpenoid that was isolated from Tripterygium wilfordii (thunder duke vine), a plant that is commonly used in traditional Chinese medicine [71]. It has been widely studied as chemopreventive and anticancer drug, and its anticancer activity has been characterized in preclinical models against non-small cell lung [72], breast [73,74], colon [75,76], ovarian [77], gastric [78], and bladder [79] cancers. ROS induction has been attributed as the primary mode through which celastrol mediates its anticancer effects. Downstream of ROS, celastrol has been reported to inhibit HSP90 function [80], induce suppressor of specificity protein (Sp) repressors [79], activate the PKCzeta-AMPK-p53-PLK 2 signaling axis [73], and activate the JNK pathway [80,81] to induce apoptosis. With respect to other ROS-mediated anticancer effects, celastrol induces ER stress [78], mitochondrial dysfunction, specifically disruption of mitochondrial membrane potential [72,78,82], and cell cycle arrest at G2/M phase [76,77] and S phase [75]. Interestingly, at low concentrations (i.e., below the cytotoxic threshold) celastrol was found to induce autophagy in gastric cancer cells through ROSmediated accumulation of hypoxia-inducible factor 1-α via the transient activation of AKT. However, at cytotoxic concentrations, autophagic flux was inhibited and celastrol mediated p53-independent apoptosis through the JNK pathway [81]. As a pro-oxidative phytochemical, two mechanisms have been reported for ROS induction by celastrol: (1) inhibition of mitochondrial respiratory chain complex I activity [80]; and (2) inhibition of peroxiredoxins, namely peroxiredoxin-1 [76] and peroxiredoxin-2 [78]. The latter is thought to be the main mechanism of action, given that peroxiredoxins are upregulated in many cancer types and are involved in tumorigenesis and chemoresistance [83][84][85]. Consequently, celastrol derivatives are currently in development to improve their potency and specificity against peroxiredoxins [76]. Celastrol has been consistently reported to enhance the sensitivity of triple-negative breast cancer cells to tamoxifen [74] and non-small cell lung cancer cells to erastin [72] and induce ROS-mediated apoptosis in doxorubicin-resistant colorectal cancer cells [75]. These findings highlight the potential of celastrol as an anticancer drug, and its safety is consistently being investigated in an open-label safety study (NCT05494112). However, the clinical translation of celastrol is hindered by poor aqueous solubility, poor bioavailability, as well as potential side effects [71]. Hence, more studies are needed on this front to overcome these limitations to advance the clinical translation of this promising pro-oxidative phytochemical. Pro-Oxidative Phytochemicals with Uncharacterized Mechanism of ROS Induction In addition to the above-mentioned phytochemicals, various other phytochemicals have also been investigated for their pro-oxidative capacity with respect to their anticancer activity. However, the mechanism of ROS induction is not well characterized for many of these phytochemicals. For example, carnosol, a polyphenol, is one such pro-oxidative phytochemical whose mechanism of ROS induction has not been characterized. It was previously identified as an active constituent of sage (Salvia carnosa) and rosemary (Rosmarinus officinalis) extracts, and its anticancer activity was first characterized in the 1900s [86]. Carnosol was reported to induce ROS-dependent autophagy and/or apoptosis in colon cancer [87], triple-negative breast cancer [88], and osteosarcoma [89] cells. Our lab further clarified the underlying mechanisms of its activity: carnosol was found to target STAT3 [90] and p300 [91] to proteasomal degradation in an ROS-dependent manner and to function as a specific inhibitor of p300 in breast cancer [91]. Additionally, we found that carnosol induced p38-mediated endoplasmic reticulum (ER) stress in an ROS-dependent manner [92], which contributed to carnosol-induced autophagy and apoptosis in triple-negative breast cancer cells. Carnosol was also reported to modulate oxidative stress through depletion of the antioxidant glutathione, which induced apoptosis in adult T-cell leukemia/lymphoma cells [93]. However, the pro-oxidative activity of carnosol appears to be cancer specific, as carnosol has been reported to reduce ROS levels in non-melanoma skin cancer cells [94]. Additionally, very few studies have assessed the anticancer activity of carnosol in vivo, and this is limited to one or two studies each for breast, prostate, and skin cancers [95]. Future research with this promising phytochemical compound should focus on other cancer types and as well as in vivo studies to assess its safety/toxicity and pharmacokinetics. Similarly, allicin, a sulfenic acid thioester, which is a major bioactive component of garlic (Allium sativum), has been reported to induce cell cycle arrest and apoptosis in an ROS-dependent manner in various cancer cells [96], but the mechanism underlying the observed ROS induction has not been elucidated, to the best of our knowledge. A select few of such pro-oxidative phytochemicals, including allicin, with anticancer activity whose mechanism of ROS induction is not well characterized are listed in Table 1 alongside the anticancer effects induced downstream of ROS induction. Table 1. Pro-oxidative phytochemical compounds that exhibit anticancer activity whose mechanism of ROS induction has not been characterized. Table 1. Pro-oxidative phytochemical compounds that exhibit anticancer activity whose mechanism of ROS induction has not been characterized. Table 1. Pro-oxidative phytochemical compounds that exhibit anticancer activity whose mechanism of ROS induction has not been characterized. Table 1. Pro-oxidative phytochemical compounds that exhibit anticancer activity whose mechanism of ROS induction has not been characterized. Lung cancer screen 1 (LCS-1; 4,5-dichloro-2-m-tolylpyridazin-3(2H)-one) was first identified in 2009 from a small molecule screen for lung adenocarcinoma cell lines. It inhibited the growth of EGFR or KRAS mutant lung adenocarcinoma cell lines by impairing the activation of MAPK and AKT signaling pathways, which regulate cell growth [102]. The same group later identified SOD1 as a protein target for LCS-1 [103]. Consistent with this finding, LCS-1 has been reported to induce its anticancer effect through induction of mitochondrial superoxide and ROS in other cancers, including breast cancer [104], colorectal cancer [105,106], multiple myeloma [107], and glioma [108], which could be attributed to inhibition of SOD1. Downstream of ROS generation, oxidative DNA damageinduced apoptosis [105], ER stress [107], loss of mitochondrial integrity [104], and loss of proteosome function [107] have been reported as anticancer effects of LCS-1. Recently, Ling et al. reported that the anticancer activity of LCS-1 was independent of p53 function and that LCS-1 induced degradation of PARP and BRCA1 [108], suggesting that it inhibited DNA repair pathways. In contrast with numerous in vitro studies, only two studies [107,108] have assessed the in vivo activity of LCS-1 owing to its poor aqueous solubility and, hence, bioavailability. However, LCS-1-loaded triple-polymer-coated magnetite nanocarrier exhibit enhanced efficacy, which could overcome this limitation [109]. Additionally, LCS-1 seems to be a potential therapeutic candidate for bortezomib-resistant multiple myeloma [107] and tamoxifen-resistant breast cancer [110]. Further preclinical research is needed on this SOD1 inhibitor to investigate its anticancer activity in other cancer types, improve its efficacy and pharmacokinetics, and understand the molecular mechanisms underlying its effects. Tetraethylthiuram Disulfide Apart from developing novel anticancer drugs, repurposing drugs that have previously been approved for other conditions can also prove beneficial for the treatment of cancers. Disulfiram (tetraethylthiuram disulfide [DSF], also known as antabuse) is one such drug that is approved by the FDA for the treatment of alcoholism. Its pharmacodynamics and pharmacokinetics are well established, and its commercial availability is widespread. Recently, DSF-Copper (Cu) complexes have been shown to act as potent ROS inducers, specifically through the MAPK/ERK and PI3K/AKT pathways [124]. Cu is known to generate ROS; however, its use in therapeutics is limited because of its tightly controlled intracellular transport. This limitation can reportedly be overcome through complex formation with the DSF derivative N,N-diethyldithiocarbamate, which induces ROS-mediated apoptosis [125]. DSF/Cu complexes have been shown to increase ROS levels and induce G0/G1 cell cycle arrest in acute myeloid leukemia [126], and specifically trigger MAPK-mediated apoptosis in gastric cancer in an ROS-dependent manner [127]. Cu dependent ROS induction through DSF has been used to target prostate cancer, breast cancer, and lymphoid malignancies in preclinical studies [128][129][130]. Apart from Cu, other metal ions have also exhibited anticancer properties and great potential as cancer therapeutics. Namely, gold (III)-dithiocarbamate complexes were shown to induce ROS in breast cancer in part by interfering with the proteosome [131], as ROS can cause cellular disruptions that lead to the inactivation of the ubiquitination-proteasome pathway. Pro-Oxidative Drugs in the Clinical Setting This section discusses pro-oxidative anticancer drugs that are being investigated in clinical trials or are already in clinical use for cancer therapy. Tables 2 and 3 summarize clinical trials for the drugs discussed in this section. Choline tetrathiomolybdate (ATN-224) is an analogue of the copper chelator tetrathiomolybdate that is used for the treatment of Wilson's disease. ATN-224 was reported to exhibit anticancer activity via inhibition of SOD1, which resulted in increased superoxide levels and subsequent induction of ROS-dependent apoptosis in multiple myeloma cells [132]. This effect was further characterized in vitro in liver, ovarian, non-small cell lung cancer, colorectal, and pancreatic cancer [2]. Consistently, ATN-244 has been investigated in various clinical trials for solid organ tumors (including prostate, breast, esophageal, colorectal, and hepatocellular cancer) and multiple myeloma ( Table 2) [2]. However, the drug has not yet entered any Phase 3 trials. Recently, ATN-224 has been increasingly promoted as an adjuvant drug that can reverse chemoresistance. Combination treatment with cisplatin was found to increase ROS levels, decrease glutathione levels, and increase Platinum-DNA adduct formation, which enhanced the anticancer activity of cisplatin both in vitro and in vivo for non-small cell lung cancer [133]. Ryumon et al. previously reported similar findings for head and neck squamous cell carcinoma [134]. Additionally, the same study reported that pretreatment with ATN-224 sensitized the cisplatin-resistant A431-CDDP-R cell line to cisplatin due to suppression of ATPase copper transporting beta, suggesting that ATN-224 can be used as a chemosensitizing agent. Further studies are needed to assess the potential clinical use or repurposing of ATN-224 as a chemosensitizing adjuvant drug for other chemotherapeutic agents and cancer types. 2 clinical trials for glioblastoma multiforme, ovarian cancer, multiple myeloma, prostate cancer, and renal cell carcinoma (Table 2), and have FDA orphan drug designation for the former three [135]. The anticancer activity of 2-methoxyoestradiol can be attributed to increased ROS production through a debated mechanism, resulting in a loss of mitochondrial membrane potential [136,137] and nuclear localization of nitric oxide (NO) synthase, resulting in increased NO production and subsequent NO-induced DNA damage [138,139], both of which trigger apoptosis. Ongoing research involving 2-methoxyoestradiol is aimed at combating its main limitation, which is its poor bioavailability. Consistently, sulphamoylated analogues [140,141] and nanomedicine-based approaches [142,143] have been reported to improve the pharmacokinetics of 2-methoxyoestradiol in vivo and in patient-derived xenograft models. Curcumin and Its Derivatives Curcumin is the bioactive component of Curcuma longa L. (turmeric) and has been actively studied in the past few decades for its various pharmacological properties, including anticancer activity. Over the years, various derivatives of curcumin have been developed to improve its bioavailability and stability [144]. ROS induction has been implicated as one of the mechanisms of the anticancer activity of curcumin and its derivatives in various cancers, including leukemias [145,146], prostate cancer [147,148], colorectal cancer [149,150], gastric cancer [151,152], lung cancer [149,153], glioma [154], breast cancer [149], and cholangiocarcinoma [155]. Curcumin induces ROS by inhibiting the activity of various ROS-related metabolic enzymes, such as CAT, SOD1, glyoxalase 1, and NAD(P)H dehydrogenase [quinone] 1 [146,149]. ROS accumulation further mediates G1 or G2/M cell cycle arrest [146,147,150,154], senescence [146], and apoptosis. Many pathways have been implicated in ROS-mediated induction of apoptosis by curcumin, including downregulation of AKT phosphorylation [145], endoplasmic reticulum stress (namely through the PERK-ATF4-CHOP axis) [150,151,153], activation of the JNK pathway [151], and inhibition of STAT3 [155]. Curcumin has been studied extensively in in vivo cancer models [156] and has been investigated in Phase I and II for various cancers, including multiple myeloma, lung, breast, colorectal, and prostate cancers, with generally beneficial results ( Table 2) [23]. Presently, there are two ongoing clinical trials for curcumin for invasive breast (NCT03980509) and unresectable pancreatic (NCT02336087) cancers. Additionally, the combination of curcumin and piperine, a pro-oxidative phytochemical that drastically increases the bioavailability of curcumin in humans [99], is also being investigated in clinical trials (NCT02598726 [Phase 1] and NCT04731844 [Phase 2]). Pro-Oxidative Drugs in Clinical Use Many of the chemotherapeutic drugs that are currently used as standard-of-care treatment for cancers are known to induce ROS. However, the induction of ROS may not necessarily be the main mechanism of action, nor a direct consequence of the drug. For example, paclitaxel, which is used for the treatment of breast, endometrial, and ovarian cancers, exerts its anticancer effect by stabilizing tubulins. This prevents the disassembly of microtubules and consequently induces mitotic arrest, which in turn induces cell death [157]. ROS induction has been suggested as a secondary/alternative mechanism of action of paclitaxel and has been reported in osteosarcoma [158], prostate cancer [159], and non-smallcell lung cancer [160] cells. However, of these studies, only one reported that the use of the ROS scavenger N-acetylcysteine abrogated paclitaxel-mediated effects [159]. Moreover, it has only recently been clarified that mitochondrial accumulation, a consequence of paclitaxel-induced mitotic arrest, causes mitochondrial oxidative stress [161]. This lack of in-depth clarification is true for other approved and clinically used drugs with proposed pro-oxidative activities, such as rituximab, used for the treatment of B-cell lymphomas [10]. Hence, it is important to clarify whether ROS induction directly plays into the anticancer effects mediated by chemotherapeutic drugs. In this section, we have highlighted cisplatin and doxorubicin as examples of approved/ standard-of-care anticancer drugs that directly induce ROS, which further contributes to their anticancer activity. Other examples include motexafin gadolinium (an electron acceptor that increases superoxide production; used in the treatment of breast cancer and malignant melanoma) [162], arsenic trioxide (inhibits SOD and TrxR; used in the treatment of relapsing acute myeloid leukemia) [135], and imexon (disrupts GSH activity, causing depletion of GSH pool; used in the treatment of ovarian cancer, multiforme glioblastoma, and multiple myeloma) [135,162]. Cisplatin Cisplatin is a platinum-based chemotherapeutic drug that is widely used for the treatment of several cancers, including ovarian, bladder, and testicular cancers. The main mechanism of the anticancer activity of cisplatin can be attributed to DNA-platinum adduct formation, which induces p53-mediated cell cycle arrest and apoptosis [163]. However, oxidative stress independent of nuclear DNA damage has been implicated as one of the mechanisms underlying its cytotoxic effect [164]. Cisplatin induces generation of mitochondrial ROS, which compromise mitochondrial function and membrane potential [165], mitochondrial DNA integrity, and promote mitochondrial biogenesis [164,166], the latter of which further increases mitochondrial ROS levels. Consistently, mitochondrial content was found to correspond to cisplatin sensitivity in ovarian cancer; pharmacological increase in ROS sensitized cisplatin-resistant ovarian cancer cells to cisplatin-induced oxidative stress-mediated apoptosis [166]. Downstream of mitochondrial ROS, cisplatin induces ROSdependent autophagy and apoptosis through the JNK [167] and Bax/Bak pathways [166], respectively. In contrast to nuclear-DNA damage-mediated cytotoxic effects, cisplatin induces oxidative stress independent of p53 status in head and neck squamous cell carcinoma [168]. The clinical use of cisplatin is often complicated by chemoresistance, as well as associated toxicities, including nephrotoxicity and hepatotoxicity. However, the use of cisplatin in combination with other therapeutic strategies can overcome these challenges. Consistently, there are numerous ongoing clinical trials that are investigating cisplatin in combination with other therapeutic strategies (Table 3). Doxorubicin Doxorubicin (Adramycin) is a widely used chemotherapeutic agent for various solid (including breast, ovarian, gastric, and thyroid cancers) and hematological cancers (including multiple myeloma, and acute lymphoblastic/myeloblastic leukemia), either alone, in combination with other drugs (such as cyclophosphamide), or as nanoformulations (such as Doxil) [169]. Although topoisomerase II inhibition is the main mechanism of action, doxorubicin also induces ROS generation through redox cycling into its unstable semiquinone, which releases ROS on conversion back to doxorubicin [170]. Additionally, doxorubicin as an iron chelator forms complexes that catalyze the conversion of H 2 O 2 and superoxide into hydroxyl radicals or iron-peroxo complexes [169] and downregulates CAT and manganese superoxide dimutase activity [171,172]. Further, ROS induction by doxorubicin mediates p53-independent apoptosis in osteosarcoma cells [172]. Currently, various drug delivery approaches are being investigated to increase the bioavailability of doxorubicin as well as to decrease its side effects, namely cardiotoxicity. Consistently, liposomal and pegylated liposomal doxorubicin are in clinical use for breast cancer [173]. Additionally, pegylated liposomal doxorubicin and other drug delivery approaches for doxorubicin are being studied in clinical trials alone or in combination with other chemotherapeutic agents (Table 3). In this regard, doxorubicin-loaded nanoparticles have also shown promise in in vitro and in vivo studies [174][175][176]. Zheng et al. reported that doxorubicin loaded arginine-glycine-aspartic acid-modified solid lipid nanoparticles exhibited better cellular uptake in both breast cancer and normal cells and higher cytotoxicity against breast cancer cells than doxorubicin in vitro [176]. Additionally, they exhibited higher plasma concentra-tions and better biodistribution (less drug concentration in kidneys and hearts) compared to doxorubicin in vivo. Concluding Remarks The paradoxical role of ROS in cancer cells presents a unique therapeutic approach that has garnered much attention in recent years. As highlighted in this review, pro-oxidative anticancer drugs, such as cisplatin and doxorubicin, are already being used clinically to treat various cancers, and other pro-oxidative anticancer drugs, such as curcumin and its derivatives, carnosol, and 15d-PGJ2, are in various stages of research and development. The pro-oxidative drugs discussed in the present review induce ROS through diverse mechanisms (Figure 2), which further regulate various signaling pathways to induce anticancer effects ( Figure 3); this has been schematically summarized in Figures 2 and 3. As presented herein, targeting the skewed redox balance of cancer cells via prooxidative drugs, specifically pro-oxidative phytochemicals, presents a promising therapeutic approach; however, further research is needed on this front. One of the main aspects which seems to be overlooked is the characterization of mechanisms through which anticancer drugs induce ROS, as reflected with respect to pro-oxidative phytochemicals in Table 1. This is of specific importance as it can help direct the selection of pro-oxidative anticancer drugs by clinicians. For example, drugs that inhibit SOD1 can potentially be used for the treatment of cancers in which SOD1 is overexpressed, such as non-small cell lung cancer and breast cancer [177]. Additionally, although numerous prooxidative phytochemicals have been reported to exhibit promising anticancer activity in vitro, there is a lack of sufficient in vivo studies. Moreover, poor bioavailability, a limitation common to many phytochemicals [178], further complicates the matter and hinders their clinical application. Consistently, nanoparticle-based approaches are being investigated to overcome this limitation; however, further research is needed on this front to circumvent the inherent challenges created by this approach, which can limit clinical translation, such as target specificity and toxicity to normal cells [179]. The context-dependent role of phytochemical compounds also needs to be investigated further, as based on the concentration used, phytochemicals, namely resveratrol, are known to function as both antioxidants and pro-oxidants. Although antioxidants have been proposed as adjuvants to counteract the side effects of chemotherapy, the rationale for the same is debatable. While ROS drive cancer cell survival and progression, the disrupted redox balance is further augmented by pro- As presented herein, targeting the skewed redox balance of cancer cells via prooxidative drugs, specifically pro-oxidative phytochemicals, presents a promising therapeutic approach; however, further research is needed on this front. One of the main aspects which seems to be overlooked is the characterization of mechanisms through which anticancer drugs induce ROS, as reflected with respect to pro-oxidative phytochemicals in Table 1. This is of specific importance as it can help direct the selection of pro-oxidative anticancer drugs by clinicians. For example, drugs that inhibit SOD1 can potentially be used for the treatment of cancers in which SOD1 is overexpressed, such as non-small cell lung cancer and breast cancer [177]. Additionally, although numerous pro-oxidative phytochemicals have been reported to exhibit promising anticancer activity in vitro, there is a lack of sufficient in vivo studies. Moreover, poor bioavailability, a limitation common to many phytochemicals [178], further complicates the matter and hinders their clinical application. Consistently, nanoparticle-based approaches are being investigated to overcome this limitation; however, further research is needed on this front to circumvent the inherent challenges created by this approach, which can limit clinical translation, such as target specificity and toxicity to normal cells [179]. The context-dependent role of phytochemical compounds also needs to be investigated further, as based on the concentration used, phytochemicals, namely resveratrol, are known to function as both antioxidants and pro-oxidants. Although antioxidants have been proposed as adjuvants to counteract the side effects of chemotherapy, the rationale for the same is debatable. While ROS drive cancer cell survival and progression, the disrupted redox balance is further augmented by pro-oxidative anticancer drugs (Figure 1). Given that many of the clinically approved anticancer drugs induce ROS either directly or indirectly, the use of antioxidants for anticancer therapy as opposed to cancer prevention seems counterproductive and could potentially decrease the ROS-dependent anticancer activity of such drugs [3,180]. Hence, it is paramount that in the future, the concentration-and context-dependent roles of phytochemicals should be carefully investigated, with specific focus on the pro-oxidative effects. Moreover, investigation on this front would further facilitate the use of pro-oxidative phytochemicals as adjuvant drugs that can synergize with or sensitize resistant cells to clinically used pro-oxidative anticancer agents; for example, PPL in combination with platinum-based drugs, such as cisplatin [18] and oxaliplatin [19]. Further, it is important to clarify whether ROS induction plays directly into the anticancer effects of chemotherapeutic drugs and to elucidate the underlying mechanisms to exploit the above-mentioned synergistic effects.	2023-05-29T15:05:43.288Z	2023-05-26T00:00:00.000	{ "year": 2023, "sha1": "4a00ceb78d50f4f746b559ba40fcce702211eda0", "oa_license": "CCBY", "oa_url": "https://doi.org/10.3390/antiox12061159", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "f49e7b61f60823697db3814dcca99fa90bbbcec6", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Medicine" ] }
258291446	pes2o/s2orc	v3-fos-license	Repurposing Text-Generating AI into a Thought-Provoking Writing Tutor Text-generating AI technology has the potential to revolutionize writing education. However, current AI writing-support tools are limited to providing linear feedback to users. In this work, we demonstrate how text-generating AI can be repurposed into a thought-provoking writing tutor with the addition of recursive feedback mechanisms. Concretely, we developed a prototype AI writing-support tool called Scraft that asks Socratic questions to users and encourages critical thinking. To explore how Scraft can aid with writing education, we conducted a preliminary study with 15 students in a university writing class. Participants expressed that Scraft's recursive feedback is helpful for improving their writing skills. However, participants also noted that Scraft's feedback is sometimes factually incorrect and lacks context. We discuss the implications of our findings and future research directions. INTRODUCTION Before the internet, conducting information search for writing was an analog and resource-intensive process. However, with the widespread adoption of the internet, this process underwent a great change, giving rise to the information age and the democratization of knowledge. We believe that text-generating AI will lead to an even more radical transform to writing, by fundamentally transforming both the writing process and writing education. One's writing skills may soon depend on one's ability to use AI writing-support tools, analogous to how information search skills depend on one's ability to use the internet. While this may exhibit efficacy in writing productivity, this presents implications in writing education, as AI may soon contribute to a shift in core writing pedagogies such as critical thinking and self-reflection. As such, we contend that AI writing-support tools should focus on preserving writing's recursive and thought-provoking nature. THE IMPORTANCE OF RECURSIVE FEEDBACK IN WRITING EDUCATION Current AI writing-support tools are typically limited to providing linear feedback to users. They facilitate a writing process where text-generating AI directly provide outputs to user prompts, greatly minimizing opportunities for recursive and iterative user involvement. We are particularly concerned that excessive exposure to such tools may deprive users of the opportunity to develop their * Both authors contributed equally. writing skills, including skills that are often co-developed with writing such as critical thinking. This is because writing is an inherently recursive process, rather than a linear one. Thus, we contend it is critical that educational AI writing-support tools preserve the recursive nature of writing. For instance, the process of argument development and feedback reception is a key element of writing. To preserve this process, AI writing-support tools can utilize recursive feedback, e.g., continuous generation of Socratic questions, as opposed to linear feedback, e.g., direct answer generation. In this work, we focus on providing recursive feedback in the form of Socratic questions, however, recursive feedback refers to the general cyclical process in which a user's understanding serves as input for an iterative exchange that continues to refine the learning experience [4]. AI writing-support tools that are limited to linear feedback only warrant 'regenerations' or edits. However, tools with recursive feedback mechanisms can enable users to improve their thinking and writing organically as the feedback is tailored to users' individual writing experience. Throughout the writing process, users can request continuous and personalized feedback from the AI writing-support tool, i.e., AI can be a personal writing tutor. SCRAFT: THOUGHT-PROVOKING AI WRITING TUTOR To realize its idea, we developed a prototype AI writing-support tool called Scraft (Figure 1) that provides recursive feedback based on the Socratic questioning method. The impetus for exploring this approach lies in its capacity to cultivate critical thinking, which serves as a fundamental element of writing pedagogy. Socratic questioning fosters a disciplined mode of thought that actively monitors, evaluates, and restructures cognitive processes, emotions, and actions in a rational manner [3]. The current prototype is designed to generate elementary recursive feedback by continuously producing Socratic questions as users engage in the writing process. When users incorporate the provided Socratic questions into their work, Scraft adapts to this new input and generates an updated set of feedback. This functionality is integrated within a conventional writing document interface, featuring a sidebar for displaying the generated questions. Additionally, users have the option to utilize an "answer the question for me" button, which enables them to receive suggested responses to the Socratic questions at any given point during the writing process. PRELIMINARY STUDY To explore how Scraft can aid with writing education, we conducted a preliminary study with 15 students in a university writing class. Participants used Scraft for a two-week period to work on a class assignment. Afterwards, we conducted 15 minutes interviews with eight students. Below we summarize the participants' praises and complaints on the current prototype. Starting with praises, many participants said they experienced an improvement in their writing skills, and attributed it to Scraft's Socratic questions. These participants said that the questions encouraged them to critically think about aspects of their writing they had not previously considered. Scraft's recursive feedback pinpointed areas in the participants' writing that require further development. By doing so, it encouraged the participants to conduct more research on the writing topic and develop a deeper understanding. Participants also mentioned that Scraft's Socratic questions were useful for anticipating potential counterarguments to their claims and preemptively strengthen their writing. The main complaint on the current prototype was that the provided Socratic questions varied in their usefulness. Participants said some questions lacked relevance to the writing topic. In other instances, the provided feedback contracted one another or lacked factual accuracy. We find this to be consistent with the findings of Dikli and Blyele [1,2]: Compared to human feedback, automated essay feedback may often yield mixed outcomes as the effectiveness of such feedback may be undermined by factors such as incomprehensibility, overly general content that is challenging to implement, and overall inaccuracies in identifying essay sections with errors or requiring correction. CONCLUSION In summary, we demonstrated how text-generating AI can be repurposed into a thought-provoking writing tutor by providing recursive feedback to users in the form of Socratic questions. Our preliminary study with Scraft show the potential of this approach. However, it also revealed limitations: Socratic questions generated by current text-generating AI technology sometimes lack relevance to the writing topic and are factually incorrect. These findings underscore the need for further research on educational AI writingsupport tools to better support the recursive nature of the writing process and uphold pedagogical principles that foster critical thinking and self-reflection in student writers. We argue our proposed approach not only maintains but enhances the natural processes of writing, i.e. drafting, which in culmination, fosters the core pedagogical aspects of writing education.	2023-04-24T01:15:16.295Z	2023-04-09T00:00:00.000	{ "year": 2023, "sha1": "1ed876d2ed4db59aabbb2be90901315182a8fb6b", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "1ed876d2ed4db59aabbb2be90901315182a8fb6b", "s2fieldsofstudy": [ "Education", "Computer Science" ], "extfieldsofstudy": [ "Computer Science" ] }
246949241	pes2o/s2orc	v3-fos-license	Charming, influencing and seducing: a portrayal of everyday coaching ABSTRACT Since a critical turn was embarked on two decades ago, research into sports coaching has increased in quality and quantity [see Jones, R. L. (2019). Studies in sports coaching. Cambridge Scholars Publishing]. Despite such welcome advances, the essence or heart of the activity remains contested terrain [Abraham, A., & Collins, D. (2011). Taking the next step: Ways forward for coaching science. Quest (grand Rapids, Mich), 63(4), 366–385. https://doi.org/10.1080/00336297.2011.10483687; Jones, R. L., Edwards, C., & Tuim Viotto Filho, I. A. (2016). Activity theory, complexity and sports coaching: An epistemology for a discipline. Sport, Education and Society, 21(2), 200–216. https://doi.org/10.1080/13573322.2014.895713]. Subsequently, the aim of this work was to inductively analyse the practice of a top-level sports coach to better understand the core of what he actually did whilst working. This was particularly in terms how he managed the working contexts and the others within it towards desired ends. In seeking a ‘bottom up’ construction of practice, the study adopted tenets from both grounded theory and phenomenological inquiry. More specifically, the fieldwork was conducted over a 6-month period at a top-level women’s basketball club, with the data collection methods being ethnographic in nature, inclusive of participant observation and informal interviews. The main findings indicated that the coach in question, together with his coaching team, were engaged in a series of social, power-related, seductive strategies designed to charm athletes and others to ‘buy into’ the given agenda. Introduction Although research into sports coaching has increased in both quality and quantity since embarking on a critical turn some two decades ago, Jones (2019) recently claimed that the field has nonetheless suffered from 'theoretical defeat' (McCarthy, 1989). This was because, in conceding 'definitial rights' to other disciplines, coaching has not been able to cultivate its own thoughts or be interpreted from within its own frame of reference (Jones, 2019). Similarly, the tendency to borrow theoretical explanations from psychology, pedagogy and particularly sociology has given rise to inevitable accusations that resultant findings are more concept-centred than they really are (Liberman, 2013); that is, the theory has dominated the data. Hence, and perhaps unsurprisingly, paradigmatic disagreements have emerged in relation to what coaching is actually about (e.g. decision-making [Abraham & Collins, 2011], empowering athletes [Duda, 2013a[Duda, , 2013b, dyadic relationships [Davis & Jowett, 2013], and interactional exchanges [Hemmestad & Jones, 2019]). This lack of consensus has manifest itself in debates ranging from the nature of coaching knowledge to how coaching should be researched and represented (Jones et al., 2016). Furthermore, despite the almost obligatory 'complex aware' rhetoric, with a few exceptions (see Jones, 2019 for a summary), most of the above-cited work continues to be starved of contextual considerations, thus 'being somewhat hollow in terms of appreciating how coaching actually plays out as situated action' (Jones et al., 2016, p. 201). Consequently, the present study marked an attempt to uncover what is actually going on when coaching takes place. Through giving primacy to naturalistic and inductive research methods, the study's objectives related to exploring how coaches manage, maneuver and influence context and others towards desired ends, particularly in terms of the performative exchanges evident. The purpose then, was to further understand the arrangements and relationships experienced between people in coaching, thus somewhat revealing what the resultant collective web of negotiation fundamentally entails (Jones, 2019). The significance of the work lies in examining the often unnoticed surreptitious everyday intentions and interactions which, nevertheless, remain central to the 'doing' of coaching. The work thus builds upon that of Armour (2015, 2017) and Jones's (2009) inductive conceptualisations of coaching to further unmask the precise intentionality of the activity within cultural constraints. Hence, it goes further than claims related to the perceived 'activity' of coaching, to an exploration of how that activity is realised or secured. In doing so, the paper also progresses Jones and Ronglan's (2018) tentative theorising about the 'quiddity' (Garfinkel et al., 1981) or 'just whatness' of coaching, thus bringing the activity back to itself. In effect, the study marks an effort to better grasp the practical accomplishments of actors in context. Although such accomplishments or actions have been variously described in previous work (as mentioned above), what remains largely missing is an exploration of the 'consciousness' of coaching; a consciousness possessing an intentionality (Zahavi, 2019). This is not in respect of having a purpose in mind when one acts, but rather an 'aboutness' or 'directedness' to the awareness; i.e. that all action (in this case coaching) is about something (Luft & Overgaard, 2012). Consciousness thus, is not considered as preoccupied in and of itself, but is always related to something; it is outward-facing and directed at an object other than itself (Zahavi, 2019). Similarly, any action or object is made conscious in a particular way, in terms of how it is perceived by an observer. In this respect, as stated, coaching has been considered to be about many things (e.g. technical knowledge, pedagogy, decision making, etc.) depending on who perceives or identifies it. No doubt, therefore, a need exists to better examine and describe this 'aboutness of coaching consciousness', thus better-placing coaches, to use a phrase from the phenomenologist Heidegger, as 'being in the world'. Finally, in helping coaches better understand the meaning of their work through its experiental components, the study possesses the potential to offer relevant support for practitioners' organisational lives. It can do so through providing more appropriate and realistic professional development provision for this intensely demanding profession (Jones & Wallace, 2005, 2006; provision founded on the idenified rules and tacit resources that underlie the 'doing' of coaching. This is particularly in relation to how coaches can better live, create, and relate in the salty, satisfying, frustrating worlds they inhabit (Moustakas, 1994). The ultimate rationale for the work then, lies not just in passively explaining coaching, but in an active pursuit of better reconstituting it. Methodology The setting where the fieldwork was undertaken comprised an adult women's basketball club, 'The Waverley Hoops' (a pseudonym). The club was located in a large provincial city and was primarily chosen due to its 'first' team, of which it principally comprised, competing at the highest national level. The club, however, also comprised two other teams, which took part in further additional competitions. The weekly rountine for the players (n = 15) and staff (n = 5) involved two team training sessions, two 'individual' training sessions, two strength and conditioning sessions, and one performance analysis meeting, in addition to regular weekend league games. The squad, from which the team was selected, and all of its activities was overseen by the head coach, Gareth, who, in turn, was supported by assistants Rhys and Henoch (all names are pseudonyms). In more detail, Gareth had been Waverley Hoops' head coach for 7 years, having earlier served as the Assistant. At the time of the research, he also had a role with the national senior squad, and a national age-group team. He was 30 years old and frequently referred to the club as 'the basketball family'. Rhys was Gareth's principal assistant and had been working with him in various capacities for some years. He also served as head coach for a Waverley team who competed in a secondary, separate competition. Rhys was 25 years old. Finally, Henoch was Gareth's 'second' assistant coach while also being the head coach of the Division 2 team. At 28, Henoch had only recently joined the club having previously coached abroad for a number of years. Borrowing traits or features from both grounded theory and phenomenology, inclusive of a search for intentionality or 'aboutness', of not being concept led, and of general discovery, inductive means were used to collect the data. The intention was to allow descriptive data wherever possible, as opposed to existing conceptualisations, to drive the exploratory process. Phenomenology sets an anti-reductionist approach to how people experience life, focusing on individual and situated experiences to uncover the essence of a phenomenon (Kerry & Armour, 2000). Such philosophical assumptions have been described as an attitude; a way of looking and interpreting the world in order to uncover 'things as they are' (Allen-Collinson, 2009). The purpose, as previously stated, was to allow for an exploration of the everyday, mundane events which comprised the coaching lifeworld under study (Cronin & Armour, 2015). Having said that, there is no claim to a pure phenomenological approach being utilised here. This is because the data collection was no doubt driven by the study's given aims as opposed to unhindered open ended inquiry. Hence, although a combination of phenomenological and grounded theory-related principles were employed in a predominantly inductive design which, in turn, enabled the collection of original and meaningful data, this was done in relation to the study's objectives. In practice, this was manifest through no theories on how the coaches managed the context being considered a priori, allowing the observed actions, reactions, emotions and perceptions to be the focus of the collection. The attempt then, through considered reflexive activity, was to 'set aside all preconceptions' and 'take nothing for granted' (Androsino, 2007, p. 38), albeit in line with a given intentionality. Methods Participant observation and informal (or conversational) interviews were utilised to collect the data. Here, I, as the first author and principal researcher, not only took part in the various weekly sessions and competition preparation but also participated in (principally) Gareth's daily office routine and coaching-related meetings. This included, for example, joining the coaches in their meetings with physiotherapists and doctors, in game analysis sessions, as well as taking part in athletes' warm-ups, preparing the game venue, fetching balls during practice, and generally sitting on the bench with injured players during training sessions. To further my integration, I also sold merchandising to help raise money for the club and distributed advertising material as appropriate. Doing so, provided opportunities to observe, listen, ask questions and take part in contextual interactions. In this respect, I also followed the recognised principles of qualitiative research as related to; openness, distance and skepticism (Rabb, 2019). Hence, I was constantly engaged in a reflexive balancing act between intense closeness and a 'simultaneous maintenance of a controlled distance [from] the social situation being investigated' (Rabb, 2019, p. 39). Of particular importance, however, and following Goffman (1989), I subjected myself the best I could to the set of contingencies experienced by the actors under study ('to take the same crap they were taking' [Goffman, 1989, p. 125f]); to emotionally act as if I could not leave at any time (which, of course, I could). In addition to observational fieldwork, which yielded rich and comprehensive data, conversational interviews (Ennis & Chen, 2012) were conducted to further complement and/or explain the observed events. The purpose here was to seek insights into the observed and perceived complexities inherent within the coaching under study. Consequently, as opposed to structured or even semi-structured interviews, no particular set of questions was arranged or categorised a priori. Rather, following Patton (2002), such conversations were treated as natural extensions of the observations in terms of clarifying or further exploring interpretations. This, or course, is not to say that such interactions proceeded without detailed knowledge and preparation, with the study's purpose and the general scope remaining at the forefront of consideration (Fife, 2005). These conversational interviews were repeatedly used throughout the fieldwork period, consisting mainly of relaxed conversations with the participants in a variety of situations; for example, before, during and after training sessions and competitions, in the coaches' offices, in parking lots, coffee shops and hotel hallways. The fieldwork experiences were recorded using written notes and a dictaphone. Here, verbal exchanges were transcribed verbatim as much, and as soon, as possible, and always within a few days of their occurrence. The information obtained from the interviews served to sharpen the data already collected through the participant observations (and vice-versa), whilst adding extra dimensions through new or contradictory insights. Due to the largely interpretative and naturalistic nature of this work, I continuously resorted to reflexivity and the use of critical friends, including my co-author, to prompt reflection, challenge pre-assumptions, and consider alternative interpretations. This involved a constant critical analysis and questioning of 'to what extent were the data obtained a result of, or were influenced by, experience, knowledge, expectations?' 'Am I sure I can label what I saw in the way I did?' 'Am I being too concept led'?''Is there anything else in the context I'm missing?' This process allowed an acknowledgment and management of personal leanings and an increased critical consideration of the information collected. In turn, the data, as related to observations or informal interviews, came to be considered as social encounters in and of themselves. Here, specific attention was paid to identifying key events, characters, and moments, and the interaction that occurred between them, which appeared crucial to the objective(s) of the work. In this respect, the reflective process undertaken included judicious discussions amongst ourselves as authors, loosely framed by the 'just whatness' of the phenomenon and associated events being observed (Jones & Ronglan, 2018). This critical reflexivity also moved beyond traditional ethical positions and considerations to the aspirational relational approach advocated by Sparkes and Smith (2014). Subsequently, in addition to requesting and gaining approval from the university's ethics committee before the fieldwork began, we also sought to emphasise elements of care and connectedness throughout the project. Such judicious consideration was not only extended to ourselves and those under study, and to between ourselves as researchers, but also to an ethical obligation to the work itself (Sparkes & Smith, 2014). In this respect, care was taken to critically examine and better understand how our personal histories and biographies influenced the data gathered and the interpretations made. No doubt, our individual experiences as both players and coaches (myself in volleyball and Robyn in football), in addition to our collective identities as scholars working at the same institution, encouraged us to 'see' some things at the expenses of others. This is where the reflexive process outlined above moved from introspection to that of intersubjective reflection and mutual collaboration (Findlay, 2002). The process engaged in then marked an effort to go beyond personal revelation, a kind of 'benign introspection' (Woolgar, 1988, p. 22), to a critical discussion of the inference contended between experience, the observed social context, and the knowledge claimed (as outlined in the paragraph above). Theorising the data, and presenting the results In giving primacy to the data and aligned with the phenomenological and grounded theory principles already mentioned, subsequent theorising was engaged with in a 'light touch' fashion. The objective here was to adhere to the inductive analysis undertaken and thus, use professional concepts only to assist in the explanation and understanding of the observed activities. Although aware of the need to make sense of the findings, every effort was consequently made not to allow existing concepts to dominate interpretations; that is, not to make the undertaken observations more concept focussed or driven than they really were (Liberman, 2013). This is not to say that arrived-at theoretical impressions did not guide the fieldwork, particularly in the later stages. Rather, that a process of both 'theoretical discovery' and 'theoretical refinement' was engaged in (Puddephatt et al., 2009, p. 15). Consequently, the novelty of the practice witnessed at Waverley came into being through the reflective iterative writing and the conceptual questioning process. In this respect, constant consideration was given to identifying the ultimate narrative themes selected in association with the primary sensitising concepts, which were loosely applied as the work unfolded. Examples of such concepts included 'power', 'charisma' and 'attraction'. The process of data analysis and collection then, ran concurrently to an increasing degree as the fieldwork progressed. Finally, a closing stage of analysis, once collection had ceased, was undertaken where a concluding 'theoretical order' was again subject to critical consideration based on the previously developed ideas. In more detail, the initial stage of data 'collection' resembled a process of 'free' memo-writing about first impressions and involved staying as close as possible to descriptive phenomenological principles. Here, I allowed myself to be guided by intuitive reflections in relation to the coaches' (principally Gareth's) presence or bearing, the influence they were having over the context, and the fundamental 'doings' of the coaching that took place. This was followed by a more structured second coding procedure, where increasingly concrete pieces of data were examined in relation to their convergence or divergence from the initially developed raw notions. Although this second stage was intended to develop some thematic substance, it also allowed an exploration of what had not been included so far. Here, following Charmaz (2006), open codes were used, which entailed constructing and applying 'labels' that further described the actions noticed in increasing detail. Rather than moving in a rather linear progression to a focused coding phase, I thirdly performed something closer to an additional phase of sense-making, where the codes developed were both associated with others and matched. The value of this stage lay in reviewing, reflecting upon, and further annotating the created codes, in order to both consolidate then, while allowing consideration of any events or ideas which had not been included so far. Consequently, a 'bigger picture' of how coaching worked at Waverley was slowly assembled; a picture which came to comprise the general topics of 'caring'; 'putting on a performance'; and 'coaches' control'. Taking care not to overly allow the concepts to lead the sense-making of the data, this process involved further searching, observing and taking notes in terms of the (above) evolving categories whilst also revisiting previous notes to ascertain the credibility of earlier interpretations. In addition, reviewing the study's objectives within this process offered some reassurance of saturation and a degree of confidence to 'move' within the analytical process. The decision to present the results of the study through a series of short vignettes was informed by a number of reasons. Principal among these was the ability of such text to both help create and communicate 'meaning in context'. In short, and following others (e.g. Gilbourne et al., 2014), we considered vignettes as better allowing situational context to be explored and influential features to be elucidated than other means of representation. In particular, vignettes were chosen due to their ability to elicit the intended meaning made of the data. In this respect, as opposed to merely describing the events observed, we wanted to portray the emotionality, the feeling, and the evident 'care-ful' actions, that took place within the setting (Ely et al., 1997). Through the use of vignettes then, an effort was made to draw readers into the experience, into the sporting setting at Waverley, and into the coaching practice. The obligation to make such tales 'credible', 'confirmable' and 'trustworthy', taken as criteria of 'good qualitative research' (Shenton, 2004), helped us to create the portrayal(s) as something akin to what was experienced. Hence, means such as plot, character development, theme, and language were all employed to render the stories both complex, relatable and significant (Ely et al., 1997). In this way, we tried to make comprehensible some often elusive notions. Acknowledging the constructed nature of vignettes forced further engagement with the critical reflexivity previously mentioned. This is because vignettes necessitate an interpretation of witnessed events and felt experiences; an interpretation that restructures complex phenomena in order to highlight whan can be learned, through a particular portrayal of them. As opposed to direct descriptions then, the presented vignettes represent composites that encapsulate what was discovered through the fieldwork undertaken. In this respect, they 'sandwich together people, places or events to reveal implicitly the signifiance of the [broader] story told' (Ely et al., 1997, p. 72). Their precise construction involved the inductive data analysis alluded to, through various iterations brought about by critical discussion, to what we considered to be the principal considerations. In this respect, the vignettes represent examples of the main interpretive themes inferred from the data and their collection process. Consequently, although certainly inspired and constructed around events as witnessed, in line with the aforementioned process engaged in, the vignettes were fashioned to highlight what came to stand out. The following 'results' section marks an attempt to capture what we considered important from the fieldwork, and is divided into four segments, each centred on a particular vignette. The segments include 'We're all in it together', 'A benevolent 'Head of family'', 'Caring for Audrey', and 'An enjoyable, meaningful climate'. Results: coaching as charming We're all in it together During the fieldwork, I initially observed the coaches 'acting like coaches' and heard familiar expressions such as 'sometimes you have to be nice, other times you just have to be tough … ' (field note no.37). However, as the fieldwork progressed, how the coaches (and in particular Gareth) acted and presented themselves to the players and others soon revealed itself as complex action far beyond professionally performing as expected. Instead, the coaches seemed consumed with presenting a persuasive, charming 'front'; that is, an image of self-designed to induce others, usually the players, to 'buy in' into the presented ideas and agendas. A part of this image or manner included that of 'being the hardest worker in the room', a positive role model to follow; that the coaches would not ask anything of the athletes that they weren't prepared to do themselves. Here, portrayals of willing sacrifice for the team and club came to the fore; with the work always being done, or portrayed as being done, in the best interests of the group (a notion that was also read as the 'best interests of the athletes'). Presenting such positive and caring features formed a major part of the attractive selfless, persona(s) constructed. Vignette 1 Everyone's waiting outside the sports venue door, as usual. Gareth and some of the players are sitting on the stairs facing the rest of the team. One of girls was excitedly sharing what had happened the previous night, but the general mood was subdued. I comment 'You guys are quiet today … '. Gareth laughs and says to all … 'We spend so much time together that we don't have anything new to talk about'. The players smile in return. He claps cheerfully getting up: 'Let's go!' I look at the venue's walls, a banner features a picture of the team gathered in a circle, hands on shoulders. It reads 'coming together is a beginning; staying together is progress, and working together is success'. Next to it, another banner with pictures of players passing and shooting, includes the statement 'success does not come to you … you must go and get it. Perseverance!' The atmosphere is one of excited enjoyment. The girls have filled the water bottles, the bibs are laid neatly on the bench; the session begins. Gareth takes the lead. The clear talk is of the need to work ever harder; 'we have to fight for it … we need to keep pushing, together!' Although directive, the messages are delivered with caring enthusiasm and sincerity. The players listen intently; Gareth's eys and words seem to hold them. On the back of one of the players' t-shirt are the names of all players and coaches; underneath is written 'ONE FAMILY'. Gareth finishes, the team 'shout', they start the warm up. Some of the players have been given the 'day off' to facilitate minor injury recovery. They're sitting on the long bench, next to me. Cara glances at my notebook; 'you must be very busy if you're shadowing Gareth'. It seems common knowledge how hard he works; still, she deems it necessary to share the admiring sentiment. He manages everything related to the team, from session planning, to choosing the video clips to analyse, what the website should look like, booking physiotherapy sessions for the injured, and managing the house bills of semi-professional players. Cara shares a look with Anne, who nods in deferential agreement. 'Do you know what time he comes in every day? And he only leaves after our sessions. I really don't know how he does it all!' she concludes. She stares at the action, then looks back at me and smiles 'He's done so much for me, you know, personally. He's the best coach I've ever had'. A benevolent 'Head of family' At the start of the season play-offs, Gareth gave every player a card. On the right hand-side, it read; 'The secret to success is in the constancy of purpose'. On the left, the message was personalised: 'Sofia, I hope you enjoy the final 4's. It has been a pleasure having you around the team and the family. I hope you have enjoyed it as much as I have … and that you've found it useful for your research!' (Fieldwork note; 16.4) The Oxford English Dictionary comprises a definition of 'charm' as: 'any quality, attribute, trait, feature, etc., which exerts a fascinating or attractive influence' (The Oxford English Dictionary, 2015), 'to act upon so as to influence, control, subdue, bind, bewitch, enchant etc.' From the data gathered, the coaches were similarly seen to construct, 'charming' personalities to convince others, principally the athletes, to engage with their wishes and desires. Gareth's positioning of the group as a 'family' was clearly meant as a message associated with close relational 'care'. The metaphor, in this respect, provided a cultural script for everyone associated with the Hoops to believe that they could depend on each other (and particularly the coaches) not only for support and understanding but also for intimacy, empathy, obligation, and compassion. It was a means to tie the group closer together. In doing so, Gareth (and to a lesser extent the other coaches) 'charmed' the athletes in terms of the latter's well-being and development being central to the practice carried out. This was regularly communicated through the coaches' dayto-day actions while constantly expressed through sayings such as 'we'll do it together'; 'you have a say … '; 'we're a family'. Such a discourse prompted in players a sense of security allowing the coaches to be perceived as trustworthy in terms of making good decisions on their behalf (i.e. the athletes' behalf). In doing so, the coaches ensured that the players 'bought into' the given agenda. Vignette 2 The players are warming up, some in pairs others individually. Gareth announces himself to all with a loud 'HI'. He sits beside me on the bench. The players carry on with the warm-up and start some drills in their own time. Some practice dribbling, others shooting (from a variety of distances). It's cold, but the players don't seem to feel it; they're working hard. I ask Gareth what motivates the players to work on their own. He replies: 'It's the club culture … Cara has been with me for 4 years now'. He starts pointing; 'she's been with me for 5'; '2 years'; 'hum.. 2 again'; 'and those two, it's their first year'. 'It might also be because they see me working hard too?' He pauses, reflects, then says: 'I guess at first, we established what the culture was, and that I don't accept passive training. But it has to do with me too. Time is my thing; you have to use it well. I question them a lot about it, because you can't come here to lose time. I mean, this session won't happen again, so everyone needs to always ask 'what did I take from today's session?' What you do with your time is very important. So, yes, these players are now in a good place where they work hard without me being on the top of them. If they were being passive though, I would be all over them!' He stands up and starts to walk around the hall, observing the players' performances. I re-visit my field notes from last week. Gareth comes back and resumes our chat: 'But don't get me wrong, sometimes there are bad days. I like players to have some ownership of their work. Today they've asked me if they could do these drills, and that wasn't my plan, but I rather have them doing something they're motivated for it, as long as it's relevant. It also shows them that I trust them, and I get a lot back from that'. 'A lot back?' I ask. 'I get more engagement, more self-motivated action, more 'buy in' if you like. That wouldn't happen if I just obviously controlled everything'. Caring for Audrey Vignette 3 Audrey moved to Waverley from overseas, essentially to for the Hoops. She was offered a room in the players' shared house and a contribution towards her living expenses. Things, however, were not working out as planned. Gareth explains; 'For a variety of reasons, one of which was the language barrier, she was struggling. She was not happy, so she came to talk to me; she wanted to leave'. I ask how he reacted, as this decision affected the team. He said it was hard because 'she came here to play with certain objectives; and we accommodated and made some sacrifices for her'. Audrey left the programme soon after that meeting. Gareth explained his thought process in the chain of events; 'I imagine in a few years' time, I'll meet her in the street, with her kids, her family. And we'll recall the time she spent here with joy. She'll remember it as a good experience: she made new friends, learnt English and improved her basketball. I want this experience to be something that impacts her life in a positive way.' Similarly, Audrey described Gareth as the best coach she ever had demonstrating a sincere will to comply with what he asked of her and the team in general. Consequently, in the grander scheme of things, the decisions made by the coaches to seemingly care in this regard resulted in the club having happier, more committed athletes 'in its service'; a situation which was considered to produce better performances. Additionally, such displays of individual athlete care did not go unnoticed by the rest of the group. This prompted or created a certain impression of the coaches as always having the best interests of athletes at heart and, as a consequence, certain feelings of reverence towards them. Therefore, not only were unexpected personal problems used to benefit performances, but also allowed the coaches' increased respect from, and consequently more influence over, the athletes. In this sense, the coaches' behaviours promoted augmented loyalty from athletes and a willingness to work harder with and for them. An enjoyable, meaningful climate Vignette 4 Today's session starts with the usual team shouteveryone, coaches and players, in a circle and hands together in the middleand then goes on to a relaxed warm up. The drill consists of a relay between two teams, with players starting by running around a basketball in a very tight circle. The point is to make them dizzy before other tasks. Players shout to encourage teammates and laugh as they run in zig-zags and shoot disorientated. They move on to play 3 × 3 on half court, with two games going on simultaneously. Gareth moves from one court to the other. Soon he observes one of the games more closely. He refers to the team in yellow bibs as 'the bananas' and the other team as the 'lefties'. He celebrates successful actions by hi-fiving players. He's constantly commentating on the game: 'Uuuhh nice steal!'; 'Oh … the lefties are kicking your butt!' In between drills, the three coaches talk and laugh. They go around chatting with the players, generally smiling. Gareth gathers all the players around and explains the next drill. He also provides some individual feedback: 'Rebecca … ' he starts, but the players start laughing 'Rebecca?' one asks amused. He laughs with them and looking at Becky he says: 'When you're being naughty, it's Rebecca!' He continues to move along the sideline as the game re-starts, frequently shouting feedback and praise. Rhys is writing notes leaning on the wall, while Henoch is squatting next to him observing the action; they both occasionally shout some words of encouragement. Gareth becomes gradually unsatisfied with the performance. He's pacing back and forth and sighing discretely. He eventually shouts 'Stop, stop!', the players move closer. 'Ok guys, I get it. You want to be quick. That's good! But you need to keep the game flow in mind. And you've just done 5 turnovers in one minute. Please let's think about what we're doing', he pleads. The game starts, Gareth marches around the court and, as he walks past me, he tilts his head in my direction and whispers 'Fuck me. Thank God it's not an actual game'. Rhys and Henoch glance at him and increase the encouragement, shouting and clapping to bzoost players' motivation. 'OK, OK; keep going, you're working hard!' As one of the players runs past Rhys, she swiftly looks at him clearly frustrated. He smiles back; 'just keep driving.' She nods in acknowledgement. Gareth yells more feedback and encouragement: 'Good job!', 'Good choice!' One of the players stretches out to the ball and manages to steal it. Gareth and Rhys both whoop at the same time, look and each other, and laugh. The players in close proximity notice it and smile. Then Cara, the captain, scores from long range. 'That's double points for you' Gareth shouts and adds 'And she knows it!' She looks back at him and smiles. At the end, everyone gathers around again to do the team shout. Gareth says: 'We've been dreaming about this. But don't just dream about the final … think about what you have to do to win the final'. After this session, walking back with Gareth, I note 'Nice session, that was good fun'. He laughs scornfully, rolling his eyes. 'Well, there's no point in putting too much pressure on them now. The upcoming finals already do that for me'. Discussion In positioning power as omni-present, the French sociologist Touraine (1981) declared that 'all social relations are relations of power' (p. 33). In taking issue with earlier conceptualisations of power as a zero-sum game, where one individual or group possesses it while the other does not, power and the social were considered indivisible. Power thus was deemed as being constitutive of, as opposed to separated from, social order; one simply cannot exist without the other (Westwood, 2002). It is a view advocated by a generation of critical thinkers from Habermas, Gramsci, Giddens, and (of course) Foucault, as part of sociology's post-structural project. Subsequently, emancipation, in whatever realm, is considered an inherently flawed concept, with the best we can work towards being a 'rationally defensible form of authority' (Cassell, 1993, p. 228). Drawing on Nietzsche, Foucault's 'take' on power, in general opposition to those before him, was not as something negative or positive but as ever-present and potentially productive. For Foucault then, power was not considered as emanating from a cult of the autonomous self, but that social identities and productions were lived into existence through the constructive influence of culture (May, 2005). Although frequently accused of anti-humanism, perhaps due to his writings on the power of institutional scripts, Foucault alternatively saw great value in examining how those with responsibility (e.g. coaches) used their power to form productive relationships with those around them (Denison & Scott-Thomas, 2011;Markula & Martin, 2007). Seen as such, power was not considered as a destructive, coercive force, but rather generative of social relations. Following this seemingly 'progressive' lead, Westwood (2002) argued for a greater understanding of power through an examination of its 'modalities', defined as 'the property of power as it is exercised' (p. 25). One such modality was claimed to be that of 'seduction', taken as the most cogent expression of performative power (Westwood, 2002). It is a modality that not only emphasises the relational base of power but also how the social comprises performative selves enacted through certain behaviours and discourse. According to Bauman (1992), the power alluded to here does not have to coerce subjects but rather seduces through the representations and signification of 'innumerable dreams'. Such dreams are, in turn, considered attainable or purchasable by those subject to the seduction. Desires are thus, recycled and often reinvented as new, albeit remain as constructions around which respective performances are organised. Indeed, according to Westwood (2002), seductions are the classic enactment of power, 'feeding desire and individual power plays, massaging egos and, generally, generating a sense of well-being, worth and the feel-good-factor' (p. 83). Through such power plays then, as often manifest in careful and sensitive actions, affective ties are further secured. Similarly, the coaches in the current study were observed to create and embody particular (friendly, approachable and affable) personas to convince others of their credibility and, hence, solicit desired responses. The resulting personas selected, according to the messages they wished to convey, could be characterised as 'charming', a potentially useful and novel concept when considering coaching. Here then, the coaches adopted compelling knowledgeable, forthcoming and caring characterisations which exerted a persuasive influence over the athletes that subsequently 'drew the latter in'. The related messages were not only spoken and performed but also embodied by the coaches, who actively created the witnessed attractive, seductive, charming 'personas'. Similarly, the massaging of egos and the generation of a sense of supported well-being were very identifiable features of the created context (in line with Westwood, 2002). Such enactments were witnessed in Gareth's positioning as the 'head of the family', his obvious hard work and evident dedication to the team and club. Although Goffman's writings have traditionally suffered from the criticism of under-theorizing power and its workings, a growing number of scholars have disputed such claims (e.g. Dennis & Martin, 2005;Jones & Davey, 2011). The alternative case made is that the microfocus adopted by Goffman clearly demonstrates a fundamental concern with power phenomena. This is particularly so with regards to the social processes through which power is enacted. Viewed as such, Goffman's work is to do with examining how the rules of collective engagement are established, enforced, challenged and broken (Dennis & Martin, 2005). Indeed, Goffman's own conceptualisation of his work was as an 'analysis of the social arrangements enjoyed by those with institutional authority ' (1983, p. 17), with the self-positioned as a 'performed character' (Goffman, 1969). In this sense, the power held by people was considered as performative. Far from being some kind of underhand scheming, however, such performances were considered benign fabrications (Goffman, 1974); that is, they involved ideas impressed on others in their perceived best interests. The coaches' 'performances' at Waverley were additionally not exclusively dependent upon intentions. Rather, they were limited by the boundaries of social structure in addition to others' validation and responses (Goffman, 1959). Indeed, being so authenticated, the coaches' actions in this respect were co-constructed by contextual cultures and the stakeholders who comprised them. In turn, the coaches' 'fronts' and 'personas', as manifest by and through Gareth, influenced or reproduced the club's values and cultures (e.g. approachability, hard work, commitment), thus further consolidating personal power. Indeed, this is what is meant when it is said that 'power is relational' (Winter, 1996); that is, power as a function of 'reciprocally enacted roles, institutions and understandings' (p. 742). Again, this moves the analysis away from power as a seemingly 'common sense' game of strateges and tactics; as something that can be 'caught, held, bowled or just lost, as in the game of cricket' (Westwood, 2002, p. 135). Although the coaches' charming performances were no doubt convincing, this is not to say that the athletes were somehow power-less in context. They were certainly not cultural dupes unquestioningly blinded by dazzling charismatic leads. Such an interpretation would take us back to where we started in relation to the 'all or nothing' zero-sum power game. The athletes here then were certainly not without power. However, far from resisting the seductive power they were subject too, as those in less powerful positions are often prone to do, they appeared compliant in its use. Although such seemingly receptive behaviour could be understood in terms of Bourdieu's notion of the complicity of the dominated (Bourdieu & Wacquant, 1992), the 'hard edged' symbolic violence grounded in a collective deception associated with such complicity, was absent. Rather, what was evident at Waverley was the athletes as willing subjects in their own seduction, with players being very aware of the 'being a family' narrative (with all its positive connotations) and Gareth's position at the head of it. Consequently, similar to those whose awareness of advertising as an idealistic as opposed to a realistic representation of material life still doesn't hinder a purchase, the players continued to find the interests that sought to compel and influence them (i.e. the coaches and their actions) attractive and appealing (Winter, 1996). Hence, the pressure and attention exerted upon the athletes were generally welcomed, as they became consciously seduced subjects' in consuming the coaching they were subject to (Westwood, 2002). In this respect, even among audiences who can decode others' performances, the seduction is enjoyed. Two further sense-making concepts worthy of mention here, to develop a more complete picture of the context under study, include Weber's (1968) charismatic authority and Blau's (1986) social exchange. Weber (1968) conceptualised charismatic authority, not in terms of certain personality leadership traits, but as being grounded in the structure of the ensuing social relationships. As opposed to the unilateral imposition of a leader's will then, what was crucial here in terms of establishing charisma's validity was its recognition and acceptance by those subject to it. Such authority, although potentially transformational and far-reaching, was considered a relational phenomenon, constructed in the social interactions between leaders and followers; in this case, between the coaches and athletes. Both coaches and players were thus considered active in the creation of the charismatic bond, with each gaining something from the association. In this respect, the created attachment was founded 'on an exchange of mutual needs, where the charismatic leader is granted authority by the followers in return for recognition, affection, and reinforcement of worth' (Hofmann & Dawson, 2014, p. 351). In developing such a notion, Peter Blau's work on exchange and power in social life (1986) explained that individuals engage in interaction in search of something and, in doing so, 'each gains something while always paying a price' (Blau, 1955, p. 108). People then choose who to associate and interact with in the expectation of (social) rewards. Individuals, therefore, need to not only recognise attractiveness (i.e. potential profits) in others, but to also demonstrate attractive qualities which will appeal to those others (Blau, 1986); something evident in the studied coaches' charming personas. According to Blau (1986), more often than not, individuals act upon the anticipation of rewards on the basis of incomplete information; that is, the profits which may be collected are not easily predictable. Moreover, individuals act in the expectation that whatever the hoped outcome of the interaction might be, it will, at least, generate a sense of gratitude and/or obligation for returning the 'investment' (Blau, 1986). Relatedly, the coaches at Waverley relied on the creation of an underlying sense of willing obligation to obtain compliance from the players. Here, as opposed to stimulating a given reaction to a specific deed, the coaches made use of the sentiments generated about them: that is, from their ongoing demonstrations of friendliness, supportiveness and interest, thereby demonstrating to the athletes that they could or perhaps should be trusted. Naturally, as discussed above, being conscious of such seductive actions, the athletes subsequently 'bought into' the scenario that engagement here was a mutually beneficial endeavour. Conclusion An in-depth analysis of the coaches' everyday actions portrayed them in a constant search for control over their working conditions and the engagement of athletes. The coaching witnessed was connected to manipulative behaviours and selections of 'fronts'. However, rather than being reflective of Machiavellian intentions, these behaviours were seen as part of both tactical and tactful strategies, employed in the athletes' best interests. Indeed, they were the result of an often unwitting, implicit understanding that the interactions were, although part of a beneficial exchange, an attempt to pool the coaches' power; that is, the expressions of care and affirmation evident were aimed at generating further power and influence for the carers (i.e. the coaches). Hence, the coaches' everyday activity featured the creation of 'charming' personas to convince others (principally the athletes) that a joint association and continual investment would be beneficial for all concerned. According to Blau (1986), individuals act towards others' best interests with the expectations of a degree of gratitude and even obligation in return. It was through balancing investments and benefits that the coaches regulated their acts, so as to maintain and negotiate their fluctuating levels of control over a relational and complex process. Such an account of coaching as a covert process for control naturally takes issue with much of the current functional literature related to coaching (e.g. Duda, 2013aDuda, , 2013b. In rejecting such potrayals we further open up possibilities to critically explore the essence or nature of coaches' work and how they go about it. A part of this process is to reclaim such concepts as seduction, charisma and power, not as ones that ought to be avoided or mythicised but embraced and better understood in relation to the quest for improvements that coaches are invariable and constantly involved in. Not unsurprisingly perhaps, we consider that such an understanding can also serve to improve the coach education fare on offer. It can principally do so through encouraging coaches to reflect critically on how and why they act in the ways they do when coaching; that is, what their actions are orientated towards, and why they think such action is beneficial to what they are trying to achieve. As previously argued (Jones & Davey, 2011), this is where coach education should increasingly venture; into the ground which, although some consider murky, we see as liberating. That is, into areas of coaches' 'intentionality' and 'whatness' of action. Naturally, such ideas are not to be used without careful, critical and considered thought, but not to have them in the dialogue results in unnecessary poverty of professional development, of limiting the promise of future possibilities, and of what can be for coaches.	2022-02-19T16:28:11.375Z	2022-02-16T00:00:00.000	{ "year": 2023, "sha1": "873218d80e7214f82f5532b2117aed218743eaf5", "oa_license": "CCBY", "oa_url": "https://www.tandfonline.com/doi/pdf/10.1080/13573322.2022.2029390?needAccess=true", "oa_status": "HYBRID", "pdf_src": "TaylorAndFrancis", "pdf_hash": "9fbda3e790f08ab6f4a7cb4b3a1982cd149ddf7d", "s2fieldsofstudy": [], "extfieldsofstudy": [] }
246609445	pes2o/s2orc	v3-fos-license	Long-term follow-up for monovision surgery by Implantable Collamer Lens V4c implantation for myopia correction in early presbyopia Purpose To investigate the long-term safety and efficacy of monovision surgery using implantable collamer lens V4c (ICL V4c) implantation in myopic patients with early presbyopia. Setting Eye and ENT Hospital of Fudan University, Shanghai, China. Design Prospective case series study. Methods This study included 64 eyes of 32 patients with early presbyopia, who underwent bilateral ICL V4c implantation for myopia correction. Parameters, including mean spherical equivalent (SE), uncorrected distance visual acuity, corrected distance visual acuity, intraocular pressure, endothelial cell density, presbyopic add power, visual acuity (logMAR) of dominant eyes (D-eye), nondominant (nD-eye) eyes, and both eyes (Bi) at 0.4 m, 0.8 m, and 5 m were recorded at the last follow-up. Results All surgeries were uneventful. At the last follow-up, the safety indices were 1.23 ± 0.18 (D-eyes) and 1.21 ± 0.18 (nD-eyes) (p > 0.05); the efficacy indices were 0.95 ± 0.27 (D-eyes) and 0.92 ± 0.28 (nD-eyes) (p < 0.05), the SE was -0.62 ± 0.47 D (D-eyes); and − 1.21 ± 0.78D (nD-eyes) (p < 0.05), presbyopic add power was 1.31 ± 0.58 D. The visual acuity (logMAR) of D-eyes, nD-eyes, and binocular (Bi) at 5.0 m were: 0.06 ± 0.15 (D-eye), 0.21 ± 0.18 (nD-eye), (p < 0.01), and 0.04 ± 0.13 (Bi); 0.8 m: 0.03 ± 0.18 (D-eye), 0.08 ± 0.16 (nD-eye), (p > 0.05), and − 0.02 ± 0.11 (Bi); 0.4 m: 0.08 ± 0.09 (D-eye), − 0.02 ± 0.08 (nD-eye), (p < 0.001), and − 0.03 ± 0.09 (Bi). Subjects were very satisfied or felt excellent with their visual acuity at near (81.25%) and far distances (87.50%), respectively (versus preoperative, p < 0.001). Conclusion Monovision surgery using ICL V4c implantation is safe and practicable for correction of myopes with presbyopia, with long-term efficacy at near and far distances and patient satisfaction. Introduction As the population ages, an increasing number of people, which was estimated to be 2.1 billion worldwide by 2020 [1], is affected by presbyopia, the age-related loss of accommodation. Presbyopes may be managed using monovision [2], in which the dominant eye (D-eye) may be fully corrected for distance vision, and the nondominant eye (nD-eye) may be undercorrected for near vision, thus producing monocular blur. Earlier, methods such as excimer laser or SMILE were used to correct myopic presbyopia [3]. In addition, glasses, contact lenses, corneal inlays, laser-assisted in situ keratomileusis (LASIK), and lens replacement showed the clinical value of monovision for presbyopia. In patients with high myopia, corneal implants [1] and corneal laser surgery [3] may induce high-order aberrations and reduce contrast sensitivity. Refractive lens replacement can cause loss of accommodation and increase the risk of retinal detachment. Visian ICL (ICL, STAAR Surgical, Nidau, Switzerland) is a phakic chamber intraocular lens (pIOL) with good performance in the correction of different degrees of myopia [4]. Previous studies have shown that ICL implantation is safe and effective in correcting hyperopia and hyperopic astigmatism [5]. With the clinical application of ICL with a central hole (hole ICL), or ICL V4c, the misgivings regarding possible effect of ICL on cataract formation in the elderly population have been dispelled as it improves aqueous humor circulation [6,7]. ICL V4c is also safe for the correction of ametropia in myopic people aged 40 years or above [8]. ICL V4c implantation is reversible; therefore, its application in monocular refractive surgery is worthy of further study. However, to the best of our knowledge, no long-term research has been conducted to observe outcomes of ICL V4c implantation in myopic presbyopes. Therefore, this study aimed to assess the safety, efficacy, and predictability of monovision surgery by ICL V4c implantation in myopic presbyopes and provide a theoretical basis for this treatment modality. The visual acuity (VA) of the D-eye, nD-eye eye, and binocular (Bi) was evaluated at near and far distances. Subjects This prospective observational consecutive case series included 64 eyes of 32 patients with early presbyopia (male/female: 12/20, average age: 43.50 ± 2.62 years old, range 40 to 50 years) who received ICL V4c implantation to correct myopia or myopia with astigmatism at the Eye and ENT Hospital of Fudan University, Shanghai, China, between April 2016 and December 2017. This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Eye and ENT Hospital of Fudan University. All procedures were performed after obtaining written informed consent from the patients. The inclusion criteria were as follows: age ≥ 40 years, stable refractive error (increase of < 0.5D/year) within 2 years, no use of soft contact lens for ≥ 2 weeks, no use of rigid gas permeable contact lens for ≥ 4 weeks, and presbyopic add power ≥ + 0.50 D. The exclusion criteria were as follows: preoperative corrected distance VA (CDVA) (Log-MAR) > 0.15, anterior chamber depth (ACD) < 2.8 mm or endothelial cell density (ECD) < 2000cell/mm 2 , history of ocular inflammation or trauma, lens opacity, glaucoma, previous eye surgery, other diseases of the eye or systemic diseases. Preoperative examinations Spherical equivalent (SE), CDVA, and uncorrected-distant VA (UDVA) were measured by an experienced ophthalmologist using a phoropter (RT-5100, Nidek Technologies, Japan). Presbyopic add power was measured with a Fusion Cross-Cylinder (FCC) at the distance of 33 cm, and the D-eyes and nD-eyes were determined by the card-hole method preoperatively. Slit lamp examination and fundus examination were completed after pupillary dilation. Intraocular pressure (IOP) was measured using a Canon Full Auto Tonometer TX-F (Canon, Inc., Tokyo, Japan); ECD by SP-2000P (Topcon Corporation, Kyoto, Japan); corneal thickness, white to white (WTW); and ACD, angle to angle (ATA) and anterior chamber volume (ACV) by Pentacam HR (Oculus Optikgerate Wetzlar, Wetzlar, Germany). WTW was also measured by IOL Master 700 (Carl Zeiss AG, Germany). Lens power and size calculation Lenses were transparent in patients in this study and they selected ICL V4c implantation to retain their natural accommodation. The ICL power was determined using an online calculator provided by the manufacturer (STAAR Surgical). The D-eyes were targeted for approximately − 0.75 to 0 D (− 18.00 D was selected as the target refraction correction at the spectacle plane for eyes < − 18.00 D) and the nD-eyes for around − 2.25 to − 0.50 D, according to the presbyopic add power of each patient. In the patient with D-eyes < − 18.00D, − 18.00D was selected as the target refraction correction, resulting in residual myopic diopters as the target refraction in some D-eyes, which was on trial in frame glasses preoperatively and accepted by the patients. The ICL size calculation was based on the horizontal WTW, ACD, and ATA distances. We applied an adjustment to the WTW value from Pentacam and referred value from the IOLmaster measurements for ICL sizing. Implantable collamer lens surgical procedure All ICL V4c implantation surgeries were successfully performed by two experienced doctors (Dr. Xingtao Zhou and Dr. Xiaoying Wang). The patients were administered antibiotic eye drops 4 times a day for 3 days before the operation. Performed under topical anesthesia, the ICL V4c were implanted into the anterior chamber which was pre-injected with a viscoelastic agent through the lateral corneal incision. Thereafter, the ICL was adjusted using the manipulator, and the viscoelastic agent was replaced with a balanced salt solution. The detailed steps have been described previously [9]. Postoperative antibiotic eye drops and steroid eye drops were administered 4 times a day for 2 weeks and tapered gradually. Postoperative examinations At 1 month and 3 months postoperatively, and at the last follow-up, data on the UDVA, SE, CDVA, IOP, ECD (except at 1 month postoperatively), WTW, ACD, and ACV, was recorded. Presbyopic add power, CDVA, and UDVA (log-MAR) of D-eyes, nD-eyes, and both eyes at 0.4 m, 0.8 m, and 5 m were recorded at the last follow-up. Measurements were performed under mesopic conditions using two trial frames [14] and two tumbling E charts (VSK-VC-J 0.4 m/0.8 m, Wehen Vision, China) for the monocular and binocular VA at the distance of 0.4 m and 0.8 m, respectively, and a phoropter was used for monocular and binocular distant VA at 5 m. The safety index (SI) was defined as postoperative CDVA over preoperative CDVA, efficacy index (EI) was defined as the postoperative UDVA over preoperative CDVA, predictability was defined as the comparison between postoperative and target SE, and stability was defined as the SE change at long-term follow-up. The subjective satisfaction of VA at near and far distances was recorded preoperatively, 3 months postoperatively, and at the last follow-up (scores 1-5: 1, very dissatisfied; 2, dissatisfied; 3, satisfactory; 4, highly satisfactory; and 5, feel excellent). Statistical analysis All statistical analyses were performed using the Statistical Package for the Social Sciences version 25.0. (SPSS, Inc., Chicago, IL, USA). The results are expressed as mean ± standard deviation. Normality of the data was checked using the Kolmogorov-Smirnov test. One-way/ repeated ANOVA was used to compare the pre-and posttreatment normally distributed data, and Wilcoxon signedrank test was used for non-normally distributed data. The Pearson correlation coefficient was employed to analyze the correlation between the add power or changes in such and other parameters. Differences were considered statistically significant at p < 0.05. Results All patients successfully and uneventfully completed the last follow-up at 43.19 ± 7.06 months (range 33 to 58 months) postoperatively. Table 1 provides an overview of the preoperative patient demographics. All types of data loss were < 5%. Safety and efficacy The safety indices for all eyes at 1 month, 3 months postoperatively, and at the last follow-up were 1.17 ± 0.17, 1.17 ± 0.21, and 1.22 ± 0.18, respectively. The corresponding efficacy indices were 1.03 ± 0.25, 1.04 ± 0.27, 0.85 ± 0.29, respectively. The safety and efficacy indices of the D-eyes and the nD-eyes are listed in Table 2. The percentage of D-eyes and nD-eyes that had an increased CDVA (Snellen lines) for ≥ 2 lines were both 21.88%; for ≥ 1 line were 53.13% and 50.00%, respectively; and for decreased CDVA were 0% and 0%, respectively (Snellen lines) (Fig. 1 A, B, and C). Monovision Presbyopic add power increased by 0.17 D/year from 0.69 ± 0.40 D preoperatively to 1.31 ± 0.58 D at the last follow-up. Five patients had a dominant eye switch. The UDVA, SE, and CDVA values are presented in Table 2. There was no significant difference between the presbyopic add power and the absolute value of SE for the nD-eyes at the last follow-up (p > 0.05). Figure 2 A provides an overview of the VA of the D-eye, nD-eye, and binocular at near to far distances. Figure 2 B, C, and D compare their performances in Snellen VA. Predictability and stability At 1 month, 3 months postoperatively and the last follow-up, 81.25%, 78.13%, and 71.88% of the D-eyes were within the range of ± 0.50 D and 100%, 96.88%, and 90.63% were within the range of ± 1.00 D, respectively. At 1 month, 3 months postoperatively and the last followup, 68.75%, 59.38%, and 59.38% of the nD-eyes were within the range of ± 0.50 D and 96.88%, 85.70%, and 84.38% were within the range of ± 1.00 D, respectively. Endothelial cell density As shown in Table 3, the ECD decreased by 483.5 ± 609.14 cell/mm 2 (13.63% ± 15.54%) at the last follow-up, and decreased by 134.93 cell/mm 2 (3.80%) every year. ECD in none of the eyes decreased to < 2000cell/mm 2 . Subjective satisfaction The subjective satisfaction with near and far distances is shown in Fig. 3. Preoperatively, 3 months postoperatively, and at the last follow-up, 6.25%, 84.38%, and 81.25% participants, respectively, were very satisfied or felt excellent (subjective satisfaction scores of 4 or 5) with their VA at near distance, and 0%, 96.88%, and 87.50% participants, respectively, were very satisfied or felt excellent with their VA at far distance. Correlation analysis There was no significant correlation between presbyopic add power and the anterior segment parameters (ACA, ACV, ACD, or vault), and there was also no significant correlation between the changes in presbyopic add power with the same parameters. The VA of the D-eye at 0.4 m showed a correlation with presbyopic add power at the preoperative level (r = 0.392, p < 0.05), the level at the last follow-up (r = 0.587, p < 0.001), and the difference of those two levels (r = 0.354, p < 0.05). The VA of the nD-eye at 5 m showed a correlation with presbyopic add power at the last follow-up (r = 0.469, p < 0.01). The binocular VA at both 0.4 m and 5 m distance showed a correlation with presbyopic add power at the last followup (r = 0.421, p < 0.05). There was no significant correlation between myopic shift (SE at the last follow-up minus that at 3 months postoperatively) and the decrease of vault (vault at Discussion Presbyopia is the most common refractive error due to agerelated loss of accommodation, which impairs the ability to change the refractive power of the crystalline lens and to focus at different distances. Presbyopes may lose their ability to accommodate at the age of 50 years when the crystalline lens loses elasticity [10]. Monovision is a conventional and appealing choice for presbyopes with good visual quality and patient satisfaction [11]. In this study, the SI and EI at the 43 months followup were 1.22 ± 0.18 and 0.85 ± 0.29, respectively. The proportion of eyes with UDVA ≥ 20/20 reached 68.75%, VA ≥ 20/40 at 40 cm reached 100%, and the satisfaction rate (subjective satisfaction scores of 4 or 5) reached 81.25% (vision at near distance) and 87.50% (vision at far distance), similar to the results of previous studies. Studies on conventional refractive surgery using LASIK have demonstrated that 84.7% of eyes had UDVA ≥ 20/20 (Snellen lines) at 3 months postoperatively, 90.7% VA ≥ 20/40 at 0.4 m, and 86.7% overall satisfaction rate [12]. And such figures were 100% ≥ 20/32, 100% ≥ 20/40 at 0.33 m, and 86.7% of the overall satisfaction rate 1 year postoperatively using SMILE with 1.03 of the SI and 1.04 of the EI [13]. It was observed that the distant VA of the patients at the last follow-up in this study was slightly lower than that at 3 months postoperatively, as shown by the significant decrease in SE and EI. Myopia drift may be related to the high proportion of patients with ultra-high myopia and the further progression of myopia. To date, research has tended to focus on monovision by laser corneal surgery rather than ICL implantation. [14]. The corresponding figures in this study were < 0.04 log-MAR after ICL V4c implantation at 0.4, 0.8, and 5.0 m. The results show a better moderate distance VA than that at far distance in monovision, which agrees with the findings of Nitta et al., which showed a similar improvement in VA at moderate distance by contact lenses [15]. The presbyopic add power at the last follow-up and the difference in add power were related to the VA of the D-eye at a 0.4 m distance, in accordance with the impaired VA at near distance of D-eye in monovision. The presbyopic add power at the last follow-up was related to the VA of the nD-eye at a 5.0 m distance, which is in accordance with the monovision concept of intentional undercorrection of the nD-eye. These results provide the first evidence of the long-term safety and efficacy of ICL V4c implantation in presbyopic myopes, with additional information on monovision refractive surgery and the long-term observation of monovision by ICL V4c implantation. There was no significant difference in the IOP at any time point, and the ECD decreased by 3.8% per year. Compared with 0.6% per year in healthy adults [16] and 0.93% per year in patients who underwent ICL V4c implantation [17], the higher rate in the present study may be related to the age of the subjects, as the enrolled subjects were over 40 years of age, while previous study mostly have tended to enroll younger subjects. We selected 3 months postoperatively as the follow-up time point as previous findings have disclosed a nonsignificant difference in ECD at this time point [17], which excluded the effect of surgical procedures on ECD. ACD, ACA, and ACV showed no significant difference at any postoperative time point, although they were significantly reduced compared with the preoperative level. All vaults were greater than 150 μm after surgery, which was considered the minimal safe value [18]. In those patients with larger decreases in vault, no significant myopic shift was observed at the end of follow-up, meaning the vault may not be the explanation behind the myopic shift. Table 3 The clinical parameters and biometric values of the eyes before and after the implantable collamer lens V4C implantation The axial length change during the follow-ups period can provide a clue for the myopic shift. However, this study did not include enough data for statistical analysis of this. Further study is required to demonstrate the AL change in ultra-high myopes in patients over 40 years of age. The current study was unable to analyze contrast sensitivity or near stereoacuity. Five patients had their D-eye switched, which could be attributed to constant reading or prolonged close work. Individuals, such as programmers, civil servants, or clerks, may get used to seeing at a closer distance. Further research is recommended on why and how the D-eye switch occurs. To the best of our knowledge, there is yet to be any research concerning lens replacement by multifocal intraocular lenses (MIL). Further work is required to compare these two surgeries and to establish a method to evaluate MIL in patients after ICL-V4c implantation. In conclusion, the present study provides additional evidence with respect to good binocular vision and longterm safety and efficacy of monovision surgery by ICL V4c implantation in presbyopic myopia.	2022-02-07T14:34:16.911Z	2022-02-07T00:00:00.000	{ "year": 2022, "sha1": "dd337b16214d8273d2df6d9fec670d7e4ba79675", "oa_license": "CCBY", "oa_url": "https://link.springer.com/content/pdf/10.1007/s00417-021-05545-x.pdf", "oa_status": "HYBRID", "pdf_src": "Springer", "pdf_hash": "dd337b16214d8273d2df6d9fec670d7e4ba79675", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
53207541	pes2o/s2orc	v3-fos-license	A Novel Approach to Anticancer Therapy: Molecular Modules Based on the Barnase:Barstar Pair for Targeted Delivery of HSP70 to Tumor Cells. One important distinction between many tumor cell types and normal cells consists in the translocation of a number of intracellular proteins, in particular the 70 kDa heat shock protein (HSP70), to the surface of the plasma membrane. It has been demonstrated that such surface localization of HSP70 on tumor cells is recognized by cytotoxic effectors of the immune system, which increases their cytolytic activity. The mechanisms behind this interaction are not fully clear; however, the phenomenon of surface localization of HSP70 on cancer cells can be used to develop new approaches to antitumor immunotherapy. At the same time, it is known that the presence of HSP70 on a cell's surface is not a universal feature of cancer cells. Many types of tumor tissues do not express membrane-associated HSP70, which limits the clinical potential of these approaches. In this context, targeted delivery of exogenous HSP70 to the surface of cancer cells with the aim of attracting and activating the cytotoxic effectors of the immune system can be considered a promising means of antitumor immunotherapy. Molecular constructs containing recombinant mini-antibodies specific to tumor-associated antigens (in particular, antibodies specific to HER2/neu-antigen and other markers highly expressed on the surface of a wide range of cancer cells) can be used to target the delivery of HSP70 to tumor tissues. In order to assess the feasibility and effectiveness of this approach, recombinant constructs containing a mini-antibody specific to the HER2/ neu-antigen in the first module and HSP70 molecule or a fragment of this protein in the second module were developed in this study. Strong selective interaction between the modules was ensured by a cohesive unit formed by the barnase:barstar pair, a heterodimer characterized by an unusually high constant of association. During testing of the developed constructs in in vitro models the constructs exhibited targeted binding to tumor cells expressing the HER2/neu antigen and the agents had a significant stimulating effect on the cytotoxic activity of NK cells against the respective cancer cells. INTRODUCTION The search for novel approaches to cancer immunotherapy remains relevant, although a large number of studies have focused on this problem [1][2][3]. One of the reasons why malignant neoplasms emerge and develop in the organism is that the surface of tumor cells is devoid of antigens that can activate the cytotoxic effectors of the immune surveillance system which are responsible for the elimination of transformed cells. In this context, targeted modification of a tumor cell's surface with molecular structures that are recognized by intracellular level of HSPs is the universal protective response of cells, which is associated with the unique ability of these proteins to prevent stress-induced aggregation of intracellular proteins and their denaturation, as well as to ensure repair of partially damaged proteins or their proper elimination if irreversible damage occurs. The listed functions and involvement in the folding of newly synthesized polypeptides and transport of intracellular proteins are referred to as the so-called "chaperon" properties of the constitutive pool of HSPs, which is expressed in cells under normal physiological conditions in the absence of stress [4,5]. However, localization of HSPs is not confined to the intracellular space. In a large series of studies, these proteins were found on the cell surface. In particular, surface HSPs were detected on the plasma membrane in normal [6,7] and tumor cells [8][9][10][11][12][13][14], virus-infected lymphocytes [15], and apoptotic T cells [16][17][18]. It was demonstrated that HSPs with various molecular weights are expressed on the cell surface, but that surface localization is most typical of 70 kDa HSPs (HSP70). The phenomenon of unusual surface expression of HSPs was described not only for in vitro cultured cells, but also for the cells of different patient-derived tissues [12,14]. The functions of HSPs exposed on the cell surface remain virtually unstudied. At the same time, a hypothesis has been suggested that these cell surface proteins are immunologically important, as their emergence on the plasma membrane can be a signal for the immune system to activate cytotoxic effectors and ensure the elimination of infected, transformed, and damaged cells [19]. Indeed, it is well known today that different subpopulations of T cells and NK cells are capable of recognizing highly conserved determinants of various HSPs. In particular, recognition of the membrane-resident HSP70 and Grp75 by γδ-T cells is MHC-nonrestricted [20]; recognition of Hsp70 (the inducible form of HSP70) by NK cells is also MHC-nonrestricted [21,22]. Surface-resident HSPs of tumor cells attract NK cells: their count can increase up to 500-fold in tumors expressing these proteins on their surface [23]. Data in the literature is indicative of in vitro MHC class I-restricted recognition of Hsp70 by human NK cells on the surface of human K562 erythroleukemia cells and human sarcoma cells exposed to heat shock [24]. It was also demonstrated that surface HSP70 proteins cause a strong humoral and cell-mediated adaptive immunity response. According to a number of studies, HSP70 can be attributed to tumor-associated antigens recognized by various types of T cells, such as CD4 -CD8 - [25], αβand γδ-lymphocytes [26,27], and natural killer (NK) cells [10,11,21]. The recognition of both the constitutive and inducible forms of HSP70 by MHC-restricted and nonrestricted immune cells indi-cates that surface HSP70 proteins play a crucial role in antitumor immune responses. Based on this fact, a model of immune surveillance was suggested where these cells ensure the first line of defense against infectious agents carrying HSPs on their surface, protect against virus-infected or transformed cells, and against damaged autologous cells. The lymphocyte pool recognizing conserved HSPs is probably induced during ontogenesis as the skin and intestinal microflora develop. The periodic reactivation of these lymphocytes can be caused by common viral and bacterial infections, as well as various stressful stimuli [19]. Application of HSP70 in antitumor therapy attracted the attention of researchers exploring various approaches to this problem [28][29][30][31][32]. However, most of these approaches are based on the ability of HSP70 to form strong complexes with tumor-specific peptides, rather than on direct recognition of membrane-associated HSP70 by cytotoxic effectors of the immune system. This is possibly related to the fact that in vivo expression of these proteins on cancer cells is observed not in all types of tumor tissues. This circumstance serves as the basis for the assumption that induction of HSP70 translocation onto the surface of tumor cells or targeted delivery of these molecules into malignant neoplasms to attract and activate cytotoxic immune effectors of the immune system is a new, promising direction in antitumor immunotherapy [33]. It has been recently found by a number of researchers and in our preliminary studies that both fulllength HSP70 molecules and synthetic analogues of some HSP70 fragments exhibit an activating effect on natural killer cells. In particular, addition of synthetic HSP70 fragments to a human NK cell culture significantly stimulated the production of IFN-γ by natural killer cells, identically to how this took place in the experiments with recombinant HSP70 [34,35]. Therefore, HSP70 molecules and fragments of this protein can be regarded as promising structures to be used in bioengineering approaches to the fabrication of molecular constructs for targeted modification of the surface of tumor cells in order to potentiate the antitumor cytotoxic immune response. Targeted delivery of such "cytolytic markers" can be performed by incorporating recombinant mini-antibodies against tumor-specific antigens into the recombinant construct module being designed. In particular, antibodies specific to the HER2/neu antigen (р185 HER2 ) or to other cancer markers expressed on the surface of a wide range of malignant neoplasms can be used as such mini-antibodies. This study was aimed at developing a method for targeted HER-2/neu-specific delivery of HSP70 or its fragment to the surface of tumor cells using a twomodule construct, with the barnase:barstar pair em-ployed as a cohesive linker for protein modules. In this construct, the function of the first module carrying a high-specificity anticancer antibody and barnase consists in targeted binding to the surface of cancer cells. In its turn, barnase exposed on tumor cells due to this interaction acts as a site of selective binding between the second module consisting of barstar and HSP70 (or its fragment) and the target cells. In the approach being designed, the selective interaction between the first and the second modules is ensured by an unusually high constant of barstar binding to barnase. This protein heterodimer forms a complex with K d ~ 10 -14 M, which is comparable only to that of the streptavidinbiotin system (K d ~ 10 -15 M). In our previous studies, we have proved that the barnase:barstar complex shows a high potential as an agent for the targeted delivery of various drugs to tumor cells [36][37][38][39]. EXPERIMENTAL The principles of building two-module molecular constructs for the targeted delivery of HSP70 to tumor cells In order to build a supramolecular complex containing the HSP70 protein and the targeting mini-antibody, the barnase:barstar module had to be used to bind HSP70 to one of its components, barstar. It is known from experimental data [10] that the C-terminal domain of HSP70 is responsible for the stimulation of the cytotoxic and proliferative activities of NK cells. Therefore, the C-terminus of HSP70 in the recombinant protein being constructed had to remain unbound and accessible for interaction with natural killer cells, while barstar had to be attached to the N-terminus via a flexible peptide linker ensuring unrestricted rotation of functional domains in the target recombinant protein. These theoretical considerations were taken into account when constructing a plasmid encoding the target recombinant protein His 6 -barstar-HSP70, which consisted of the HSP70 protein (the inducible form of human HSP70 -Hsp70) linked to barstar with its N-terminus via the hinge peptide of human immunoglobulin IgG3 (ThrProLeuGlyAspThrThrHisThrSerGly) and carrying the hexahistidine tail at its N-terminus (Fig. 1). Similar procedures were conducted to build the second variant of the effector module that carried the 16 kDa C-terminal Hsp70 fragment (His 6 -barstar-Hsp70/16), instead of the full-length Hsp70 molecule. The previously designed 4D5 scFv-dibarnase construct [36] was used in this study as the first (targeting) module carrying specific anti-HER2/neu mini-antibodies. The expression level of the tumor-associated HER-2/neu antigen on the surface of the cultured cell lines was tested using fluorescence microscopy and the previously designed recombinant constructs for visualizing cancer cells expressing the HER-2/neu antigen (anti-HER2/neu mini-antibody-barstar•GFP-barnase) [36,37]. It has been demonstrated that the cell culture samples being used are characterized by a sufficiently high level of expression of the HER-2/neu surface antigen (the data are not shown). Treatment of the target cells with the designed constructs Hsp70 and its fragment, Hsp70/16, were delivered to tumor cells as the components of barstar-Hsp70 and barstar-Hsp70/16 recombinant proteins. At the first stage of targeted delivery, anti-HER2/neu mini-antibody (4D5 scFv protein) within the first module of the designed supramolecular complex was bound to the respective tumor-specific antigen on the cell surface (20 µg/ml, 60 min). Next, the barstar-Hsp70 and barstar-Hsp70/16 recombinant proteins were also strongly Amp ori lac I adsorbed onto the cell membrane (50 µg/ml, 60 min) due to the barnase:barstar interaction. Assessment of the efficiency of the binding between the designed constructs and target cells Flow cytofluorimetry was used to assess the efficiency of the targeted delivery of the heat shock protein to the surface of the target cells. The samples of the cells that had interacted with the first and second modules of the designed supramolecular complex were stained according to the conventional procedure [14] using BRM22 antibodies (Sigma, USA) specific to the C-terminus of HSP70 and anti-mouse IgG-FITC (Sigma, USA) as the second antibodies. The measurements were performed on a FACScan laser flow cytometer (Becton Dickinson, USA). At least 10,000 cells were analyzed for each sample. The statistical analysis was performed using the WinMDI software for processing the histograms recorded during the cytofluorimetric analysis. Laser scanning confocal microscopy was used to visualize the targeted delivery of Hsp70 and its Hsp70/16 fragment to the surface of the tumor target cells. In these experiments, the target cells sequentially treated with the first and second modules of the designed constructs were stained with anti-HSP70 antibodies and second antibodies conjugated to AF488 fluorochrome (Molecular Probes, USA) using the conventional staining procedure. The cell precipitate obtained after centrifugation was placed onto a microscope slide; a specialized Mowioll gel-like polymerizable medium (Biomeda, USA) retaining cell morphology and preventing fluorochrome photobleaching was subsequently applied. The microscope slide was covered with a coverslip and left in the dark until the microscopic analysis. The photographs of the cells were taken on an ECLIPSE TE2000-E confocal microscope (Nikon, Japan). Assessment of the effect of treating tumor cells with the designed constructs on the cytotoxic activity of NK cells against these target cells NK cells isolated from human peripheral blood were used as cytotoxic effector cells in a series of in vitro experiments conducted to analyze the antitumor effect of the designed constructs. The magnetic separation technique using an NK cell isolation kit (MACS NK cell isolation kit II, Miltenyi Biotec, Germany) was employed to isolate NK cells from the mononuclear cell fraction obtained by density gradient sedimentation of peripheral blood from donors. The level of NK cell-mediated cytotoxicity was evaluated by CytoTox96 non-radioactive cytotoxicity assay (Promega, USA) based on a quantification of the lactate dehydrogenase (LDH) released from the target cells due to the action of natural killers on tumor cells. The experiments were conducted in accordance with the manufacturer's protocol. Each experimental point was recorded in three replicas. The ratio between NK cells and the target cells placed into the wells was 7:1. BT-474 cells added to the wells 4 h prior to the experiment and subsequently treated with the tested recombinant constructs were used as the targets. At each stage of this procedure, after the addition of the components of the supramolecular complex in the wells and subsequent incubation of the target cells for 30 min at 4°C, the cells were precipitated by centrifugation. Supernatant was then removed, and the wells were washed to remove unbound recombinant proteins. RESULTS AND DISCUSSION The cytofluorimetric analysis demonstrated that the designed constructs can efficiently deliver Hsp70 and Hsp70/16 to the surface of tumor target cells. Similar findings characterizing the binding of Hsp70 and Hsp70/16 to the cell surface were obtained in experiments with the BT-474 and SKOV3 cell lines. Hence, below we summarize the results of the interaction between the designed constructs and BT-474 cells. The components of the 4D5 scFv-dibarnase:barstar-Hsp70(Hsp70/16) supramolecular complex efficiently binded to the cell surface: 4D5 scFv-dibarnase binded to the р185 HER2 antigen, followed by interaction of barstar-Hsp70(Hsp70/16) with 4D5 scFv-dibarnase (Fig. 2). Furthermore, our findings indicate that the barstar-Hsp70 and barstar-Hsp70/16 proteins can independently interact with the cell membrane, leading to a shift in the histogram peaks of the respective control samples towards higher fluorescence signals. According to the literature data, some types of tumor cells can adsorb exogenous HSP70 onto their surface [15,16]. BT-474 cells expressed the р185 HER2 antigen targeted by the mini-antibody in the first module. HEK 293 human embryonic kidney cells were used as a control for nonspecific binding of 4D5 scFv-dibarnase to the cell surface. The cytofluorimetric analysis showed that nonspecific binding to the cell membrane was observed for neither the first (4D5 scFv-dibarnase) nor the second (barstar-Hsp70(Hsp70/16)) module (the data are not shown). Hence, the results indicate that the barnase:barstar systems ensure highly specific and efficient delivery of constructs carrying Hsp70 or its C-terminal fragment to the surface of tumor cells expressing the HER2/neu marker. The efficiency of using the designed constructs for a targeted delivery of Hsp70 and its fragment, Hsp70/16, to the surface of BT-474 and SKOV3 cells was visualized by laser confocal microscopy. The target cells sequentially treated with the first and second modules of the designed constructs were stained with anti-HSP70 antibodies and the second antibodies conjugated to AF488 fluorochrome using the conventional staining procedure. The level of fluorescent staining was analyzed on an ECLIPSE TE2000-E confocal microscope. The results confirmed that Hsp70 and Hsp70/16 were present on the surface of the treated target cells (Fig. 3). The influence of the designed constructs on the activation of cytotoxic effectors of the immune system was studied in the in vitro model of interaction between NK cells and the target tumor cells. BT-474 cells were used as the targets. Assessment of the interaction between effector cells and the target cells in this model demonstrated that targeted delivery of both the Hsp70 and Hsp70/16 molecules to the surface of tumor cells significantly enhances the antitumor cytolytic effect of NK cells. In our experiments, targeted delivery of the full-length Hsp70 molecule and its C-terminal fragment, Hsp70/16, to BT-474 cells enhanced the cytolytic effect of NK cells by more than five-and fourfold, respectively. Treatment of the target cells with individual components of the designed supramolecular complex (4D5 scFv-barnase acting as the "targeting" module and barstar-Hsp70 and barstar-Hsp70/16 acting as "effector" modules) did not significantly influence the cytolytic effect of NK cells. These findings are shown in Fig. 4. CONCLUSIONS We have demonstrated that the designed two-module molecular construct was efficient in a targeted delivery of molecules that activate the cytotoxic effectors of the immune system (heat shock protein Hsp70 and its C-terminal fragment) to tumor target cells. The approach proposed in this study can underlie the design of novel agents for antitumor immunotherapy. This work was supported by the Russian Science Foundation (grant No. 14-24-00106, gene construction and recombinant protein production) and the Program of the Presidium of the Russian Academy of Sciences "Fundamental Research for Biomedical Technologies" (studying the cytotoxic activity of effector cells of the immune system).	2018-11-04T08:39:35.532Z	2018-07-01T00:00:00.000	{ "year": 2018, "sha1": "cbeaa913f62e9982363eb0726ed9db662d2d0795", "oa_license": "CCBY", "oa_url": "http://actanaturae.ru/2075-8251/article/download/10335/139", "oa_status": "HYBRID", "pdf_src": "PubMedCentral", "pdf_hash": "cbeaa913f62e9982363eb0726ed9db662d2d0795", "s2fieldsofstudy": [ "Medicine", "Chemistry" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
54977805	pes2o/s2orc	v3-fos-license	Cognitive Function and Quality of Life in Egyptian Children with Acute Lymphoblastic Leukemia Background and Objectives: Leukemic children face many obstacles that interfere with social activities and are at greater risk for neurocognitive dysfunction. This study aimed at highlighting the problems affecting quality of life in leukemic children given chemotherapy without irradiation. Methods: A Cross-sectional comparative study was conducted in 177 leukemic patients, with 281 subjects as a healthy control. All patients and control were subjected to thorough history and examination. Cognitive functions were examined using the Mini-Mental State Examination. Paediatric quality of life was measured using paediatric quality of life inventory version 4. Results: Health and activity, emotion, social relationship and school achievement were significantly impaired among leukemic children. Orientation, registration, attention, recall and language were adversely affected in leukemic children in comparison to healthy children. Conclusion: Neurocognitive function and quality of life in leukemic children treated with chemotherapy were significantly impaired as compared with the healthy group. Introduction Similar to children suffering from chronic diseases, leukemic patients face many obstacles that interfere with their academic achievement and social activities. This is in part attributed to frequent school absenteeism, feeling of loneliness and the worry of school personnel about dealing with them [1]. Moreover leukemic children are at greater risk for neurocognitive dysfunction owing to treatment regimens used in childhood acute leukaemia. Both chemotherapy particularly methotrexate and craniospinal irradiation are generally associated with poor neurocognitive outcome [2]. The functional neurocognitive domains that are adversely affected by cancer treatments are attention, executive functioning, processing speed, working memory, and ability to learn, which impair the academic performance and social relationship of childhood cancer survivors [3]. Different mechanisms tried to explain these neurological and intellectual disabilities however damage involving the cortical and subcortical white matter is the most highlighted one [4][5][6]. Establishment of delicate balance between effective cancer therapy and the lowest morbidity remains a real challenge among oncologists. As adoption of more intensive treatment strategy even though better cure rate may be achieved, it can lead to unacceptable toxicity [2]. Focusing upon effects of chemotherapy alone during childhood acute lymphoblastic leukemia (ALL) treatment on neurocognitive function as well as quality of life (QOL) did not receive the same researchers' interest as the craniospinal irradiation did. In this study we aimed at highlighting the neurocognitive deficits and problems affecting the quality of life (QOL) in children treated with ALL chemotherapy alone without using craniospinal irradiation. Patients and Methods This study is a cross-sectional comparative study that was carried out in Pediatric Oncology Unit -Oncology Center Mansoura University during the period from 1 st December, 2013 to 30 th November, 2014. All children (198) of both sexes, aged from 9 to 15 years old with ALL who received chemotherapy without craniospinal irradiation for childhood ALL for at least 6 months or those who finished chemotherapy and followed at the study locality were enrolled. Patients exposed to craniospinal irradiation, suffering from mental deficits (IQ below 70), medical or psychiatric disorders were excluded. The comparison group included 295 children from a school near Mansoura University Hospital (from 3 rd Primary school class till 3rd Preparatory school) with age range from 9-15 years old. The comparison group was selected through systematic random sampling method from the list of students. All children (both ALL patients and survivors as well as comparison group) were interviewed using Mini International Neuropsychiatric Interview for children MINI-KID Arabic version [7,8] to exclude any children with psychiatric disorders from studied sample. In the group of patients and survivors, 9 children refused to participate in this study and 8 others were excluded from the study due to their psychiatric disorders (6 major depressive disorders and 2 children obsessivecompulsive disorders). Further, 4 other children were excluded due to their inability to complete the study tools. Therefore, the final total number of patients and survivors who actually participated in the study was 177 from the initial 198. In the healthy group, 13 children refused to participate in this study and another one child could not understand study tools, and they were excluded from this study. Therefore, the final total number of the healthy control group was 281 from the initial 295. Data was using a special designed sheet to include sociodemographic data (age, sex, residence, consanguinity, birth order); treatment factors (treatment compliance, frequency of hospital admission, frequency of follow up in the outpatient clinic). Cognitive functions were examined using the Mini-Mental State Examination (MMSE) [9,10] and the trail-making test (Part A and Part B) which is a quick and easily administered test of visuo-motor tracking, conceptualization, and mental set shifting was used [11,12]. Pediatric quality of life was measured using pediatric quality of life inventory version 4 (PedsQL 4.0) [13] which is Child self-report measured in children ages 5-18 years. Generic Core Scales have been used internationally and translated into many languages including Arabic. There are 4 subscales of the PedsQL 4.0 Generic Core Scales: Physical functioning (8 items), Emotional functioning (5 items), Social functioning (5 items), and School functioning (5 items). Study protocol was approved by Medical Research Ethics Committee of Mansoura University-Faculty of Medicine. Written consent from the parents and their children were obtained at the beginning of the assessment. Data was analyzed using SPSS version 16. Qualitative variables were presented as number and percent. Chi square test was used for comparison between groups. Quantitative variables were tested for normality distribution by Kolomogrov-Smironov test. Normally distributed variables were presented as mean (SD) and unpaired t-test was used for comparison between groups. Non-parametric variables were presented as median (min-max) and Mann-Whitney test was used for comparison between groups. P ≤ 0.05 was considered statistically significant. Results The socio-demographic data were demonstrated in Table 1. No statistical significance was noted as regards to age, sex, residency, consanguinity or birth order between both groups. Quality of life components (health and activity, emotion, social relationship with others and school achievement) all were significantly impaired among ALL cases when compared to their healthy peers (The median for health and activity problems was 450 vs. 800, for emotion was 325 vs. 500, for social relationship was 350 vs. 500 and for school achievement was 200 vs. 500 respectively [p ≤ 0.001] ( Table 2). It is worth mentioning that some patients gave a score of zero as regard health and activity problems, emotion and school achievement according to the PedsQL 4.0 Generic Core Scales [13]. Similarly, cognitive function; measured by MMSE (orientation, registration, attention, recall, language and total MMSE) and Tail making tests (A & B) were adversely affected in ALL children in comparison to healthy children with statistically significant results (p ≤ 0.001for orientation, attention, recall, language and total MMSE while p=0.039 for registration) ( Discussion Although a great attention has been paid to assess the adverse effect of cranial irradiation on the neurocognitive function for childhood leukemia survivors, much less efforts have been spent to evaluate similar effect caused by chemotherapy. In this study we tried to evaluate the cognitive function and quality of life for those children exposed to chemotherapy for leukemia treatment so that detecting those who are in need for psychiatric support or special education services in their academic career. Major cognitive functions (orientation, registration, attention, recall, language and tail making tests) were significantly impaired in leukemic patients and survivors in the current study as compared to their age and sex matched controls. This is consistent with the results of Lofstad et al. [14] who evaluated cognitive functioning of 35 children and adolescents in long-term remission from ALL who were treated with central nervous system prophylactic chemotherapy only and found long-term squeal in global cognitive functions. This is explained by delayed brain changes, most commonly reduction in the white matter volume, intra cerebral calcifications especially with increased intensity of chemotherapy regimens including intrathecal (IT) methotrexate, and changes in glucose utilization and abnormalities in event related potential in those patients [15][16][17][18]. Moreover, Brown et al. [19] reported on 48 patients with moderaterisk ALL treated with a 3-year course of systemic and IT chemotherapy only and concluded that the participants had significantly poorer performance on tests of attention and memory as well as visual construction ability than those who had recently been diagnosed. Furthermore, these cognitive deficits tend to be not only persistent but also progressive throughout the adulthood decades after treatment from childhood ALL resulting in reduced educational attainment and unemployment [20,21] making them at higher risk of suicidal ideation compared with siblings [22]. These results are to some extent different from those of Copeland et al. [23] who assessed the cognitive function of 99 children with different types of cancer, with 73% of them being leukemia and lymphoma through neuropsychological tests and concluded that chemotherapy had only a slight effect on neurocognitive status and was confined to perceptual motor skills. This difference may be attributed to diversity of cancers in nature with different treatment protocols. In addition, quality of life (QOL) in this study among childhood leukemia survivors was significantly lower as compared to age and sex matched healthy children. This is matched with Hicks et al. [24] who evaluated QOL for 13 children with ALL through phenomenological qualitative study and they concluded that QOL among ALL children was certainly affected and the most influencing factor attenuating their QOL was fatigue because of its adverse effect on their activity. This group of children usually face much more mental and emotional stress during their unpleasant visits to the clinics and hospitals, which makes them more vulnerable to fatigue [25]. Moreover, their school performance is adversely affected by poor social adjustment, including problems with peer relations, social withdrawal, and reduced social skills [26]. Furthermore, Zeltzer et al. [27] compared ALL survivors with healthy subjects. Survivors had lower physical, emotional, and vitality functions than normal population. In contrast to the results in this study, other studies reported no difference or even higher scores in the group of childhood cancer survivors as compared to their healthy peers [28][29][30]. The possible explanation of the latter results is the desire of the survivors to be ''as normal as possible,'' causing a response shift or ''paradox of satisfaction'' [31]. Moreover, childhood cancer survivors sometimes deny difficulties on QOL measures and to report high QOL even under difficult living conditions [32]. In conclusion we found that both neurocognitive function and QOL in leukemic children treated with chemotherapy alone without using craniospinal irradiation were significantly impaired as compared with healthy controls. So screening of intellectual and QOL aspects for those patients is mandatory so as to select those who have problems in achieving intellectual tasks or in leading normal lives and offer psychological and educational supports for them.	2019-03-17T13:01:52.199Z	2015-07-06T00:00:00.000	{ "year": 2015, "sha1": "2dd588934a51cafce7c05289a81b7c1f34a381b6", "oa_license": "CCBY", "oa_url": "https://www.omicsonline.org/open-access/cognitive-function-and-quality-of-life-in-egyptian-children-with-acute-lymphoblastic-leukemia-2155-9864-1000291.pdf", "oa_status": "HYBRID", "pdf_src": "MergedPDFExtraction", "pdf_hash": "eb740ad7333d57733eb5e82fc73e93b84f000b09", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
207959711	pes2o/s2orc	v3-fos-license	Functions of the WNT Signaling Network in Shaping Host Responses to Infection It is well-established that aberrant WNT expression and signaling is associated with developmental defects, malignant transformation and carcinogenesis. More recently, WNT ligands have emerged as integral components of host responses to infection but their functions in the context of immune responses are incompletely understood. Roles in the modulation of inflammatory cytokine production, host cell intrinsic innate defense mechanisms, as well as the bridging of innate and adaptive immunity have been described. To what degree WNT responses are defined by the nature of the invading pathogen or are specific for subsets of host cells is currently not well-understood. Here we provide an overview of WNT responses during infection with phylogenetically diverse pathogens and highlight functions of WNT ligands in the host defense against infection. Detailed understanding of how the WNT network orchestrates immune cell functions will not only improve our understanding of the fundamental principles underlying complex immune response, but also help identify therapeutic opportunities or potential risks associated with the pharmacological targeting of the WNT network, as currently pursued for novel therapeutics in cancer and bone disorders. THE WNT SIGNALING NETWORK The WNT signaling network is a central regulator of embryonic development and tissue homeostasis. WNT proteins are phylogenetically highly conserved secreted, cysteine-rich glycolipoproteins (1). Nineteen individual WNT proteins have thus far been described in mammals (2). Best known functions of WNT proteins include regulation of cell cycle, cellular differentiation, cell motility, cellular polarity, and cell death (3). WNT proteins act as directional growth factors that orchestrate patterning, expansion and differentiation of tissues in the organized formation of body plans, and are central regulators of stem and progenitor cell development and maintenance both during embryogenesis and adult homeostasis (4,5). Dysregulation of WNT signaling is implicated in a multitude of diseases, including cancer, fibrosis, bone density disorders, metabolic and neurodegenerative diseases (6). WNT proteins are highly hydrophobic due to post-translational modification by palmitoleic and palmitic acid at conserved cysteine residues. This is afforded through action of the acyltransferase Porcupine (PORCN) in the endoplasmic reticulum (Figure 1). WNT acylation has been shown to be required for the release, receptor interactions, and functions of WNTs (1). The chaperone Wntless (WLS) facilitates transport of acylated WNT ligands to the plasma membrane and aids in FIGURE 1 \| Posttranslational modification and secretion of WNT ligands. Upon translation, WNT proteins undergo acylation in the endoplasmic reticulum by Porcupine (PORCN), a modification required for WNT release (7), receptor interactions (8), and functions (7,9). Wntless (WLS) facilitates transport of acylated WNT ligands to the plasma membrane and aids in WNT release (10)(11)(12). WNT secretion and extracellular transport appears to occur via multiple mechanisms including multi-vesicular bodies and exosomes, cytonemes, lipoproteins, and heparan sulfate proteoglycans (1). WLS protein is recycled via clathrin-mediated endocytosis (13). WNT release (14). WNT proteins act as morphogens in a concentration-dependent manner through the formation of gradients within tissues. How the hydrophobic WNT ligands act at short distances as well as at longer ranges is incompletely understood. Multiple mechanisms that facilitate WNT transport beyond the boundaries of the producing cell have been described, including chaperones, lipoproteins, exosomes, and cytonemes (1). Macrophages infected by viruses or intracellular bacterial pathogens release exosomes and microvesicles that contain pathogen-derived components alongside host membrane proteins (e.g., MHC-I, MHC-II) and immune mediators (e.g., cytokines) that modulate functions of bystander cells (15)(16)(17)(18). Observations of elevated WNT5A protein levels in serum of patients with severe sepsis (19) highlights the possibility that WNT proteins produced in response to infection may act not only locally but also systemically, and thereby shape immune cell differentiation and functions at distant sites. WNT ligands initiate intracellular signaling by binding to cell surface-expressed WNT receptors and co-receptors, including Frizzled (FZD) 7-transmembrane domain receptors, low-density lipoprotein-related proteins (LRP5, LRP6), as well as receptor tyrosine kinases ROR and RYK (20). Cytoplasmic scaffolding proteins of the disheveled family (DVL) are central to initiating intracellular signaling downstream of FZD receptors (21). The functional outcome of WNT interactions with target cells is decided at the level of receptor engagement. Depending on the receptor context, WNT ligands activate distinct intracellular pathways, which can be grouped into β-catenin-dependent and β-catenin-independent signaling events (Figure 2). Individual modalities of β-catenin-dependent and β-catenin-independent WNT signaling have been reviewed in detail elsewhere (3,5,20). Briefly, β-catenin-dependent WNT signaling is mediated by cytoplasmic stabilization of β-catenin, which is controlled by the β-catenin destruction complex. The destruction complex FIGURE 2 \| WNT signaling pathways. (A) WNT/β-catenin signaling. The destruction complex is comprised of APC, AXIN1, CK1, and GSK3β. Phosphorylation of β-catenin by CK1 and GSK3β within the destruction complex results in β-catenin ubiquitination mediated by β-TrCP resulting in proteasomal degradation of β-catenin (22,23) The transcriptional repressor Groucho suppresses expression of genes controlled by TCF/LEF transcription factors. Binding of WNT ligands to Frizzled (FZD) receptors and LRP co-receptors promotes recruitment and clustering of DVL, forming signalosomes (21,24), facilitating recruitment of the destruction complex and stabilization of cytoplasmic β-catenin. Nuclear translocation of β-catenin enables its functions as a transcriptional co-activator for transcription factors of the TCF/LEF family (3). (B) WNT/JNK signaling via FZD, alone or in conjunction with co-receptors (e.g., ROR RYK) activates the small GTPases RAC1 and RHOA engaging the actin cytoskeleton, as well as JNK MAP kinase activation (25)(26)(27). (C) WNT/Ca 2+ signaling downstream of FZD receptors is mediated by phospholipase C (PLC) activation leading to enhanced levels of cytosolic Ca 2+ , resulting in calmodulin/calmodulin-dependent kinase II activation and NF-AT-regulated transcriptional responses (28), and engagement of the actin cytoskeleton. RYK has been implicated as a co-receptor for WNT/Ca 2+ signaling. Figure created with Biorender.com. is comprised of scaffolding proteins adenomatous polyposis coli (APC), axis inhibition protein (Axin), and the kinases casein kinase 1 (CK1) and glycogen synthase kinase 3β (GSK3β). In the absence of WNT ligand binding to FZD and LRP co-receptors, phosphorylation of β-catenin by CK1 and GSK3β within the destruction complex results in βcatenin ubiquitination by beta-transducin repeat-containing E3 ubiquitin protein ligase (βTrCP), fueling continuous degradation of β-catenin by the proteasome (Figure 2A). Binding of WNT ligands to FZD/LRP results in recruitment of DVL and the destruction complex, inhibiting GSK3β and CK1 activity and stabilization of cytoplasmic β-catenin. This enables nuclear translocation of β-catenin where it functions as transcriptional co-activator for transcription factors of the TCF/LEF family (Figure 2A). WNT/JNK-[described as planar cell polarity (PCP) pathway in Drosophila] and WNT/Ca 2+ -signaling are modes of β-catenin-independent WNT signaling. WNT/JNK signaling results in FZD/DVL-mediated activation of the small GTPases RAC1 and RHOA, directing cytoskeletal rearrangements, cell polarization and motility. Activation of JNK can drive c-Jun-and AP-1-controlled transcription ( Figure 2B). WNT/Ca 2+ signaling downstream of FZD receptors and DVL leads to phospholipase C (PLC) activation and enhanced levels of cytosolic Ca 2+ , which activates calmodulin/calmodulin-dependent kinase II and NFATregulated transcriptional responses ( Figure 2C). Tight regulation and precise targeting of WNT signaling is essential, as emphasized by the evolutionary investment in multiple layers and modes of WNT pathway modulation. WNT signaling is negatively regulated by secreted Frizzledrelated proteins (sFRP) and WNT inhibitory factor 1 (WIF-1), which directly bind WNT proteins interfering with receptor interactions (3). The palmitoleoyl-protein carboxylesterase Notum was shown to facilitate serine de-palmitoleoylation of WNT ligands, thereby negatively regulating WNT functions (29). Members of the Dickkopf (DKK) and Sclerostin/SOST families, as well as the glycoprotein Dorsal Inhibitory Axon Guidance Protein (DRAXIN) interact with LRP5/6 and interfere with WNT binding (30)(31)(32). FZD receptor surface availability is regulated through the E3 ubiquitin ligases, Zinc and Ring Finger 3 (ZNRF3) and Ring Finger protein 43 (RFN43), which ubiquitinate FZD receptors destining them for proteasomal degradation (33). ZNFR3 and RFN43 serve as negative feedback regulators for WNT signaling, as they themselves are encoded by WNT target genes (5). WNT RESPONSES TO INFECTION Early studies identified WNT5A as a highly responsive gene in human macrophages upon microbial encounter (19,34). WNT5A has also been found to be highly expressed by tumor-associated macrophages (35), synoviocytes in rheumatoid arthritis (36), macrophages in atherosclerotic plaques (37), and adipose tissueresident macrophages in obesity (38). This has directed initial attention toward elucidating immune functions of WNT5A. However, it is increasingly evident that the host response to infection encompasses differential expression of multiple WNT ligands, receptors and regulators (39)(40)(41)(42)(43). Thus, detailed understanding of how the concerted actions of WNT ligands and potentially concurrent WNT signaling events define host responses to infection is key to firmly establishing immune functions of the WNT signaling network. Bacterial Infections Gram-Negative Bacteria WNT responses to infection have been studies in the context of experimental infection with a limited number of Gram-negative bacterial pathogens (Table 1). WNT pathway activation and functions in the context of Salmonella infection have largely been focused on in a model of gastroenteritis in antibiotic-pretreated mice, as well as in epithelial cell lines in vitro. Salmonella (S.) enterica serovar Typhimurium infection of streptomycinpretreated mice increased mRNA expression of Wnt3, Wnt6, Wnt9a, and protein expression of Wnt2 and Wnt11 in intestinal tissues (43,50,56). In vitro studies indicated that colonization of murine intestinal epithelial cells with S. Typhimurium induced elevated mRNA expression of Wnt2 and Wnt11 (also confirmed at protein level), Fzd2, Fzd4, Fzd6, Fzd7, Fzd8, Fzd9, with limited or no effects on the expression of other Wnt and Fzd genes (50,56). Induction of Wnt2 and Wnt11 expression was attributed at least in part to Salmonella AvrA (50, 56), a bacterial effector that has been implicated in the regulation of β-catenin ubiquitination and stabilization (64)(65)(66)(67). With an increasing understanding of the complex WNT response in Salmonella infection, future studies should explore WNT network activation in macrophages, innate immune cells that are important in the host control of Salmonella infection. Thus far, it has been noted that Wnt5a and Fzd4 expression in S. Typhimurium-infected murine peritoneal macrophages was modestly increased, albeit the impact on the expression of other WNT signaling components was not explored in this study (44). Ehrlichia (E.) chaffeensis infection of human THP-1 macrophage-like cells transiently increased mRNA expression of WNT6, WNT10A, FZD5, and FZD9, while decreasing expression of WNT5B, WNT7B, and FZD7, as determined by pathwayspecific qPCR arrays (42). Expression of WNT regulators such as DKK3 and sFRP2 was suppressed or enhanced, respectively, and a significant number of WNT-target genes were differentially expressed (42). WNT responses upon encounter of pathogenic and nonpathogenic Escherichia (E.) coli have been investigated to some extent in mouse models in vivo. Mono-colonization of mice with E. coli F18 enhanced expression of Wnt2 in the intestine compared to germ-free mice (50). Bladder infection with uropathogenic E. coli (UPEC) induced rapid downregulation of Wnt5a expression in the urothelium of infected mice, which was partially attributed to the bacterial virulence and adhesion factor, FimH (58). This observation seems to contrast a small increase of WNT5A expression described in a human urothelial cell line infected with UPEC in vitro (59). Yet, exposure of mouse thioglycolate-elicited peritoneal macrophages exhibited a marked decrease in Wnt5a mRNA expression when exposed to a non-pathogenic E. coli strain, while expression of all other WNT ligands remained unaltered at the time point While several studies reported WNT5A expression to be responsive to macrophage encounter with Gram-negative bacterial pathogens, it remains to be defined whether opposing directions of the regulation of WNT5A expression reflect pathogen-specific responses, cell-type-dependent variations, or species-specific differences between humans and mice. Some indication that the latter apect might indeed be of importance comes from studies of macrophages stimulated with lipopolysaccharide (LPS), a major cell wall component of Gram-negative bacteria. Increased WNT5A expression has been noted in LPS-stimulated human monocytes, the human monocytic THP-1 cell line, as well as human bronchial epithelial BEAS-2B cells (19,34,53,60,61). LPS-induced WNT5A expression in human macrophages occured downstream of Toll-like receptor signaling and nuclear factor kappa B (NF-κB) activation, and was amplified by inflammatory cytokines such as tumor necrosis factor (TNF) and interferon γ (IFNγ) (19,34,60). In contrast, LPS stimulation did not significantly increase the relatively low Wnt5a mRNA expression of mouse bone marrow-derived macrophages (40), yet LPS stimulation did enhance Wnt5a mRNA expression by mouse RAW264.7 macrophages (37). Nevertheless, the amplitude of the WNT5A response might also be governed by the nature of the invading bacterial pathogen as suggested by observations that Porphyromonas gingivalis LPS induced WNT5A mRNA expression in THP-1 cells to a greater extent compared to E. coli-derived LPS (60). Importantly, however, the LPS-induced WNT response encompasses more than WNT5A. LPS stimulation elevated expression of WNT3 in human monocyte-derived macrophages (52), and WNT1 in differentiated human THP-1 cells and murine bone marrow-derived macrophages (40,48). The latter cells also displayed increased expression of Wnt2b, Wnt5b, Wnt6, and Wnt10b upon LPS stimulation, whereas expression of Wnt5a, Wnt10a, and Wnt11 remained unchanged, and expression of the remaining Wnt genes was below the detection limit (40). Systemic challenge of mice with a sub-lethal dose of LPS in vivo induced rapid elevation of Wnt5b, Wnt10a, Wnt10b, Wnt11, Fzd1, and Fzd5 mRNA expression in spleen tissue, accompanied by elevated expression of the WNT target gene Wisp1. In contrast, expression of Wnt6, Fzd7, and Fzd8 was decreased, whereas Wnt3, Wnt4, Wnt5a, Wnt16, Fzd3, Fzd6, Fzd9, and Fzd10 were not differentially expressed (39). In an ovine model of fetal intra-amniotic LPS exposure, elevated expression of Wnt7b, and reduced mRNA expression of Wnt1, Wnt2, Wnt4, and Ctnnb1 were observed in fetal lung tissue (49). With patterns beginning to emerge in the WNT response to Gram-negative bacteria (e.g., WNT5A expression by macrophages), more detailed insight is required to delineate the impact of cell type-, species-, and pathogen-specific contributions to the amplitude and composition of infection-associated WNT responses. Gram-Positive Bacteria WNT responses upon infection with Gram-positive bacteria are just beginning to be explored ( Table 1). Staphylococcus (S.) aureus infection of Drosophila melanogaster led to enhanced expression of Wnt4 (55). Expression of other WNT ligands was not determined in this study, leaving it to be determined how S. aureus, and other pathogens, affect WNT expression in Drosophila. S. aureus infection of Caenorhabditis elegans induced elevated expression of the WNT2 homolog cwn-2, and suppressed expression of the WNT1 homolog mom-2 and the FZD homolog mom-5 (45). A comprehensive analysis of WNT expression in the Pacific white shrimp Litopenaeus vannamei revealed pronounced upregulation of the mRNA expression of multiple WNT ligands, including LvWnt5, LvWnt6, LvWnt9, and LvWntA in different organs upon S. aureus infection (46). Expression of Wnt5a and Fzd4 by murine macrophages marginally increased upon infection with S. aureus (44). In contrast, infection of RAW264.7 mouse macrophages with Streptococcus (S). pneumoniae has been reported to suppress Wnt5a protein expression (57). Sequencing analyses of lung tissue of mice vaccinated intranasally with S. pneumoniae deficient for the autolysis-inducing factor pep27 revealed enhanced expression of Wnt4, Wnt5b, Wnt7a, and Wnt7b, and impaired Wnt2b, Wnt3, Wnt6, Wnt9a, and Wnt9b mRNA expression (51). Kinase activity profiling in mouse lung tissue of S. pneumoniae-infected mice indicated a reduction in β-cateninstabilizing signals associated with a decrease in β-catenin protein expression (62). Thus, due to the paucity of information it is currently largely unknown if host cell encounter with pathogenic Gram-positive bacteria directly modulates WNT responses and signaling capabilities. Mycobacteria Mycobacterial infections induce significant alterations in the expression of WNT signaling components in infected tissues of a variety of host organisms ( Table 1). Macrophage-associated WNT5A expression was initially described in tuberculosis lung granulomas (34), and WNT5A and FZD4 mRNA expression was significantly elevated in peripheral blood mononuclear cells of tuberculosis patients (44). Mycobacterium (M.) tuberculosis infection of C57BL/6 mice enhanced lung mRNA expression of Wnt1, Wnt6, Wnt10a, Fzd1, and Fzd5, while reducing expression of Wnt2, Wnt2b, Wnt3a, Wnt4, Wnt5a, Wnt7a, Wnt8a, Wnt10b, as well as Fzd3, Fzd7, Fzd8, Fzd9, and Fzd10 (40,41). M. marinum infection of zebrafish enhanced expression of wnt5a, yet suppressed expression of multiple other WNT ligands, receptors and WNT pathway regulators (54). Regulation into opposing directions was noted for some WNT network components, depending on the virulence of the infecting M. marinum strain (54). Macrophages are major host cells for mycobacteria and have been identified as a significant source of WNT expression during mycobacterial infection. In vitro studies showed that infection of monocytes and macrophages of human and mouse origin with mycobacteria across a virulence spectrum (M. tuberculosis, M. avium, M. bovis Bacillus Calmette-Guérin) greatly enhanced expression of WNT5A (34,44). Importantly, expression and induction of WNT5A in human macrophages was more pronounced compared to mouse cells. In M. tuberculosis-infected mice, expression of Wnt6 was localized to macrophages in lung granulomas, and Wnt6 mRNA expression was significantly elevated in murine bone marrow derived macrophages infected with M. tuberculosis or M. avium (40). Taken together, the experimental evidence to date suggests that upregulation of WNT5A by mycobacteria-infected macrophages may be evolutionarily conserved between humans, mice and possibly other species. Nevertheless, expression of other WNT ligands by infected macrophages remains to be explored more systematically across species. Moreover, WNT/WNT receptor expression in infected tissues requires cellular context for more detailed understanding of where WNT responses occur upon encounter of pathogenic mycobacteria. Toward Defining Patterns in the Host WNT Response to Bacterial Infections A WNT response consistently reported for human, and to some extent murine, macrophages to diverse microbial challenges appears to be regulation of WNT5A expression. Yet, as it becomes clear that host WNT responses to bacterial infection reach well beyond differential expression of WNT5A, it will be essential to delineate whether patterns of WNT pathway activity are stereotypical responses of distinct host cell types and tissues to microbial insult, and/or how these responses are defined by the nature of the invading pathogen. With increasing insights into WNT responses to infection arises the need to understand WNT responses in human disease. Studies in patients with severe sepsis and septic shock highlight the complex nature of the host WNT response to microbial insult. Comparisons of blood gene expression patterns in patients with septic shock compared to healthy controls, revealed elevated expression of WNT5B and WNT11, whereas the expression of WNT1, WNT2B, WNT3, WNT6, WNT7A, WNT9A, WNT10A, WNT10B, and WNT16 was significantly reduced (39). Patients with severe sepsis had elevated WNT5A serum levels, and patients with sepsis-associated acute respiratory distress syndrome displayed elevated WNT5A protein expression in lung tissue (19,68,69). An increase of WNT5A protein serum concentrations appeared to correlate with disease progression, whereas a decrease was associated with recovery in critically ill sepsis patients (68). However, WNT5A mRNA expression in whole blood was very low and not significantly different between healthy controls and septic shock patients, whereas alterations in the expression of other WNT ligands was more readily detectable (39). Whether dynamic changes in the expression of WNT pathway components accompanying severe acute infections can be exploited for the development of easily assessible biomarkers remains to be determined. Signatures that might enable patient stratification or rapidly identify classes of causative bacteria are worth exploring. Protozoal and Fungal Infections WNT responses to infections with protozoa and fungi are less well-explored ( Table 2). In mice intraperitoneally inoculated with the protozoan parasite Trypanosoma (T.) cruzi, protein expression of Wnt3a, Wnt5a, and β-catenin in splenic mononuclear cells increased with disease progression (74). Similar patterns were observed for Wnt3a and Wnt5a mRNA and protein expression in murine bone marrow-derived macrophages (BMDMs) (74). In vitro experiments indicated enhanced expression of Wnt3a and Wnt5a, Fzd4, Fzd6, Fzd8, and Fzd9 upon T. cruzi infection of murine BMDMs. In contrast, Leishmania donovani infection of mouse RAW264.7 macrophages resulted in diminished expression of Wnt5a, whereas other WNT ligands and sigaling components were not assessed (80). In human corneas infected with the fungus Aspergillus (A.) fumigatus, WNT5A expression was found to be significantly higher than in uninfected corneal tissues. WNT5A mRNA and protein expression were also enhanced by A. fumigatus infection of human THP-1 macrophages (78). Murine peritoneal macrophages infected with Candida albicans, A. fumigatus, or A. flavus or stimulated with the fungal and bacterial cell wall component Curdlan displayed elevated Wnt5a expression (79). More comprehensive profiling of the WNT network will be required to assess the quality of WNT responses by protozoal and fungal infections and determine to what extent WNT expression and signaling are defined by the host cell vs. the nature of the encountered pathogen. Viral Infections WNT responses to viral infections have been studied in the context of a limited number of viral infections ( Table 2). HIV infection elevated WNT2B and WNT10B expression by human primary astrocytes (71), whereas expression of WNT1, WNT3, WNT5B, WNT9A, WNT9B, and WNT16 remained unaffected, and WNT2, WNT3A, WNT4, WNT5A, WNT6, WNT7A, WNT7B, WNT8A, WNT8B, WNT10A, and WNT11 expression was below the detection limit of the assay (71). HIV infection of mouse neuronal cells of the spinal dorsal horn elevated Wnt5a mRNA expression (77). WNT5A expression was also upregulated in Epstein Barr virus (EBV)-infected nasopharyngeal carcinoma epithelial cells (75). Influenza A infection of mice resulted in impaired expression of Wnt2, Wnt3a, Wnt10b, Fzd2, Lrp4, and Tcf3 in infected lung tissues (72). Human cytomegalovirus (HCMV) infection of human foreskin fibroblasts was associated with WNT5A and WNT5B downregulation (76), whereas HCMV infection elevated WNT2 expression in human mesenchymal stem cells (73). HCMV infection of dermal fibroblasts, placental extravillous trophoblasts, and foreskin fibroblasts was associated with degradation of β-catenin (83). In contrast, β-catenin stabilization was observed in human B cells infected with EBV (86), vaccinia virus-infected HEK293T cells (87), hepatitis B virus-infected Huh7 cells (84), and hepatitis C virus-infected HEK293T cells (85). These reports indicate responsiveness of the WNT signaling network to viral infections. Modulation of β-catenin stabilization might be indicative of viral exploitation of host cell replication and apoptosis. Yet, the WNT responses associated with viral infection noted thus far show no discernible patterns, likely due to the paucity of comprehensive analyses. Systematic comparisons of host cells and different viral classes are required to assess whether there are WNT network signatures that are indicative of a viral infection. WNT FUNCTIONS IN THE HOST RESPONSE TO INFECTION The realization that the WNT network is responsive to infections has driven significant interest in delineating its roles in host defense and immune responses. There is increasing evidence that WNT ligands (and other ligands for WNT receptors) contribute to the host control of phylogenetically diverse pathogens in non-vertebrates and vertebrates (57,74,80,88,89). Some associations between polymorphisms in WNT network genes, and susceptibility and quality of the immune response to infection have been suggested (90)(91)(92)(93). Professional antigen-presenting cells (APCs) such as macrophages and dendritic cells have been studied intensively as sources and targets of WNT ligands (19,40,44,94,95). Roles for WNT ligands in orchestrating phagocytosis, antimicrobial defense and inflammatory cytokine responses have been indicated (Figure 3, Table 3) (48,98,100). WNT ligands have also been implicated in the cellular differentiation and functional polarization of APCs and T cells, bridging of innate and adaptive immune responses (34,94), and shaping lymphocyte functions (107)(108)(109)(110)(111)(112). Considerations for Experimentation Experimental approaches to deciphering WNT ligand-driven immune functions include utilization of mouse models with genetic deletion of individual WNT ligands or receptors. Use of cell-specific deletion (95, 113) or heterozygous mice (40,57) is often indicated due to the deleterious impact of global deletion of individual WNT ligands on embryonic development. SiRNA-mediated knock-down of endogenous WNT components (104,114), interference with WNT/WNT receptor interactions using neutralizing antibodies and recombinant WNT regulators (e.g., sFRPs, DKK) (34,115), as well as plasmid-based overexpression of WNT ligands, receptors and regulators (63,116) are commonly utilized, in particular in in vitro cellbased studies. Conditioned media from WNT-overexpressing cells and recombinant WNT proteins have also been proven as valuable tools for deciphering WNT functions. Of note, some biological responses of innate immune cells observed upon exposure to recombinant WNT protein preparations have been attributed to Toll-like receptor activation, rather than known WNT receptors (61,117). The biological importance of this requires further clarification. As it becomes increasingly evident that multiple WNT ligands are differentially expressed in response to microbial insults, and that WNT ligands are likely to arise from different cellular sources during infection, strategies that broadly target the WNT response as opposed to individual WNT ligands are increasingly employed. Cell-targeted conditional deletion of WLS and PORCN in mouse models, and the use of small molecule inhibitors targeting PORCN activity have proven useful for in vitro and in vivo studies (39,42,80,98,112,118). Similarly, genetic and pharmacologic interference with β-catenin functions as a transcriptional co-activator have been employed to delineate functions of β-catenin-mediated WNT signaling (39,112,119). It is important to note, however, that β-catenin stabilization is not exclusively indicative of WNT/WNT receptor engagement, and that microbial ligands and growth factors can induce β-catenin stabilization (120)(121)(122). Thus, here we chose to focus mainly on studies that link WNT ligands, their receptors and regulators with host defense to infection. Phagocytosis WNT-induced engagement of the actin cytoskeleton suggests that WNT ligands may play an active role in phagocytosis. Indeed, the D. melanogaster glypican dally is a co-receptor in wingless signaling and has been implicated in promoting phagocytosis of a non-pathogenic virus (white spot syndrome virus) by S2 phagocytes. Functional interactions of dally with frizzled 2 and wnt2 in this process were deduced from gene expression analyses (96). In mouse RAW264.7 macrophage-like cells, it has been reported that exposure to Wnt5a-conditioned medium or recombinant Wnt5a enhanced uptake of non-pathogenic E. coli DH5α, as well as latex beads. In contrast, Wnt3aconditioned medium did not enhance phagocytosis (80,98). Fzd5, as well as Rac-1, PI3K and IKK signaling were implicated in mediating the Wnt5a-driven phagocytic activity. Treatment with liposome-encapsulated Inhibitor of WNT Production-2 (IWP-2), a small-molecule PORCN inhibitor (123), impaired macrophage uptake of E. coli DH5α (98). A follow-up study described that exogenously added recombinant Wnt5a also enhanced RAW264.7 cell phagocytosis of S. pneumoniae (Grampositive) and P. aeruginosa (Gram-negative) mediated by Rac-1 and Dvl (57). Mice pre-treated with IWP-2 displayed enhanced bacterial burden within peritoneal cells at 2 h postintraperitoneal infection, and within lung homogenates 5 h after intranasal infection with P. aeruginosa. Similarly, more viable P. aeruginosa were recovered from peritoneal cell lysates of Wnt5a +/− mice compared to Wnt5a +/+ mice (57). These observations further suggest a role for Wnt5a, and potentially other WNT ligands in host cell uptake of P. aeruginosa. However, treatment of RAW264.7 cells with recombinant Wnt5a did not alter internalization of L. donovani (80), and siRNAmediated knock-down of endogenous WNT5A did not impair phagocytosis of M. tuberculosis by human monocyte-derived macrophages (102). Thus, the effects of WNT5A on phagocytosis of bacterial pathogens requires further investigation, including comparisons of extracellular alongside intracellular pathogens and macrophages of different origins. E. chaffeensis is an obligate intracellular pathogen that infects mononuclear cells through caveolae-mediated endocytosis and resides in intracellular vesicles that retain characteristics of early endosomes (124). E. chaffeensis tandem repeat proteins (TRPs) are secreted effectors of E. chaffeensis that have been shown to interact with host cell proteins, including components of the WNT signaling network (125,126). Phagocytosis of TRP120coated microspheres by human monocytic THP-1 cells was impaired by small molecules targeting intracellular signaling components that are also part of the WNT signaling network, such as β-catenin/TCF interactions (FH535), CamKII (KN93), and Rac-1 (NSC23766) (42). In contrast, a PORCN inhibitor (IWP-2) did not impair TRP120-microsphere phagocytosis, suggesting that secreted WNT proteins may not have been directly involved in driving this process. In contrast to phagocytic cells, WNT11 over-expression, but not WNT2 overexpression, by human intestinal epithelial HCT116 cells has been suggested to decrease invasion by Salmonella enterica Typhimurium (50,56). The cellular mechanisms facilitating this protection are unknown and it remains to be established how induction of WNT11 expression by Salmonella infection might contribute to pathogenesis in vivo. Autophagy Several studies have begun to address how WNT ligands might affect the ability of host cells to control pathogenic bacteria. In the case of non-pathogenic E. coli DH5α, exogenous addition of Wnt5a enhanced phagocytosis, but did not alter the ability of RAW264.7 macrophages to rapidly kill the engulfed bacteria (98). In contrast, RAW264.7 macrophages exposed to recombinant Wnt5a displayed a more rapid decline in viable intracellular S. pneumoniae and P. aeruginosa within the first 2-3 h of infection. Wnt5a-induced killing within the first hours of infection was suggested to be mediated by Rac-1 and Dvl. Mechanistically, the authors implicated enhanced autophagy as the mechanism of Wnt5a-induced enhanced control of engulfed S. pneumoniae and P. aeruginosa (57). While S. pneumoniae is targeted by autophagy in non-phagocytic cells (127), the contribution of autophagy in macrophages to controlling this bacterium had not been reported previously. In contrast, the contributions of autophagy to macrophage control of P. aeruginosa require further clarification as beneficial effects for the host as well as the bacteria have been suggested (63,(128)(129)(130)(131). It is noteworthy, however, that after the sharp initial decline of viable intracellular S. pneumoniae and P. aeruginosa in Wnt5a-treated RAW264.7 macrophages, from day 1 onwards the intracellular bacterial burden declined more slowly and at a similar rate in both Wnt5aand control-treated cells (57). Thus, the cellular mechanisms accelerating the initial bacterial killing might be transient, and could be specific to some pathogens as they did not affect macrophage killing of non-pathogenic E. coli DH5a (98). With Wnt5a expression reported to be suppressed by S. pneumoniae and P. aeruginosa infection of macrophages (57), roles of other WNT ligands responsive to infection (e.g., Wnt4, Wnt5b, Wnt7a, Wnt7b) (51) and the net-outcome of WNT signaling in infected cells will need further exploration. Of note, overexpression of β-catenin in RAW264.7 macrophages has been reported to accelerate killing of engulfed P. aeruginosa, which was associated with suppression of autophagy (63). Beneficial or detrimental impact of WNT-autophagy-crosstalk might be defined by a pathogen's ability to exploit intracellular niches for replication and survival. Intracellular bacterial burden in E. chaffeensis-infected THP-1 cells was diminished when cells were exposed to IWP-2, as well as the β-catenin/TCF-1 inhibitor FH535, or the CamKII inhibitor KN93. Small interfering RNAmediated knock-down of WNT pathway components, including WNT5A, FZD5, FZD9, LRP6, CTNNB1, and DVL2 diminished intracellular bacterial burden over 1-2 days of infection, further supporting the notion that intracellular survival of E. chaffeensis in this cell line was facilitated by the action of endogenous WNT ligands (42). A subsequent study indicated that DVL signaling suppressed autophagy and phago-lysosomal maturation in E. chaffeensis-infected cells (132). WNT pathway activation (e.g., by Wnt5a) upon infection with M. bovis BCG has been reported to interfere with IFNγ-induced activation of autophagy in mouse macrophages, a process facilitated by arachidonate lipoxygenase. The same mechanisms have also been implicated for Shigella flexneri and Listeria monocytogenes infection (103). A recent study suggested that in human monocyte-derived macrophages infected with M. tuberculosis, WNT5A contributed to enhancing autophagy resulting in a small decrease in intracellular bacterial burden. In this study, WNT5A-mediated autophagy was suggested as an effector mechanism of IL-36γ (102). However, as WNT5A expression in human macrophages is rapidly induced by M. tuberculosis infection (34), this mechanism might represent an amplification of the WNT5A response of these cells as indicated for other cytokines such as TNF (19). Exogenous addition of Wnt3a conditioned medium suppressed association of intracellular M. bovis BCG with autophagy machinery in RAW264.7 macrophages, which was associated with impaired mRNA expression of autophagy effectors (e.g., Atg5, Atg7, Atg12, p62) (99). With evidence for bi-directional regulation between WNT signaling and autophagy (133)(134)(135), and the notion that some pathogens might exploit this for their intracellular survival, the functional consequences of this crosstalk for pathogen control is an area for future pursuit. Reactive Radicals Additional cell-intrinsic host defense mechanisms that may be regulated by WNT signaling include the formation of reactive radicals. Treatment of RAW264.7 macrophages with recombinant Wnt5a induced NADPH oxidase-mediated ROS production, which has been suggested to contribute to the macrophage control of L. donovani (80). Exogenous addition of recombinant Wnt3a or Wnt3a-conditioned medium to human umbilical vein endothelial cells induced elevated expression of endothelial NADPH oxidase and production of hydrogen peroxide (101), and GSK3β has been implicated as a negative regulator of LPS-induced NADPH-oxidase 1 expression and production of reactive oxygen species production by macrophages (136). These observations could implicate β-catenin-stabilizing WNTs as drivers of ROS production. Yet, treatment of RAW264.7 macrophages with Wnt3aconditioned medium did not affect ROS production upon P. aeruginosa infection (100). Thus, contributions of WNT ligands, in particular endogenously expressed WNTs to ROS production as an anti-microbial defense mechanism require further investigation. Wls-deficiency in BMDMs of Wls fl/fl -Lyz2-Cre mice has been reported to significantly increase mRNA expression of inducible nitric oxide synthase (iNOS, encoded by Nos2) (137), a phenotype also observed in macrophages isolated from myocardial infarct tissue of cfms-icre Wls fl/fl mice (138). This may be reflective of suppression of iNOS expression by autocrine/paracrine WNT signaling. The human iNOS promoter has TCF-4 binding sites and Nos2 expression and nitric oxide production were shown to be positively regulated by β-catenin and TCF-4 (139). These observations suggest that the balance of β-catenin-dependent and -independent WNT signaling could be important for fine-tuning iNOS expression and activity. Whether this bears consequences for pathogen control needs to be investigated. Nevertheless, enhanced iNOS expression by Wlsdeficient macrophages may indicate compensatory mechanisms associated with the inability to release WNT proteins from producing cells and significant elevation of WNT gene expression observed in these cells (137). However, such alterations in WNT expression may be cell specific as F4/80 + liver macrophages of Wls fl/fl -Lyz2-Cre mice did not show significant differences in Wnt4 and Wnt6 expression (118). With some indication that WNT ligands may determine a cell's ability for production of reactive oxygen and nitrogen species, there is also evidence that ROS and NO produced in response to microbial insult may regulate WNT responses. For example, peritoneal macrophages isolated from Nos2 −/− mice showed lower induction of Wnt5a, Fzd4, and Lrp5 mRNA expression upon M. bovis BCG infection compared to wild type control cells. Treatment with an NO-donor restored Wnt5a, Fzd4, and Lrp5 expression in Nos2-deficient macrophages (44), implicating reactive nitrogen species as potentiators of WNT signaling initiation. Dectin-1/Syk-mediated ROS production by murine RAW264.7 macrophages contributed to β-catenin stabilization (79), although how this might intersect with WNTdriven cellular activation remains to be explored. Antimicrobial Peptides Beta-catenin-stabilizing WNT ligands may also play a role in the expression of antimicrobial peptides. A recent study reported that Wnt3a-conditioned medium elevated the P. aeruginosa-induced mRNA expression of cathelicidin-related antimicrobial peptide (CRAMP) and β-defensin 1 in RAW264.7 mouse macrophages, which correlated with a small increase in bacterial killing by these cells (100). Stabilization of β-catenin has also been linked to production of the α-defensins cryptdin-1 and cryptdin-6 by murine intestinal crypts (140). In C. elegans, it has been shown that expression of the antimicrobial peptide clec-60 (human homolog RegIIIγ) upon S. aureus infection is dependent upon the β-catenin homolog bar-1 (45). These observations implicate βcatenin in the transcriptional control of a range of antimicrobial peptides. This encourage analyses on the potential roles of infection-responsive endogenous WNTs in the expression of antimicrobial peptides by infected cells. Tryptophan Metabolism Indoleamine 2,3dioxygenase (IDO) catalyzes the first ratelimiting step in the catabolism of the tryptophan for the formation of active metabolites (141). IDO activity is essential for host resistance to some infections where IDO activity limits the pathogen's access to the essential amino acid tryptophan (142,143). The PORCN inhibitor IWP-L6 and the β-catenin inhibitor iCRT14 enhanced IDO expression and activity in T. cruzi-infected murine macrophages, which was associated with enhanced control of intracellular parasites (74). This suggests that endogenous WNT expression and associated β-catenin stabilization in T. cruzi-infected macrophages suppressed IDO expression in this context. It will be interesting to explore whether induction of WNT/β-catenin signaling by T. cruzi is an active strategy of subverting host defense mechanisms. Importantly, β-catenin activity in CD11c + APCs has been associated with induction of IDO expression and the attainment of a tolerogenic phenotype in DCs (144,145). Whether these apparent differences are reflective of the cellular context (macrophages vs. CD11c + dendritic cells) or the immune responses (parasite infection vs. sterile inflammation) are worth further investigations. Anti-viral State and Type I Interferon Responses GSK3β activity and β-catenin functions have been implicated in the positive or negative regulation of type I interferon (IFN) responses associated with protection or susceptibility of cells to viral infection (82,97,(146)(147)(148)(149)(150)(151)(152). In some studies, direct contributions of endogenous WNT ligands has been confirmed. For example, siRNA-mediated knock-down of Wnt5a in mouse bone marrow-derived macrophages and RAW264.7 cells impaired Chandipura virus-induced IFNβ production associated with enhanced viral load in infected cell cultures (104). WNT2B and WNT9B were identified as negative regulators of Sendai virus-induced interferon beta (IFNβ1) expression, and inhibition of GSK3β-controlled virus-induced type I IFN responses in a β-catenin-dependent manner in a range of human cell lines and primary cells (97). SiRNA-mediated knockdown experiments in human bronchio-epithelial cells (HBECs) identified WNT5A and DKK1 as positive, and FZD5, DVL3, SFRP5, WNT7B, WNT9B as negative regulators of influenza A PR8 replication (114). Knock-down of WNT2 and WNT3 (but not WNT1, CTNNB1, or LEF1) impaired infection of HeLa cells by Dengue virus (153). Enhanced control of flaviviruses was associated with enhanced type I IFN signaling via interferon regulatory factor (IRF)-3 activation and interferon response gene expression. It was proposed that this was facilitated by crossregulation and physical interactions between TANK-binding kinase-1 (TBK-1, which phosphorylates IRF-3) and GSK3β (153). Inflammation WNT signaling has been ascribed both pro-inflammatory and immune-regulatory properties. The paradigm developed over the past decade or so suggests that WNT ligands triggering β-catenin-independent signaling exert pro-inflammatory functions, whereas WNT ligands driving β-catenin stabilization have anti-inflammatory or immune-modulatory effects. These emerging concepts of WNT ligands orchestrating inflammation and immune cell functions have been reviewed and commented on extensively over time (19,120,(155)(156)(157)(158)(159)(160)(161)(162)(163). Here we have chosen to specifically focus on examples for pro-inflammatory and regulatory effects of endogenous WNT ligands. It is increasingly recognized that the WNT response upon infection or microbial challenge comprises complex changes across multiple WNT ligands, receptors and regulators (Tables 1, 2). Moreover, WNT receptors exhibit a degree of promiscuity for WNT ligands (164,165). Thus, the concerted action of WNT ligands and the consequences for local and systemic inflammation in the context of infection require careful consideration. Use of small molecule inhibitors of PORCN (e.g., IWP-2) indicated net pro-inflammatory roles of WNT ligands in mouse models of LPS-induced endotoxemia and E. coli-induced bacterial peritonitis (39,98). Moreover, two studies utilizing small molecule inhibitors of β-catenin functions as transcriptional co-activator (ICG001, iCRT3) independently revealed pro-inflammatory functions of β-catenin in LPSinduced endotoxemia and cecal ligation and puncture (CLP)induced peritonitis (39,119). This challenged the current paradigm of anti-inflammatory roles of β-catenin stabilization and urges further studies to understand the contributions of βcatenin in different (immune) cells to inflammatory responses in vivo. Moreover, which of the individual WNT ligands responsive to infection are responsible for the pro-inflammatory functions in vivo, and what role selective downregulation of regulatory WNTs might play in this context remains to be explored in more detail. Significant focus by some of the earliest studies has been on WNT5A, a WNT family member implicated in driving proinflammatory cytokine responses by myeloid cells via β-cateninindependent signaling (19,(34)(35)(36)166). Endogenous WNT5A has been shown to positively contribute to pro-inflammatory cytokine production by monocytes and macrophages in the context of Mycobacterium and E. coli infection, as well as LPS stimulation (19,34,104). Knockdown of WNT5A in primary human bone marrow stromal cell also impaired basal and LPSinduced release of pro-inflammatory cytokines and chemokines (167). Inhibition of endogenous Wnt5a in a mouse model of HIV-induced neuroinflammation reduced gp120-induced proinflammatory cytokine responses in vivo (77). However, Wnt5a has also been implicated in impairing dendritic cell functions and creating an immune suppressive environment in a mouse melanoma model. Importantly, this was attributed to Wnt5a mediated β-catenin stabilization (168), which contrasts the proinflammatory roles of Wnt5a affected by β-catenin-independent signaling upon microbial challenge. This highlights that the receptor/signaling context rather than the WNT ligand might guide the functional outcome of WNT signaling. Evidence for net anti-inflammatory functions of WNT ligands can be deduced from enhanced pro-inflammatory cytokine release and decreased release of regulatory TGF-β by T. cruziinfected murine macrophages in the presence of PORCN (IWP-L6) and β-catenin/TCF inhibitors (iCRT14) (74). In this study, it was noted that neither PORCN nor β-catenin inhibitors affected T. cruzi-induced IL-10 production by infected macrophages in vitro (74). Similar results were observed in an in vivo LPSinduced endotoxemia model (39). These observations highlight that IL-10 may not be susceptible to WNT regulation in all contexts. An example of infection-induced expression of a specific endogenous WNT ligand being associated with suppression of pro-inflammatory cytokine responses comes from M. tuberculosis-infected mouse macrophages. Bone marrow-derived macrophages from Wnt6-deficient mice displayed elevated TNF expression and secretion upon M. tuberculosis infection (40). That immune-suppressive roles of individual WNT ligands could be vital for host survival upon bacterial infection has been demonstrated for WntD in Drosophila. WntD-deficiency rendered flies more susceptible to L. monocytogenes infection and this was attributed to WntD curbing lethal inflammation by negatively regulating expression of the inflammatory mediator edin via suppression of Dorsal, an NF-κB family member (88). Inhibition of intracellular cell signaling cascades that drive pro-inflammatory cytokine expression (e.g., NF-κB) has been implicated as one of the mechanisms by which β-cateninstabilizing WNT ligands negatively regulate inflammation (169,170). Evidence on how this contributes to shaping cellular immune responses and inflammation during infection in complex in vivo settings will be invaluable to further affirm this regulatory feedback mechanism. FUNCTIONAL FATE OF MACROPHAGES AND DENDRITIC CELLS WITH IMPLICATIONS FOR T-CELL RESPONSES WNT ligands have been implicated in defining the functional polarization and differentiation of macrophages and dendritic cells. These innate immune cells are critical in shaping inflammation and antimicrobial defense, and in instructing adaptive immune responses in their role as professional antigen presenting cells (APCs). Macrophage Polarization Macrophages exhibit functional plasticity along a multidimensional spectrum directed by external and internal stimuli such as microbial products, cytokines, oxygen availability and cellular metabolism (171,172). Accordingly, phenotypic classification of macrophages based on relative induction or suppression of the transcription of individual genes has limitations. Nevertheless, expression of iNOS is commonly associated with (M1-type) inflammatory macrophages, whereas elevation of arginase 1 (Arg1) expression has been associated with (M2-type) alternatively activated macrophages. Nevertheless, Arg1 activity is also found in M1 macrophages regulating NO production by iNOS (171). Wls deletion in resting mouse bone marrow-derived macrophages was accompanied by elevated expression of Nos2, Tnf, and Il6, and reduced expression of the M2-associated gene Mrc1 (macrophage mannose receptor), without affecting Arg1 expression (137). This suggests that basal Wls activity (and by inference the net impact of released WNT ligands) contributed toward M2 polarization of these macrophages. In contrast, several studies indicated that Arg1 expression is regulated by WNT ligands in macrophages upon pathogen encounter. For example, the PORCN inhibitor IWP-L6, but not the β-catenin inhibitor iCRT14, decreased Arg1 expression in T. cruzi-infected mouse macrophages, yet without impacting production of reactive nitrogen intermediates (74). Wnt6-deficient macrophages expressed less Nos2 and Arg1 in response to M. tuberculosis infection, yet reactive nitrogen production was not impaired relative to wild type controls (40). Exogenous addition of Wnt3a-conditioned medium promoted the expression of Arg1 in M. tuberculosis-infected murine BMDMs (41). sFrp1-overexpression, which was accompanied by impaired β-catenin signaling, led to reduced expression of Arg1 and macrophage mannose receptor, CD206 (173). Albeit not evident of endogenous WNT ligands contributing to macrophage polarization, it is worth considering that in vitro exposure of macrophages to recombinant WNT ligands (including Wnt1, Wnt3a, Wnt5a, Wnt7a) have returned varying results on their ability to elicit phenotypic changes indicative of alternatively activated macrophages or macrophages tolerized against LPS activation (61,105,106,117). Dendritic Cell Maturation and Functions The impact of exogenously added or endogenously released WNT ligands and contributions of β-catenin signaling on the expression of functional surface markers of DCs (e.g., MHC-I and MHC-II, co-stimulatory molecules, PD-L1, PD-L2) and DC endocytic capacity has been analyzed in a number of studies returning varying results (115,137,(174)(175)(176)(177)(178)(179)(180)(181)(182). Such variability is likely governed by the use of cells from different species; differentiation and culture conditions; use of exogenous modulation through recombinant WNTs, conditioned media, WNT regulators vs. perturbation of endogenous WNT ligands and signaling events, for example by using small molecule inhibitors or genetic perturbations. Moreover, the utility of recombinant proteins and the possibility of alternative receptors interacting with WNT ligands requires further validation (61,95,117,183,184). Nevertheless, β-catenin activity in myeloid cells has emerged as a rheostat in immune-regulation and tolerance, specifically elucidated in in vivo models of autoimmunity, gut mucosal homeostasis and cancer (95, 120,162,[183][184][185]. Recent studies implicate direct roles for WNT ligands that act via engagement of LRP co-receptors in this regulatory mechanism. Selective deletion of LRP5/6 in CD11c + APCs (which includes DC and macrophage populations in the intestinal mucosa) rendered mice more susceptible to dextran sodium sulfate (DSS)-induced colitis (95,144). This was associated with elevated expression of pro-inflammatory cytokines (e.g., TNF, IL-6, IL-1β) and reduced expression of anti-inflammatory/regulatory effectors (e.g., IL-10, IDO), and functional bias toward fostering Th1 and Th17 responses at the detriment of T regulatory cells (Tregs). The microbiome has been implicated as a driver of inflammation in mice with LRP5/6-deficient CD11c + APCs with expression of a stabilized form of β-catenin specifically in CD11c + APCs ameliorating disease pathology and proinflammatory responses in the DSS colitis model (144). Similar experimental approaches confirmed a regulatory role for β-catenin expression in CD11c + APCs in mouse models of experimental autoimmune encephalitis (EAE), collageninduced arthritis, and tumorigenesis (94,(183)(184)(185). It is interesting to note that the adjuvant utilized in the EAE model contains mycobacterial antigens and that LRP5/6-deficient DCs exhibited reduced pro-inflammatory and enhanced regulatory cytokine responses upon mycobacterial stimulation in vitro (94), suggesting that infection-associated WNT responses might direct APC functions in Treg vs. Th1 and Th17 differentiation. In an OVA-expressing tumor model, Wnt1-overexpression by DCs was associated with reduced T cell receptor stimulation, granzyme B secretion and cytotoxicity by CD8 + T cells (186), whereas conditional knockout of LRP5/LRP6 in CD11c + cells resulted in an increase in granzyme B production by CD8 + T cells (185). Thus, WNT-mediated activation of APCs also bears consequences for subsequent T cell functionality. Of note, there is some evidence indicating that WNT-mediated β-catenin signaling also orchestrates the differentiation of plasmacytoid DCs (187)(188)(189)(190), but consequences for pDC functions remain to be explored. The aforementioned studies support the view that βcatenin-stabilizing WNT signaling engaging LRP5/6 co-receptors can mediate an immune-regulatory profile of DC functions. In contrast, inducible deletion of Wnt5a and one of its receptors, Ror2, rendered mice more resistant to DSS-induced colitis (113). This was accompanied by diminished proinflammatory cytokine responses, including IL-12 expression, and selective impairment in the differentiation of IFNγproducing CD4 + T cells, without impact on IL-17-and IL-10producing CD4 + T cells (113). It was implicated that Wnt5a in this context arose from non-hematopoietic cells such as fibroblasts, whereas Ror2 signaling occurred in the hematopoietic compartment including DCs. Nevertheless, cultured Wnt5adeficient and Ror2-deficient colonic DCs showed impaired proinflammatory cytokine profiles upon LPS stimulation including enhanced IL-12 production and increased responsiveness to IFNγ (113). These observations support the notion of proinflammatory roles of Wnt5a expressed by myeloid cells. They also align with data indicating that myeloid cell-derived WNT5A, and likely other WNT ligands, bridge innate and adaptive immunity by perpetuating the IL-12-IFNγ axis in T cell and natural killer T (NKT) cell responses (34,112,174). Importantly, however, the roles Wnt5a plays in shaping DC functions may be defined by the receptor/signaling output. This is highlighted by findings that melanoma-derived Wnt5a effected a metabolic shift in DCs from glycolysis to oxidative phosphorylation, which was attributed to β-cateninand PPARγ-mediated cellular activation. This resulted in tolerogenic DCs that promoted IDO activity and regulatory T cell differentiation. Relevance of this mechanism was translated into an in vivo melanoma model in mice (145). It will be important to delineate whether factors specific to the pathophysiological context (e.g., immune regulatory molecules, cytokine milieu) explain the apparently opposing outcomes of WNT exposure on DC functions in melanoma vs. inflammatory disorders. T Cell Functions During Infection Genetic deletion of β-catenin in CD11c + cells was associated with only a small increase in the frequency of CD4 + T cells, but no significant changes in the frequency of CD8 + T cells, TCRγδ + T cells, NKT cells, Tregs, or T follicular helper cells were observed (183,191). These findings suggest that β-catenin functions in CD11c + myeloid cells define the quality of T cell responses due to the functional capabilities of APCs, rather than by significantly affecting lymphocyte differentiation. Nevertheless, β-catenin and TCF activation play distinct roles in the development, differentiation and function of innate-like and adaptive lymphocytes, and direct contributions of WNT ligands to these processes have been shown (110,192,193). In a mouse model of lymphocytic choriomeningitis virus (LCMV) infection, TCF-1-deficiency had no effect on the expansion and functions (e.g., IFNγ production and cytolysis) of effector CD8 + T cells (194,195), whereas others reported an increase in effector CD8 + T cells associated with enhanced IFNγ and TNF expression (196). In contrast to the apparently opposing observations for effector T cells, these studies consistently showed reduced numbers of memory CD8 + T cells, reduced IL-2 expression, and impaired expansion of memory cells upon rechallenge (194)(195)(196). However, it was suggested that these TCF-1-mediated effects may not be attributable to βcatenin functions, as conditional knockout of β-catenin in mature T cells did not affect memory T cell numbers or functions upon LCMV and L. monocytogenes infection (197). Yet, in a transgenic mouse model of constitutively activated βcatenin/TCF-1-signaling, an increased proportion of memory CD8 + T cells and increased IFNγ expression during LCMV, vaccinia virus and L. monocytogenes infection were reported (198). These studies indicate that TCF-1 is likely required for CD8 + T cell memory formation and functions after infection. The role β-catenin might play in this and whether WNT ligands have a direct contribution to these signaling events requires further investigation. In an in vitro system, depletion of WNT1, 2B, 3 and 5B from astrocyte-conditioned medium reduced the differentiation of CD8 + T cells toward a CD4 dim CD8 bright T phenotype in cultures of human peripheral blood mononuclear cells. CD4 dim CD8 bright T cells in the central nervous system are thought to be effector memory T cells important in the control of HIV (71). While this study implicated direct involvement of WNT ligands in the formation of this CD8 + T cell subset, it remains to be determined whether WNT ligands mediated this differentiation by acting on the CD8 + T cells, or indirectly via APCs (e.g., by shaping the cytokine milieu). To our knowledge, there are thus far only very few links between WNT ligands and CD4 + T cell functions during infection. In a susceptible mouse model of Leishmania major infection, an inhibitor of Dkk1, which should increase WNT/β-catenin signaling, exhibited reduced numbers of CD4 + T cells in the draining lymph node, with subsequent reduced IL-4 and IL-10 expression after ex vivo stimulation (199). An in vitro study utilizing neutralizing antibodies against WNT5A and FZD5 showed impaired antigen-specific IFNγ production by human PBMCs of antigen-experienced donors re-stimulated with M. tuberculosis antigen. As human T cells expressed FZD5, it was hypothesized that WNT signaling can facilitate memory T cell activation (34). However, these studies did not demonstrate that these effects were driven directly by WNT signaling in CD4 + T cells, nor did they exclude WNT effects on APC functions. Detailed analyses of the WNT receptor and WNT regulator repertoire of different T cell lineages and subsets should guide targeted interventions with WNT signaling events to delineate the roles infection-associated WNT responses play in shaping T cell effector and memory formation and functions. CONCLUDING REMARKS AND FUTURE PERSPECTIVES The WNT signaling network has been firmly established as an evolutionary conserved integral component of host responses to infection. In-depth understanding of how WNT ligands define immune cell functions is beginning to offer mechanistic insights into the contributions of WNT responses to pathogen control and inflammation. Experiments establishing how infectionassociated endogenous WNT responses shape immune cell functionality in vivo will be key to deciphering WNT functions in shaping complex immune responses. Thus far, macrophages and DCs, as well as T cells have been a major focus of delineating WNT-mediated immune functions. Knowledge of how WNT ligands shape the functions of other immune cells, including neutrophils, mast cells, natural killer cells, natural killer T cells, innate lymphoid cells, B cells, etc. is required to begin to understand the complexity of immune-related WNT responses. Considering that the WNT signaling outcome is largely decided by the cellular context at the level of receptor engagement (20), functional redundancy of WNT ligands, or lack thereof, in orchestrating cellular responses of functionally diverse cells in complex tissue environments is an important factor. With a clearer understanding of the WNT receptor and WNT regulator repertoire expressed by different immune-and nonimmune cells in responses to infection, it will be important to determine if there are species-specific differences in the consequences of WNT exposure of functionally similar cells. This is especially critical when investing in utilizing animal models for understanding human pathology and calls for systematic analyses of WNT responses in infected tissues across different species. Reporter mice for WNT ligand and receptor expression as well as WNT signaling activity (200,201) will be invaluable for the temporal and spatial documentation of WNT responses in complex in vivo settings, including infections. Comparisons with human specimens, wherever possible, will be critical. While some consistent patterns of WNT responses begin to arise (e.g., WNT5A regulation in human macrophages), it remains largely unclear whether stereotypical WNT responses to infection exist regardless of the invading pathogen, or whether the nature of the pathogen dictates the WNT response. Comparative studies using phylogenetically diverse pathogens covering spectra of virulence and pathogenesis mechanisms will be essential to distinguish stereotypical and selective responses to microbial infection. In depth understanding of the molecular drivers and regulators of WNT ligand and receptor expression during infection will be invaluable in delineating which microbial factors drive WNT responses. Whereas our understanding of WNT responses and functions during viral and bacterial infections is taking shape, WNT contributions to parasitic and fungal infections remain to be explored in more breadth and depth. Knowledge of the investment of pathogens into actively manipulating the WNT signaling network (202)(203)(204) will inform our understanding of pathogenesis mechanisms and roles of WNT signaling in the host defense against infection. Such insights will be essential when exploring WNT response patterns as biological indicators supporting diagnosis, prognosis and choices for clinical management of infectious diseases (205). Due to the central role of WNT signaling in maintaining tissue homeostasis, including epithelial barrier functions, consequences of immune-related WNT responses reach beyond leukocyte functions. Indications that WNT/WNT receptor interactions shape chemokine responses (186) and cellular metabolism (145) deserve particular attention in the context of immune responses to infection and beyond. Aberrant WNT expression and/or WNT signaling underlying carcinogenesis, fibrosis, and osteoporosis has generated considerable interest in pharmacologically targeting the WNT signaling network (206)(207)(208)(209). Understanding the functional nature and temporal regulation of WNT responses in the host response to infection, and other immune settings, is essential for identifying therapeutic opportunities, but also potential risks of pharmacologically targeting the WNT signaling network. AUTHOR CONTRIBUTIONS JL, JK, and AB wrote the manuscript. TT contributed sections to the manuscript. JL and AB conceived and designed figures and tables. All authors read and approved the final version of the manuscript.	2019-11-14T14:13:17.884Z	2019-11-08T00:00:00.000	{ "year": 2019, "sha1": "8130e8d4fedbc16a6e720eff59e8b9dfc7bb4d32", "oa_license": "CCBY", "oa_url": "https://www.frontiersin.org/articles/10.3389/fimmu.2019.02521/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "8130e8d4fedbc16a6e720eff59e8b9dfc7bb4d32", "s2fieldsofstudy": [ "Biology", "Medicine" ], "extfieldsofstudy": [ "Medicine", "Biology" ] }
3916212	pes2o/s2orc	v3-fos-license	Positive impedance humidity sensors via single-component materials Resistivity-type humidity sensors have been investigated with great interest due to the increasing demands in industry, agriculture and daily life. To date, most of the available humidity sensors have been fabricated based on negative humidity impedance, in which the electrical resistance decreases as the humidity increases, and only several carbon composites have been reported to present positive humidity impedance. However, here we fabricate positive impedance humidity sensors only via single-component WO3−x crystals. The resistance of WO3−x crystal sensors in response to relative humidity could be tuned from a negative to positive one by increasing the compositional x. And it was revealed that the positive humidity impedance was driven by the defects of oxygen vacancy. This result will extend the application field of humidity sensors, because the positive humidity impedance sensors would be more energy-efficient, easier to be miniaturized and electrically safer than their negative counterparts for their lower operation voltages. And we believe that constructing vacancies in semiconducting materials is a universal way to fabricate positive impedance humidity sensors. Resistivity-type humidity sensors, which can perceive and record the change in electrical resistance in response to that in environmental humidity, have been investigated with great interest due to the increasing demands in industry, agriculture and daily life [1][2][3][4][5] . To date, most of the available humidity sensors have been fabricated based on negative humidity impedance, in which the electrical resistance decreases as the humidity increases. However, due to their lower operation voltages, positive humidity impedance sensors would be more energy-efficient, easier to be miniaturized and electrically safer than their negative counterparts. Thus they would have wider applications in protectors for integrated circuits from humidity, energy-efficient automatic air humidifiers, and so on. But so far only several carbon composites have been reported to present positive humidity impedance [6][7][8] . In sensing materials, semiconductor metal oxides are one of the most promising candidates for solid-state chemical sensors due to their high sensitivity, and quick response and recovery 9,10 . Among them, tungsten oxides are very important semiconducting materials, finding applications in gas sensing together with photocatalysis and electrochromism 11 . Focusing on gas sensors, tungsten oxides can be applied for a variety of gases, such as H 2 S, O 2 , NO x , CO x , NH 3 and so on [12][13][14] . Particularly, the sensors for H 2 O (humidity) based on WO 3 (the only reported tungsten oxide based sensors in literature) are WO 3 nanowire humidity sensor on chip manufactured using CMOS-MEMS technique 15 and WO 3 thin-film sensor fabricated using deposition technology 16 . But in most cases, they also functionalize in a composite, just like poly-2,5-dimethoxyaniline/WO 3 composites 17 , the mixture of Cr 2 O 3 and WO 3 18 , and polyaniline/WO 3 composites 19 . And none of them exhibits positive-sensitive property to himidity. As for the sensing mechanism, the response of WO 3 to relative humidity (RH) is generally attributed to the water dissociative chemisorptions process that would result in the formation of hydroxyl groups on the surface of WO 3 crystals; and then, electrons are accumulated on the WO 3 surface. As a result, the resistance of WO 3 crystals decreases with increasing RH 17,20 . To the best of our knowledge, no study focuses on the influence of oxygen vacancies density of metal oxides on humidity sensing property. Furthermore, unlike most of the oxygen-deficient metal oxides, which are not stable (especially in humid condition), WO 3−x crystals with a variety of oxygen-deficient stoichiometries, such as WO 2.72 , WO 2.8 , WO 2.83 and WO 2.9 , can be easily prepared, since they are stable, ordered phases with precise stoichiometries. And the early studies revealed that oxygen vacancy can consistently account for the defect level and trap assisted conduction in semiconducting oxides [21][22][23][24] . Among them, Gillet and co-workers 24 even indicated that the density of oxygen vacancy in WO 3 would be affected by water vapor when the experiments were performed in air. These facts inspire us to design and fabricate various WO 3−x humidity resistors in which the different densities of oxygen vacancy might induce and modulate the humidity sensitivity. Therefore, here we developed an approach to prepare oxygen-deficient tungsten oxides (WO 3−x ) nano-/ micron-structures (NMS) only by heating WO 3 powder in S atomsphere in a vacuum tube furnace, and with the structured WO 3−x crystals, humidity sensors were fabricated simply by screen-printing them onto ceramic substrates with Ag-Pd interdigital electrodes. Surprisingly, a positive humidity-sensitive property was found in the sensors prepared by single-component WO 3−x crystals with high density of oxygen vacancies. And the resistance of WO 3−x crystal sensors in response to relative humidity could be tuned from a negative to positive one by increasing the compositional x. We believe that our method not only provides a new avenue for fabricating highly effective positive humidity sensors by various metal oxides, but also creates a powerful platform to understand and design desirable semiconducting oxides humidity sensors. In addition, the findings on the positive resistance characteristics of single-component material humidity sensors can not only extend the application of humidity-sensitive resistor in different types of miniaturized devices, but also enrich and compensate for the humidity-sensing principles. Materials composition and structure After a systematical investigation 25 , NMS samples (see Extended Data Fig. 1) with different compositions could be obtained. The phase structure of the samples was investigated by X-ray diffraction (XRD). Typical XRD patterns are shown in Fig. 1a. All the diffraction peaks of the sample prepared at 950 °C can be indexed to those of the already known monoclinic W 10 To determine the chemical state of the elements in the obtained samples, X-ray photoelectron spectroscopy (XPS) analysis was carried out. The results are shown in Fig. 1b. For tungsten, a complex energy distribution of W4f photoelectrons was obtained. The W4f core-level spectrum could be deconvoluted into three doublets (six peaks), which are also shown in this figure, where the red line corresponds to the fitted spectrum. The binding energies of the first doublet peaks (solid blue curve) are 35.85 and 37.9 eV for W4f 7/2 and W4f 5/2 lines, respectively, which can be assigned to those of the W 6+ oxidation state 26 . The second doublet peaks (dash dot green curve) have binding energies at 34.1 and 37.1 eV, corresponding to W4f 7/2 and W4f 5/2 lines, respectively, which can be attributed to those of the W 5+ oxidation state 26 . The last doublet peaks (dot rose red curve consist of W4f 7/2 line at 32.7 eV and W4f 5/2 at 35.2 eV, indicating the existence of W 4+ oxidation state on the sample surface 27 . The presence of three oxidation states for W ions reveals that the as-synthesized NMS are all of oxygen-deficient stoichiometries, and from the area ratio of W 6+ over W 5+ and W 4+ in the spectra, it was calculated that the formula of the three samples are WO 2.9 (synthesized at 950 °C), WO 2.89 (at 1050 °C) and WO 2.72 (at 1150 °C), respectively, in which the mean valence of their W ions decreased from 5.58 to 5.27 (see Extended Data Table S1). In addition, all the annealed samples were completely oxidized, containing only W 6+ atoms, without any W 5+ or W 4+ atoms. And all these results are well consistent with those from XRD analysis. It should be noted that because all the annealed samples present the same XRD and XPS results, so in the discussion about WO 3 samples, we just choose one typical annealed sample for comparison. The electron spinning resonance (ESR) spectra of the as-prepared tungsten oxides NMS recorded at room temperature are displayed in Fig. 2. From this figure, it could be easily found that the ESR spectra of the oxygen-deficient WO 3−x crystals exhibit a sharp signal at g = 2.28, while that of WO 3 presents no obvious signals. The signal of the present oxygen-deficient WO 3−x compounds could be attributed to oxygen vacancies. In literature, it was reported that the peak assigned to oxygen vacancies in metal oxides always has a g factor of 2.01 [28][29][30][31] , and most of the excessive electrons are localized at the oxygen vacancies sites. Usually, one oxygen vacancy bounds one extra electron. However, the situation in oxygen-deficient WO 3−x compounds is more complex. Di Valentin and Pacchioni 32 ever analyzed experimentally and computationally the spectroscopic data of WO 3−x , indicating that sometimes the oxidation state of the under-coordinated W ions is still formally of a chemical valence of + 6, while two extra electrons are trapped in the vacancy voids, which can be explained by the following synthetic expression: W 6+ /V O (2e − )/W 6+ (V O is the oxygen vacancy). Thus, the shift of the g value for the present oxygen-deficient WO 3−x compounds to a higher one might be owing to the presence of two charge centers (two unpaired electrons) trapped in the oxygen vacancy. Moreover, the sharp g signal of metal oxides becomes stronger with increasing concentration of oxygen vacancies [28][29][30][31] . For the present oxygen-deficient WO 3−x compounds, as the value of x increased, the sharp g signal gets stronger and stronger, indicating that the density of oxygen vacancies in them increases. But there is no signal in the ESR spectrum of WO 3 , implying that the annealed samples were almost completely oxidized, containing very little or even no oxygen vacancies. This result is also in good agreement with the XPS spectra. Humidity sensitivity of WO 3−x sensors. The dependence of impedance on RH was measured for the sensors fabricated with the obtained WO 3−x crystals (see Fig. 3), in which all the dried sensing WO 3−x films were of about 171 μ m in thickness. From the curves, it can be clearly seen that for all the four kinds of samples, under low-humidity environment (here 11% RH), the sensors fabricated with WO 2.72 present the lowest impedance, and the impedance of the sensors fabricated with WO 2.89 is lower than that with WO 2.9 , while the resistance of the sensors with WO 3 is the highest. That is to say, under low RH environment, the higher the density of oxygen vacancies in tungsten oxide crystals, the lower the impedance of the senores fabricated with them, which is similar with the conduction behavior of many semiconducting oxides 22 . The reason for this phenomenon is that oxygen vacancies are the centers of positive charges, which bound electrons easily, and the electrons around oxygen vacancies are easily excited to the conduction band; thus the conductivity of semiconducting materials can be improved with increasing density of oxygen vacancies. Moreover, from Fig. 3, it can also be seen that the resistance of the sensors fabricated with WO 2.72 NMS increases remarkably at low humidity (11~75% RH), and still increases somewhat at higher humidity (75~95% RH), presenting a completely positive resistance sensitivity to RH. With decreasing density of oxygen vacancies in tungsten oxide crystals, however, the fabricated sensors will present different sensing behaviors. The sensors fabricated with the as-prepared WO 2.89 nanorods exhibits a positive resistance response to humidity in the range of 11-85% RH with slightly increased resistance. But at the humidity higher than approximately 85% RH, a negative response could be observed in such sensors, presenting an extreme point there (see the inset of Fig. 3). When the density of oxygen vacancies was further reduced (here in the WO 2.9 nanorods), the response curve recorded from the sensors still exhibits a similar sensing profile with that of WO 2.89 sensors, but the extreme point was found at a lower humidity of about 54% RH. When it comes to WO 3 NMS, with increasing RH, the impedance of the sensors fabricated with the annealed NMS rapidly decreased monotonously, presenting a completely negative resistance characteristic to RH, which is in accordance with the already reported humidity sensitivity of the sensors fabricated with other forms of WO 3 materials 15,16 (also see Extended Data Fig. S2). In summary, the humidity sensitivity of the sensors fabricated with the present structured tungsten oxide crystals will display a gradual transition from a positive humidity-sensitive property to negative one depending on the density of oxygen vacancies in the sensing materials. When the density of oxygen vacancies is high (in WO 3−x samples with x > 0.11 in this study), they will present a completely positive humidity-sensitivity in the entire RH range (here from 11-99%). With a medium x (0.1 ≤ x ≤ 0.11 in this work), they will exhibit a positive humidity-sensitive property at low RH, but still a negative one at high RH. In such case, the extreme point may gradually move to a lower RH with decreasing x. When the density of oxygen vacancies is low (in samples with x < 0.1), for example with x = 0 as in the present WO 3 NMS, the fabricated sensors may show an almost completely negative humidity sensitive characteristics in the entire RH range. It is well known that the response and recovery behavior is an important characteristic for evaluating the performance of humidity sensors. Figure 4 shows the response and recovery characteristic curves for one cycle (corresponding to the adsorption and desorption processes of water molecules) of the sensors fabricated with the obtained tungsten oxide crystals. It can be seen that, as the humidity increased from 11-95% RH, the impedance of the sensors with WO 2.72 and WO 2.89 increased, showing a positive resistance characteristic, but the humidity-sensitive resistors fabricated with WO 2.9 and WO 3 presented a negative humidity impedance characteristics, i.e., the impedance of the sensors decreased as the humidity increased. For the sensors with WO 2.72 , the impedance increased from 6.9-26 kΩ, presenting a gain of 276.8% as the humidity increased from 11-95% RH. As the density of oxygen vacancies decreased, the impedance gain of the fabricated sensors would decrease. For example, the sensors with WO 2.89 still had a gain of about 33% with the impedance from 837-1114 kΩ. But when it comes to WO 2.9 , the impedance of the as-fabricated sensors decreased from 936-627 kΩ under similar RH environment. And the resistance of the sensors with WO 3 even presented a sharp decrease from 20716-424 kΩ (also see Extended Data Fig. 3). Such phenomenon is correspondent with the results as shown in Fig. 3. From the response and recovery characteristic curves as shown in Fig. 4, the response time of the sensors with the obtained WO 3−x crystals (defined by that a sensor reaches 90% of the total impedance change as the humidity increases from 11-95% RH) and their recovery time (defined by that a sensor reaches 90% of the total impedance change as the humidity decreases in the opposite direction and range) can be calculated. Figure 5 shows their statistical response and recovery times. It can be seen that, due to the different densities of oxygen vacancies in the sensing WO 3−x crystals, both of the response and recovery times changed dramatically as the environment humidity varied. The response time of the sensors with WO 2.72 NMS was about 6 s, indicating that such sensors have a very good response to humidity, which is comparable with those of the well-known WO 3 based sensors [17][18][19] . The sensors with the present WO 3 NMS also displayed a smiliar, quick response time of 4 s in the humidification process. But for the sensors with WO 2.89 and WO 2.9 NMS, they should take a longer response time (approximately 12 and 98 s, respectively) in the humidification process. The sensors with WO 2.72 or WO 3 NMS would display a quick response time, because both of them simply present a completely positive or negative humidity sensitive characteristics in the entire RH range (11-99% RH). However, because the sensors with WO 2.89 and WO 2.9 NMS exhibit a positive humidity-sensitive property at low RH but still a negative one at high RH, their response times become longer. This phenomenon might be correlated with the strong competition between the oxygen vacancies induced positive humidity sensibility and water-related negative humidity sensitive property. As for the recovery time of the sensors fabricated with the obtained WO 3−x crystals, as shown from the Fig. 5, it would be reduced as the density of oxygen vacancies decreased, which could be understood from the fact that the desorption kinetic energy of water from the surface of WO 3 is generally smaller than the one of hydroxyl groups from the oxygen vacancies 33,34 . In addition, it may be worth noting that the response curve of WO 2.9 (yellow in Fig. 4) is very special and different from those of the other sensors. When the RH decreases from 99-11%, it will pass through the extreme point of the curve, where the humidity-sensitive property changes from a positive to negative one. In such case, the response curve will increase initially, and then decrease, as shown in Fig. 4. This phenomenon is correlated with the conduction mechanism of WO 2.9 , in which oxygen vacancies induced conduction dominate at low RH, but water-related electrolytic conduction at high RH. Figure 6 presents typical humidity hysteresis of the sensors fabricated with WO 2.72 and WO 3 NMS (also see Extended Data Fig. 4), respectively. The black lines in this figure were measured from low to high RH (for the adsorption process), and the red lines were done in the opposite direction (for the desorption process). In accordance with the response and recovery characteristics, the impedance of the sensors fabricated with WO 2.72 increased as the humidity increased, and the pathway of its desorption process was located at the higher position of the loop, revealing that the rate of the desiccation of the adsorbed water was slower than that of the adsorption. On contrary, besides the expected negative humidity impedance characteristic, the desorption process of the WO 3 sensors was located at the lower position of the loop, which is opposite to that of WO 2.72 sensors. Moreover, it was calculated that the largest humidity hysteresis of the sensors fabricated with WO 2.72 and WO 3 were about 45.8% at 33% RH and 16.7% at 75% RH, respectively. The sensors with WO 2.72 exhibited a relatively wide hysteresis loop, indicating a somewhat slower desorption process, which is consistent with the results on their recovery characteristics as shown in Fig. 4. A possible qualitative mechanism to explain the humidity sensing properties of the present structured tungsten oxide crystals is proposed hereafter. Due to the totally different humidity impedance characteristics displayed by these sensors, their humidity mechanisms may have a distinctive difference. For the WO 3 sensors, due to the very little of oxygen vacancies existed in the sensing materials, the large increase in conductivity with increasing RH can be assigned to the adsorption of water molecules on the surface of the WO 3 crystals, because the water-related electrolytic conduction mainly functionalizes as a surface mechanism 35 . At low humidity, only very few water molecules are adsorbed, so the coverage of water molecules on the surface of the WO 3 crystals is, most probably, not continuous, based on which, the electrolytic conduction is difficult to act; thus the conductivity of WO 3 sensor was poor. At high humidity, one or several layers of water molecules might be formed onto the surface of WO 3 crystals; thus, the electrolytic conduction takes place along with the weak protonic transport, and even becomes dominating in the transport-process 3 . In a word, the impedance of the sensors fabricated with WO 3 decreases greatly as the humidity increases. On the other hand, for the sensors with the present structured oxygen-deficient WO 3−x crystals, the conducting mechanism due to the oxygen vacancies induced conduction would compete with the water-related surface mechanism, and even become dominating in the transport process at low humidity. In literature, the frequently observed semiconducting oxide conductivity has been often attributed to oxygen vacancies. It was reported that, one oxygen vacancy could bound two electrons as free carriers 36,37 and oxygen vacancies could introduce donor levels between the conduction and valence bands 24,38 , which would result in increased conductivity for semiconducting oxides. At low humidity, the electrons related to the ionization of donor centers (oxygen vacancies) of the present WO 3−x crystals are the effective charge carriers, thus enhancing the conductivity of WO 3−x crystals when the density of oxygen vacancies increases. With increasing RH, water molecules are adsorbed in oxygen vacancies. The adsorbed water is easy to dissociate from oxygen vacancies, being converted into two bridging hydroxyl groups per initial vacancy via proton transfer to a neighboring bridging oxygen atom 33,34 . Then the number of electrons bounded in the oxygen vacancies decreases, reducing the conductivity of WO 3−x . At high humidity, one or several layers of water molecules are formed on the WO 3−x crystal surface, so the water-related electrolytic conduction would happen to a large extent. Then the oxygen vacancies induced conduction and water-related electrolytic conduction would compete with each other. When x > 0.11, the oxygen vacancies induced electrical conduction is the dominant mechanism in the sensing materials, so their sensors exhibit a positive humidity impedance characteristic in the entire RH range from 11 to 99%, due to the decreased number of oxygen vacancies induced electrons with increasing RH. In such case, under high level of humidity environment, due to the competition of water-related electrolytic conduction, the rate of impedance increase of the sensors may slow down to some extent, like the behavior of WO 2.72 sensors as shown in Fig. 3. With decreasing x to 0.11 (in WO 2.89 sensors), the oxygen vacancies induced electrical conduction of the sensors is the primary mechanism at low humidity (lower than 85% RH), but the water-related electrolytic conduction would become dominating at high humidity (85-95% RH in this case), thus resulting in the turning point of impedance at 85% RH. As the density of oxygen vacancies is further reduced, the turning point of the sensor impedance will move toward a lower humidity (here at 54% RH for the WO 2.9 sensors), and even disappear, presenting a totally negative humidity impedance characteristic (as shown in the present WO 3 sensors). In summary, humidity sensors were successfully fabricated with WO 3−x crystalline NMS of different densities of oxygen vacancies. The resistance of WO 3−x crystal sensors in response to relative humidity could be tuned by changing the compositional x. When the density of oxygen vacancies is high (in WO 3−x samples with x > 0.11 in this study), they will present a completely positive impedance humidity-sensitivity in the entire RH range (here from 11-99%). With 0.1 ≤ x ≤ 0.11, they will exhibit a positive impedance humidity-sensitive property at low RH, but still a negative one at high RH. In such case, the extreme point may gradually move to a lower RH with decreasing x. When the density of oxygen vacancies is low (in samples with x < 0.1), for example with x = 0 as in the present structured WO 3 crystals, the fabricated sensors may show an almost completely negative impedance humidity sensitive characteristics in the entire RH range. Moreover, the sensors fabricated with WO 2.72 crystals possess high sensitivity with a short response time of about 6 s, a recovery time of approximately 100 s and the largest humidity hysteresis of about 45.8% at 33% RH. The humidity sensitivity of the present tungsten oxides sensors may be controlled by the combination of oxygen vacancy induced electrical conduction and water-related electrolytic conduction. Materials preparation and characterization. To fabricate the proposed tungsten oxides NMS, a high-temperature thermal evaporation process via a horizontal quartz tube furnace was used 25 . In an optimum process, 1 g commercially-bought reagent-grade WO 3 powder (Tianjin Fuchen Chemicals, China) was loaded in a quartz boat located at the center of the furnace, while another boat with 1 g Aladdin-reagent S powder was located at the upstream from the WO 3 powder. Before heating, the quartz tube was evacuated and flushed with Ar gas repeatedly for several times to deplete the residual gases. Then the furnace was heated up to the selected temperatures of 950, 1050 and 1150 °C with a dwelling time of 1 h. After that, the furnace was cooled down naturally to room temperature. Finally, powder-like products could be collected. For the purpose of comparison, a portion of the collected powders was annealed in air at 500 °C for 2 h in a muffle furnace. The morphology of the obtained samples was examined by a field emission scanning electron microscope (FE-SEM, S4800, Hitachi, Japan). The chemical composition was measured by an energy-dispersive X-ray (EDX) spectroscope attached to the SEM. The phase structure and composition were identified by XRD (D/max-RB, Rigaku Corp., Japan; Cu Kα radiation, and λ = 1.5418 Å) in continuous scanning mode with a rate of 6°/min. The chemical state of the elements in the samples was investigated by XPS (Thermo escalab 250Xi, ThermoFisher Scientific, USA; non-monochromated Mg Kα radiation, photon energy 1253.6 eV), and the results were calibrated by C1s line (binding energy, 285 eV). The unpaired electron and defect in the samples were investigated by ESR (JEOL JES-FA200, Japan) at room temperature with a microwave frequency of 9.44 GHz. Diphenylpicrylhydrazyl was used for the g value calibration. Sensors fabrication and measurement. Extended Data Fig. 5 shows schematically the sensors, in which the inset (a) shows a blank device and (b) a device coated with the sensing materials. During the fabrication, 0.2 g each of the collected powders was firstly milled and mixed with 2 mL deionized water to form a paste. Then 0.1 mL of the prepared pastes was spinning-coated by a coater (KW-4 A, China) at a rotational speed of 1000 rpm for 20 s onto an alumina ceramic substrate (Company Elitetech, China) with a size of 6 mm in length, 3 mm in width and 0.5 mm in thickness, where the screen had already been printed with two Ag-Pd interdigital electrodes of five fingers with a distance of 0.15 mm. Finally, a humidity sensor was obtained after the film was dried at ambient temperature (about 25 °C) in air for 1 h. After drying, the thickness of the sensing materials was measured by an optical microscope (BX53F, Olympus, Japan). Because the sensing film thickness also affects the performance of the sensors significantly (see Extended Data Fig. S6), all the film sensors were prepared under the same conditions excepting the presently investigated composition of the sensing WO 3−x materials. The characteristic humidity sensitivity of the as-fabricated sensors, including impedance vs. relative humidity, response property and humidity hysteresis, was examined by a Keithley 2410 analyzer (USA). During the measurement, the applied operation voltage was AC 1 V and operation frequency was 1000 Hz. And the controlled humidity environment was achieved by a series of super-saturation aqueous solutions with different salts of LiCl, MgCl 2 , Mg(NO 3 ) 2 , NaCl, KCl and KNO 3 , which could present a relative humidity at 25 °C of approximately 11%, 33%, 54%, 75%, 85% and 95%, respectively. In typical measurement, each sensor was soaked at 25 °C in an atmosphere of different RH levels in the six chambers with different salt solutions till it reached the adsorption-desorption equilibrium for water, and then the impedances of the sensors with the RH and time were measured, respectively.	2018-04-03T05:45:40.397Z	2016-05-06T00:00:00.000	{ "year": 2016, "sha1": "e0fa330766d01d9c6fbde60c4500c90c5a504525", "oa_license": "CCBY", "oa_url": "https://www.nature.com/articles/srep25574.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "e0fa330766d01d9c6fbde60c4500c90c5a504525", "s2fieldsofstudy": [ "Materials Science" ], "extfieldsofstudy": [ "Materials Science", "Medicine" ] }
235713590	pes2o/s2orc	v3-fos-license	Combined Action of Anti-MUC1 Monoclonal Antibody and Pyrazole-Platinum(II) Complexes Reveals Higher Effectiveness towards Apoptotic Response in Comparison with Monotherapy in AGS Gastric Cancer Cells MUC1 mucin is a transmembrane glycoprotein aberrantly overexpressed and underglycosylated in most epithelium origin cancers. Combining chemotherapeutics with monoclonal antibodies toward cancer-related antigens is one of the new strategies in cancer therapies. In this study, we assessed the effectiveness of 10 μM cisplatin (cisPt), two pyrazole-platinum(II) complexes (PtPz4 and PtPz6), and 5 μg/mL anti-MUC1 used as monotherapy, as well as cisplatin and its derivatives combined with mAb on apoptotic response and specific cancer-related sugar antigens in AGS gastric cancer cells. Flow cytometry, RT-PCR, Western blotting, and ELISA tests were applied to determine the influence of examined compounds on analyzed factors. PtPz6 combined with anti-MUC1 revealed the strongest apoptotic response compared to control and monotherapy. The combined action of both cisPt derivatives and anti-MUC1 was more effective than monotherapy in relation to Bad, Bcl-xL, Bcl-2, caspase-9, caspase-3, as well as pro- and cleaved caspase-3 protein, and T, sialyl Tn sugar antigens in cell lysates, and Tn, T, sialyl Tn, sialyl T antigens in culture medium. Additionally, PtPz4 administrated with mAb was revealed to be more potent than used alone with regard to Bax protein and Bid expression, and PtPz6 used in complex with anti-MUC1 revealed more efficient action towards Akt and sialyl T antigen expression. These data indicate the rationality of the potential application of combined treatment of anti-MUC1 and cisPt derivatives in gastric cancer therapy. Introduction Gastric cancer (GC) has been stated as one of the leading causes of tumor-related deaths globally [1]. Most of the GC patients are diagnosed with late stages, with poor prognosis and limited possibilities of effective therapies. Cisplatin, a common chemotherapeutic, is widely applied in the treatment of various human cancers [2,3]. Its action in cancer cells is based on the potential to crosslink with the DNA purine bases, attenuating DNA repair, leading to DNA damage, and consequently inducing apoptosis. While cisplatin is effective, its usage can be limited by its severe side effects, as neurotoxicity, anaphylaxis, cytopenias, and others. Furthermore, resistance to this drug can also reduce its application in many tumors [4,5]. Therefore, in the last years, numerous platinum-based compounds were synthesized with the goal of reducing adverse effects and overcoming drug resistance [6]. Recent studies on breast cancer cells revealed that pyrazole-platinum(II) complexes might induce cell cycle arrest and apoptosis, therefore constituting potential anti-cancer agents [7]. Apart from that, there have been attempts to apply combination therapies of cisplatin with other drugs. Cisplatin (Sigma, St. Louis, MO, USA), PtPz4, and PtPz6 were dissolved in dimethyl sulfoxide (DMSO; Sigma, St. Louis, MO, USA) (concentration of the compounds in stock solutions was 10 mM) and diluted with supplemented medium to obtain a final concentration of 10 µM. Cells were incubated for 24 h with FBS-free medium alone (control), medium containing PtPz4, PtPz6, cisplatin (cisPt), and monoclonal anti-MUC1 antibody (5 µg/mL). For combined treatment, AGS cancer cells were pretreated with anti-MUC1 mAb 5 h before platinum and platinum complexes addition. After the washing step (Phosphate Buffered Saline (PBS); pH 7.4), cell lysis was performed using Radio-Immunoprecipitation Assay lysates and culture media were collected, frozen at −70 • C, and then used for Western blot analyses and ELISA tests. To assess the protein concentration in the cell lysates, a BCA Protein Assay Kit (Pierce, Rockford, IL, USA) was used. For real-time PCR assay, the monolayers were washed three times with 10 mM PBS. Then the cells were collected into sterile plastic tubes and disrupted by a sonicator (Sonics Vibra cell; Sonics & Materials, Leicestershire, UK) (10 W, three times for 15 s on ice). For RNA isolation, aliquots of the homogenate were used. Pharmaceutics 2021, 13,968 3 of 19 mAb 5 h before platinum and platinum complexes addition. After the washing step (Phosphate Buffered Saline (PBS); pH 7.4), cell lysis was performed using Radio-Immunoprecipitation Assay (RIPA) buffer (Sigma, St. Louis, MO, USA) blended with protease inhibitors (Protease Inhibitor Coctail; Sigma, St. Louis, MO, USA) (dilution 1:200 in RIPA buffer) for 20 min at 4 °C. Upon intensive vortexing and centrifugation at 1000× g for 5 min at 4 °C, the cell lysates and culture media were collected, frozen at −70 °C , and then used for Western blot analyses and ELISA tests. To assess the protein concentration in the cell lysates, a BCA Protein Assay Kit (Pierce, Rockford, IL, USA) was used. For real-time PCR assay, the monolayers were washed three times with 10 mM PBS. Then the cells were collected into sterile plastic tubes and disrupted by a sonicator (Sonics Vibra cell; Sonics & Materials, Leicestershire, UK) (10 W, three times for 15 s on ice). For RNA isolation, aliquots of the homogenate were used. Cell Viability Assay The measurements of the viability of the cultured cells were performed in accordance with Carmichael et al. [19], with 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma, St. Louis, MO, USA). Cells were cultured on six-well plates, and after obtaining 70% of confluency, they were incubated for 24 h with 2-100 μM concentrations of PtPz4, PtPz6, and cisPt. After that, the cells were incubated for 4 h in MTT solution (0.5 mg MTT/mL of PBS) in a 5% CO2 atmosphere at 37 °C . The absorbance of the transformed dye in living cells was measured at 570 nm. The viability of AGS cells in the presence of studied compounds was calculated as a percentage of control cells (with 100% cell viability). Cell viability determination for anti-MUC1 was published previously in the work of Radziejewska et al. [20]. Annexin V Binding Assessment To evaluate the cell death induced by the examined compounds, Apoptosis Detection Kit II (BD Pharmingen, San Diego, Ca, USA) was used according to the manufacturer's instructions. Shortly after cells were trypsinized, they were resuspended in F-12 medium and then in binding buffer. Subsequently, they were stained with FITC Annexin V and Cell Viability Assay The measurements of the viability of the cultured cells were performed in accordance with Carmichael et al. [19], with 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma, St. Louis, MO, USA). Cells were cultured on six-well plates, and after obtaining 70% of confluency, they were incubated for 24 h with 2-100 µM concentrations of PtPz4, PtPz6, and cisPt. After that, the cells were incubated for 4 h in MTT solution (0.5 mg MTT/mL of PBS) in a 5% CO 2 atmosphere at 37 • C. The absorbance of the transformed dye in living cells was measured at 570 nm. The viability of AGS cells in the presence of studied compounds was calculated as a percentage of control cells (with 100% cell viability). Cell viability determination for anti-MUC1 was published previously in the work of Radziejewska et al. [20]. Annexin V Binding Assessment To evaluate the cell death induced by the examined compounds, Apoptosis Detection Kit II (BD Pharmingen, San Diego, CA, USA) was used according to the manufacturer's instructions. Shortly after cells were trypsinized, they were resuspended in F-12 medium and then in binding buffer. Subsequently, they were stained with FITC Annexin V and propidium iodide (PI) at RT (room temperature) in the dark (15 min). Annexin V bound with high affinity to phosphatidylserine was used to identify cells in the early stages of apoptosis, and PI was applied to determine late apoptotic and dead cells. Cells cultured in a drug-free medium were used as a negative control. The obtained results were evaluated on FACSCanto II cytometer (BD Biosciences, San Jose, CA, USA) with FACSDiva software (BD Biosciences, San Jose, CA, USA). Cells incubated with 3% formaldehyde in buffer for 30 min on ice were considered as positive control and applied to find the optimal parameter settings. RNA Isolation and Quantitative Real-Time PCR Total RNA was extracted using Total RNA Mini Plus Concentrator (A&A Biotechnology, Gdynia, Poland) in accordance with the manufacturer's guidelines. Isolated RNA concentration was evaluated using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA). Equal amounts (1 µg) of total RNA were used as the template for complementary DNA (cDNA) synthesis using SensiFAST TM cDNA Synthesis Kit (Bioline, London, UK). The reaction mixture (20 µL) containing 4 µL of 5× TransAmp Buffer, 1 µL of Reverse Transcriptase, RNA template, and DEPC-treated water was incubated for 10 min at 25 • C, 30 min at 45 • C and 5 min at 70 • C. Real-time PCR was performed with CFX96 detection system (Bio-Rad, Hercules, CA, USA) using SensiFAST TM SYBR Kit (Bioline, London, UK). 10 µL of reaction mixtures contained SensiFAST SYBR No-ROX Mix (5 µL), 3-times diluted cDNA template (2 µL), 0.4 µL (10 µM) of each target-specific primer purchased from Genomed (Warsaw, Poland) and DEPC-treated water. Primers for the experiment (forward and reverse) are listed in Table 1. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a control gene. The cDNA amplification conditions were as follows: 95 • C for 2 min for DNA polymerase activation and 40 cycles consisted of 95 • C for 10 s (denaturation), 60 • C for 15 s (annealing), and 72 • C for 20 s (elongation). Then, the melting curve was run to evaluate the specificity of each amplification. To confirm single product formation, melting point analysis and agarose gel electrophoresis were performed. Water, instead of mRNA samples, was used as a negative control. Each sample was analyzed in triplicate. Relative changes in mRNA expression were normalized with GAPDH and calculated using the delta-delta cycle threshold (∆∆Ct) method. Western Blot Analysis The samples prepared for electrophoresis were mixed with probe sample buffer (4:1) containing 2.5% sodium dodecyl sulfate (SDS) (Sigma, St. Louis, MO, USA). The same amount of proteins (20 µg) of cell lysates were separated on 7.5-13% polyacrylamide gels (concentration of the gel depending on the molecular weight of tested proteins) and immunoblotted on an Immobilon-P transfer membrane (Millipore, Bedford, MA, USA) according to Towbin et al. [21]. Next, the membranes were blocked with 5% non-fat milk in Tris-buffered saline (50 mM Tris-HCl, pH 7.5, 150 mM NaCl) supplemented with 0.05% Tween 20 (Sigma, St. Louis, MO, USA) (TBS-T) for 1 h at RT. Then, the membranes were washed three times (with TBS-T) and incubated overnight at 4 • C with specific monoclonal antibodies (listed in Table 2) diluted 1:600 in TBS-T containing 3% bovine serum albumin (BSA). To detect immunoreactive complexes, proper secondary horseradish peroxidase-conjugated antibodies diluted 1:2500 in TBS-T containing 5% non-fat milk were applied. To eliminate non-specific bindings, TBS-T buffer instead of monoclonal antibodies was used as a negative control. In certain analyses, membranes were stripped using a Restore Western Blot Stripping Buffer (Thermo Fisher Scientific, Waltham, MA, USA) and reused. Bound antibodies were visualized by enhanced chemiluminescence with Westar Supernova, a chemiluminescent substrate for Western blotting (Cyangen, Bologna, Italy). Band intensity was quantified by densitometry with an imaging densitometer (G:BOX, Syngene, Cambridge, UK) and normalized for β-actin. ELISA Tests To assess the relative level of Tn, T, sialyl Tn and sialyl T carbohydrate antigens, as well as MUC1 in cell lysates and culture media, ELISA tests were applied. To determine carbohydrate antigens, the following biotinylated lectins (Vector, Burlingame, CA, USA) at concentrations 5 µg/mL were used: VVA (from Vicia villosa) with affinity to Tn antigen (GalNAcα1-O-Ser/Thr), PNA (from Arachis hypogaea (peanut)) with affinity to T antigen (Galβ1-3GalNAcα1-O-Ser/Thr), SNA (from Sambucus nigra) with affinity to sialyl Tn (SAα2-6Gal/GalNAc), and MAAII (from Maackia amurensis) with binding preference to sialyl T (SAα2-3Gal). To detect MUC1 mucin, the anti-MUC1 monoclonal antibody ( Table 2) was applied. 50 µL of cell lysates or culture medium (100 µg protein/mL) were coated on microtiter plates (NUNC F96; Maxisorp, Roskilde, Denmark) and incubated overnight (RT). Then, the wells were blocked 1 h at RT (with 100 µL of blocking ELISA reagent; Roche Diagnostics, Mannheim, Germany), washed three times (with 100 µL of washing buffer PBS, 0.05% Tween), and incubated with lectins or anti-MUC1 mAb 2 h at RT. In the next step, the plates were incubated 1 h at RT with 100 µL of horseradish peroxidase avidin D (Vector, Burlingame, CA, USA) for carbohydrate antigens detection or with secondary horseradish peroxidase-conjugated anti-mouse IgG antibody for MUC1 detection. The colored reaction was developed using 100 µL of ABTS (2,2 -azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (Sigma, St. Louis, MO, USA). Absorbance at a wavelength of 405 nm was read after 20-40 min. The samples were analyzed in triplicate. A 1% BSA solution instead of the samples and washing buffer instead of lectins and primary antibody were used as negative controls. Statistical Analysis The results are presented as mean ± standard deviation (SD) from at least 3 independent determinations. Statistical analyses were carried out using the Statistica package (StatSoft, Tulsa, OK, USA). To determine statistical differences, one-way ANOVA followed by Duncan's multiple range post hock test was applied; p < 0.05 was considered statistically significant. Cytotoxic Effects of PtPz4, PtPz6, cisPt, and Anti-MUC1 To assess the effects of PtPz4, PtPz6, and cisplatin on the viability of gastric cancer cells CRL-1739, a MTT assay was applied ( Figure 2). The IC50 values for PtPz4 was 14 µM, for PtPz6 24.5 µM, for cisPt > 100 µM after 24 h of incubation. In all experiments performed in the study, we used 10 µM concentrations of PtPz4, PtPz6, and cisPt. IC50 for anti-MUC1was determined previously as 7 µg/mL [20]. In the present study, mAb with 5 µg/mL was applied. Statistical Analysis The results are presented as mean ± standard deviation (SD) from at least 3 independent determinations. Statistical analyses were carried out using the Statistica package (StatSoft, Tulsa, OK, USA). To determine statistical differences, one-way ANOVA followed by Duncan's multiple range post hock test was applied; p ˂ 0.05 was considered statistically significant. Cytotoxic Effects of PtPz4, PtPz6, cisPt, and Anti-MUC1 To assess the effects of PtPz4, PtPz6, and cisplatin on the viability of gastric cancer cells CRL-1739, a MTT assay was applied ( Figure 2). The IC50 values for PtPz4 was 14 μM, for PtPz6 24.5 μM, for cisPt > 100 μM after 24 h of incubation. In all experiments performed in the study, we used 10 μM concentrations of PtPz4, PtPz6, and cisPt. IC50 for anti-MUC1was determined previously as 7 μg/mL [20]. In the present study, mAb with 5 μg/mL was applied. Inhibitory Effect of Anti-MUC1 Monoclonal Antibody on MUC1 RT-PCR and ELISA tests were used to assess the effect of anti-MUC1 mAb action on MUC1 mucin expression in gastric cancer cells. Following 24 h of incubation of gastric cancer cells with monoclonal antibody (5 μg/mL) we observed a significant decrease in MUC1 mRNA expression compared to control ( Figure 3A). The inhibition of the mucin expression was also observed in cell lysates ( Figure 3B) and culture medium ( Figure 3C), with the highest significant decrease in MUC1 released to the culture medium. Inhibitory Effect of Anti-MUC1 Monoclonal Antibody on MUC1 RT-PCR and ELISA tests were used to assess the effect of anti-MUC1 mAb action on MUC1 mucin expression in gastric cancer cells. Following 24 h of incubation of gastric cancer cells with monoclonal antibody (5 µg/mL) we observed a significant decrease in MUC1 mRNA expression compared to control ( Figure 3A). The inhibition of the mucin expression was also observed in cell lysates ( Figure 3B) and culture medium ( Figure 3C), with the highest significant decrease in MUC1 released to the culture medium. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on the MUC1 Cytoplasmic Domain Western blotting was applied to investigate the effects of examined compounds and mAb on the MUC1 cytoplasmic tail expression. The cells were incubated for 24 h with PtPz4, PtPz6, cisplatin, anti-MUC1 used in monotherapy and combined (PtPz4 + anti-MUC1, PtPz6 + anti-MUC1, cisPt + anti-MUC1). We found that cisplatin used alone and in combination with anti-MUC1 significantly increased mucin cytoplasmic domain expression ( Figure 4). After combined cisPt and anti-MUC1 treatment lower stimulating effect was observed. Anti-MUC1 used in monotherapy and in combination with PtPz4 and PtPz6 reduced MUC1 cytoplasmic tail expression. Impact of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Apoptosis Annexin-V and PI staining was used to assess the influence of examined compounds on apoptosis in gastric cancer cells ( Figure 5). Annexin V allowed establishing cells in the early stages of the programmed cell death, while propidium iodide identified late apoptotic and dead cells. Analysis revealed that both platinum complexes used in the study did not significantly induce an apoptotic response in comparison to control. In cisplatin-treated cells, there were 16.9% apoptotic cells vs. 10.6% in control cells. After incubation with anti-MUC1, 15.14% of apoptotic cells were observed. The strongest pro-apoptotic effect on gastric cancer cells was revealed by combined treatment of PtPz6 and anti-MUC1 (26.8% of apoptotic cells). PtPz4 and cisPt combined with mAb also exhibited pro-apoptotic potential; there were 18 and 20.05% of apoptotic cells, respectively. 3.5. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on NF-κB mRNA NF-κB signaling is said to be critical in cancer development and can be deregulated in gastric cancer. RT-PCR determinations revealed that PtPz6, cisPt, and anti-MUC1 used alone as well as cisPt administrated with mAb significantly decreased NF-κB mRNA in comparison to untreated control ( Figure 6). However, such inhibitory effects were not observed after monotherapy with PtPz4 and after combined action of anti-MUC1, PtPz4, and PtPz6. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on AKT mRNA Akt controls the survival of the cells and inhibits apoptosis by phosphoryla cific components participating in the cell death program. RT-PCR results showed t PtPz6 monotherapy and this compound combined with anti-MUC1 significan pressed Akt mRNA expression in comparison to control (Figure 7). There was no cally significant impact of other tested variants of therapy on the Akt signaling p The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Pro-Apoptotic Factors To investigate if examined compounds initiate apoptosis by activation of pr totic factors, Bax mRNA and Bax protein, as well as Bad, Bid, and Bim mRNAs w termined. We found that Bax gene expression was significantly suppressed by PtP otherapy, and the rest of the examined compounds used in monotherapy and co with anti-MUC1 had no significant impact on Bax mRNA. These results were n lated with Bax protein expression, analyzed by Western blotting. We found tha The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on AKT mRNA Akt controls the survival of the cells and inhibits apoptosis by phosphorylat cific components participating in the cell death program. RT-PCR results showed t PtPz6 monotherapy and this compound combined with anti-MUC1 significan pressed Akt mRNA expression in comparison to control (Figure 7). There was no cally significant impact of other tested variants of therapy on the Akt signaling p The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Pro-Apoptotic Factors To investigate if examined compounds initiate apoptosis by activation of pr totic factors, Bax mRNA and Bax protein, as well as Bad, Bid, and Bim mRNAs w termined. We found that Bax gene expression was significantly suppressed by PtP otherapy, and the rest of the examined compounds used in monotherapy and co with anti-MUC1 had no significant impact on Bax mRNA. These results were n lated with Bax protein expression, analyzed by Western blotting. We found tha cisPt, and anti-MUC1 when used alone, and PtPz4 and cisPt combined with mAb cantly stimulated Bax protein expression ( Figure 8A). As shown in Figure 8B, Bad To investigate if examined compounds initiate apoptosis by activation of pro-apoptotic factors, Bax mRNA and Bax protein, as well as Bad, Bid, and Bim mRNAs were determined. We found that Bax gene expression was significantly suppressed by PtPz6 monotherapy, and the rest of the examined compounds used in monotherapy and combined with anti-MUC1 had no significant impact on Bax mRNA. These results were not correlated with Bax protein expression, analyzed by Western blotting. We found that PtPz6, cisPt, and anti-MUC1 when used alone, and PtPz4 and cisPt combined with mAb significantly stimulated Bax protein expression ( Figure 8A). As shown in Figure 8B, Bad mRNA increased as a result of PtPz4 action, and this expression was enhanced by anti-MUC1 addition. Significant stimulation of Bad mRNA was also observed after PtPz6 combined with anti-MUC1. Bid mRNA expression was significantly inhibited by PtPz6 when used alone and stimulated by PtPz4 combined with anti-MUC1 ( Figure 8C). PtPz6 and cisPt in monotherapy and PtPz6 together with mAb significantly stimulated Bim mRNA expression ( Figure 8D). However, the addition of anti-MUC1 to PtPz6 did not enhance the stimulatory effect in comparison with PtPz6 monotherapy. Pharmaceutics 2021, 13, 968 10 monotherapy and PtPz6 together with mAb significantly stimulated Bim mRNA exp sion ( Figure 8D). However, the addition of anti-MUC1 to PtPz6 did not enhance the s ulatory effect in comparison with PtPz6 monotherapy. The intensities of the bands were quantified by densitometric a ysis. Data represent the mean ± SD of the relative expression levels (from three assays) normal to β-actin. * p < 0.05; p < 0.01; * p < 0.001. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Anti-Apoptotic Factors Bcl-xL and Bcl-2 are considered crucial pro-survival proteins. To analyze the effec cisPt, its derivatives, and anti-MUC1 on Bcl-xL, RT-PCR and Western blotting were plied. We found that all the examined compounds significantly inhibited Bcl-xL mR expression, and the effect was enhanced after combining PtPz4 and PtPz6 with anti-MU , anti-MUC1 (5 µg/mL), PtPz4 + anti-MUC1 (10 µM + 5 µg/mL), PtPz6 + anti-MUC1 (10 µM + 5 µg/mL), and cisPt + anti-MUC1 (10 µM + 5 µg/mL). mRNAs were assessed by RT-PCR. The result is shown as a relative fold change in mRNA gene expression in comparison of gene in control where expression was set at 1 ± SD are the mean of triplicate cultures. * p < 0.05; p < 0.01; * p < 0.001. Expression of Bax protein in cell lysates was determined by Western blotting. The intensities of the bands were quantified by densitometric analysis. Data represent the mean ± SD of the relative expression levels (from three assays) normalized to β-actin. * p < 0.05; p < 0.01; * p < 0.001. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Anti-Apoptotic Factors Bcl-xL and Bcl-2 are considered crucial pro-survival proteins. To analyze the effect of cisPt, its derivatives, and anti-MUC1 on Bcl-xL, RT-PCR and Western blotting were applied. We found that all the examined compounds significantly inhibited Bcl-xL mRNA expression, and the effect was enhanced after combining PtPz4 and PtPz6 with anti-MUC1 in comparison with monotherapy ( Figure 9A). Such enhancement was not observed after joining cisPt with anti-MUC1. Significant suppression of Bcl-xL protein was observed only after PtPz6 and PtPz4 combined with anti-MUC1. We revealed that Bcl-2 mRNA was significantly inhibited by cisPt, anti-MUC1, and PtPz4 and PtPz6 and cisPt in combination with mAb ( Figure 9B). A large stimulation of such inhibition after anti-MUC1 addition was observed. ting. The intensities of the bands were quantified by densitometric analysis. Data represent the m  SD of the relative expression levels (from three assays) normalized to β-actin. * p < 0.05; p < 0 The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Caspases Expression of initiatory caspase-9 and executive caspase-3 were assessed in our stu The results presented in Figure 10A revealed that PtPz4, PtPz6, cisPt, and anti-MUC1 nificantly increased caspase-9 mRNA expression in comparison to untreated contr Such results were also observed for PtPz4 and PtPz6 combined with mAb; however, w out the stimulatory effect of anti-MUC1 in comparison with monotherapy. Surprisin such stimulatory effect was not observed as an effect of anti-MUC1 combined with c and its derivatives on caspase-9 protein expression ( Figure 10B). Anti-MUC1 alone combined with PtPz4, PtPz6, and cisPt significantly inhibited pro-caspase-9 express Pro-caspase-9 was stimulated only by PtPz4 monotherapy. Cleavage of caspase-9 was s pressed by both cisPt derivatives, anti-MUC1 and mAb combined with PtPz4. There w no effect of cisPt, anti-MUC1 combined with PtPz6 and cisPt on cleaved caspas Caspase-3 mRNA expression significantly increased as a result of anti-MUC1 action after combined treatment of mAb with cisPt and its derivatives ( Figure 10C). Pro-caspa , anti-MUC1 (5 µg/mL), PtPz4 + anti-MUC1 (10 µM + 5 µg/mL), PtPz6 + anti-MUC1 (10 µM + 5 µg/mL), and cisPt + anti-MUC1 (10 µM + 5 µg/mL). mRNAs were assessed by RT-PCR. The result is presented as a relative fold change in mRNA gene expression in comparison of gene in control (where expression was set at 1) ± SD are the mean of triplicate cultures. p < 0.01; *** p < 0.001. The expression of Bcl-xL protein in cell lysates was determined by Western blotting. The intensities of the bands were quantified by densitometric analysis. Data represent the mean ± SD of the relative expression levels (from three assays) normalized to β-actin. * p < 0.05; ** p < 0.01. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Caspases Expression of initiatory caspase-9 and executive caspase-3 were assessed in our study. The results presented in Figure 10A revealed that PtPz4, PtPz6, cisPt, and anti-MUC1 significantly increased caspase-9 mRNA expression in comparison to untreated controls. Such results were also observed for PtPz4 and PtPz6 combined with mAb; however, without the stimulatory effect of anti-MUC1 in comparison with monotherapy. Surprisingly, such stimulatory effect was not observed as an effect of anti-MUC1 combined with cisPt and its derivatives on caspase-9 protein expression ( Figure 10B). Anti-MUC1 alone and combined with PtPz4, PtPz6, and cisPt significantly inhibited pro-caspase-9 expression. Pro-caspase-9 was stimulated only by PtPz4 monotherapy. Cleavage of caspase-9 was suppressed by both cisPt derivatives, anti-MUC1 and mAb combined with PtPz4. There was no effect of cisPt, anti-MUC1 combined with PtPz6 and cisPt on cleaved caspase-9. Caspase-3 mRNA expression significantly increased as a result of anti-MUC1 action and after combined treatment of mAb with cisPt and its derivatives ( Figure 10C). Pro-caspase-3 expression significantly increased after monotherapy with PtPz6, cisPt, and anti-MUC1 compared to control ( Figure 10D). Such effect was not observed after PtPz4 used alone. The addition of mAb to cisPt and its derivatives enhanced the noticed effect. All the examined compounds stimulated cleaved caspase-3 expression with intensified outcome after combined therapy. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Cancer Related Carbohydrate Antigens Expression Finally, using ELISA tests, we checked the expression of cancer-related sugar antigens Tn, T, sialyl Tn, and sialyl T in cell lysates and culture medium after incubation of cancer cells with examined drugs. Reaction with VVA lectin allowed the dectection of the PtPz4 + anti-MUC1 (10 µM + 5 µg/mL), PtPz6 + anti-MUC1 (10 µM + 5 µg/mL), and cisPt + anti-MUC1 (10 µM + 5 µg/mL). mRNAs were assessed by RT-PCR. The result is presented as a relative fold change in mRNA gene expression in comparison of gene in control where expression was set at 1 ± SD are the mean of triplicate cultures. * p < 0.05; p < 0.01; * p < 0.001. Pro-and cleaved caspase expressions in cell lysates were assessed by Western blotting. The intensities of the bands were quantified by densitometric analysis. Data represent the mean ± SD of the relative expression levels (from three assays) normalized to β-actin. * p < 0.05; p < 0.01; * p < 0.001. The Effect of PtPz4, PtPz6, cisPt, and Anti-MUC1 on Cancer Related Carbohydrate Antigens Expression Finally, using ELISA tests, we checked the expression of cancer-related sugar antigens Tn, T, sialyl Tn, and sialyl T in cell lysates and culture medium after incubation of cancer cells with examined drugs. Reaction with VVA lectin allowed the dectection of the Tn antigen. We found that both cisPt derivatives as well as anti-MUC1, PtPz4, PtPz6, and cisPt combined with mAb significantly decreased Tn antigen expression in culture medium compared to untreated controls ( Figure 11A). All the drugs did not affect the expression of this antigen in cell lysates. T antigen, detected after reaction with PNA lectin was significantly reduced in cell lysates by PtPz4 and anti-MUC1 action as well as by combined therapy of mAb with PtPz4, PtPz6, and cisPt ( Figure 11B). CisPt induced this antigen expression in culture medium, and PtPz6, anti-MUC1 as well as PtPz4, PtPz6, and cisPt combined with anti-MUC1 reduced T antigen expression. The inhibitory effect was intensified after combined treatment. Sialylated form of Tn antigen expression was determined using SNA lectin. We revealed inhibition of sialyl Tn antigen expression in cell lysates after combined action of anti-MUC1 with PtPz4 and PtPz6 ( Figure 11C). Monotherapy with cisPt and anti-MUC1 inhibited this antigen expression in the culture medium compared to controls. The addition of anti-MUC1 to cisPt and its derivatives markedly intensified such an inhibitory effect. MAAII lectin was used to detect sialyl T antigen expression. We found that this antigen expression in cell lysates was suppressed only by combined treatment of anti-MUC1 and PtPz6 ( Figure 11D). In culture medium, a significant decrease in sialyl T antigen was revealed after PtPz4, anti-MUC1 monotherapy, as well as a result of the combined action of the drugs with anti-MUC1. Tn antigen. We found that both cisPt derivatives as well as anti-MUC1, PtPz4, PtPz6, and cisPt combined with mAb significantly decreased Tn antigen expression in culture medium compared to untreated controls ( Figure 11A). All the drugs did not affect the expression of this antigen in cell lysates. T antigen, detected after reaction with PNA lectin was significantly reduced in cell lysates by PtPz4 and anti-MUC1 action as well as by combined therapy of mAb with PtPz4, PtPz6, and cisPt ( Figure 11B). CisPt induced this antigen expression in culture medium, and PtPz6, anti-MUC1 as well as PtPz4, PtPz6, and cisPt combined with anti-MUC1 reduced T antigen expression. The inhibitory effect was intensified after combined treatment. Sialylated form of Tn antigen expression was determined using SNA lectin. We revealed inhibition of sialyl Tn antigen expression in cell lysates after combined action of anti-MUC1 with PtPz4 and PtPz6 ( Figure 11C). Monotherapy with cisPt and anti-MUC1 inhibited this antigen expression in the culture medium compared to controls. The addition of anti-MUC1 to cisPt and its derivatives markedly intensified such an inhibitory effect. MAAII lectin was used to detect sialyl T antigen expression. We found that this antigen expression in cell lysates was suppressed only by combined treatment of anti-MUC1 and PtPz6 ( Figure 11D). In culture medium, a significant decrease in sialyl T antigen was revealed after PtPz4, anti-MUC1 monotherapy, as well as a result of the combined action of the drugs with anti-MUC1. Discussion Cisplatin is one of the best and first metal-based chemotherapeutics [22,23]. It is active against a variety of solid cancers such as testicular, ovarian, bladder, lung, cervical, gastric, head and neck cancer, and some others [4,6]. However, due to severe side effects and acquired drug resistance, its clinical application can be limited. Therefore, much effort has been put into searching for new cisplatin-based anti-cancer complexes [6]. Recently, new pyrazole-platinum(II) complexes have been applied as promising anti-cancer agents leading to cell cycle arrest and apoptosis in MCF-7 and MDA-MB-231 breast cancer cells [7]. Moreover, combined therapies of cisplatin or its complexes with other drugs have been revealed to acquire better effectiveness, conquer drug resistance and lower undesirable toxicity [4]. As was mentioned in the Introduction section, tumor-associated MUC1 plays a crucial role in the development of many cancers. It has been reported to inhibit cell death and promote cell differentiation, proliferation, invasiveness, and metastasis [8]. Thus, inhibition of MUC1 activities could likely be an advantageous strategy in cancer therapy [24]. Moreover, its high expression, differential distribution pattern relative to normal cells, and altered architecture in malignant tissues made this molecule a top molecular target in anti-cancer treatment [13,14,[25][26][27]. Antibodies have been confirmed to be compelling additions to therapy for many types of cancer [28]. A number of anti-MUC1 monoclonal antibodies have been described in the literature, and some of them have been involved in pre-clinical studies [13,24,29,30]. In this study, we demonstrated the effects of combined action of an anti-MUC1 monoclonal antibody with cisplatin and two pyrazole-platinum complexes on apoptosis and cell death-related factors in CRL-1739 gastric cancer cells. We applied IgG class anti-MUC1 (BC2) monoclonal antibody reacting with a five amino acid epitope of the MUC1 core protein with low susceptibility for glycosylation. Hisatsune et al. [31] reported that MUC1 mucin on the cell surface could be internalized by anti-MUC1 antibody through the macropinocytic pathway. It has been also demonstrated that antibodies, by direct binding to their receptors, are able to inhibit their activity, resulting in the inhibition of signaling cascades that promote cell cycle and function [28]. In our research, we showed that 24 h incubation of AGS cancer cells with anti-MUC1 monoclonal antibody resulted in reduced expression of MUC1 gene as well as mucin expression in cell lysates and culture medium. Upon such results, we assumed that anti-MUC1 mAb potentially inhibited the activity and function of MUC1. Such a conclusion was supported by revealing an inhibitory effect of mAb on the MUC1 cytoplasmic tail. This part of MUC1 was reported to be involved in intracellular signal transduction Discussion Cisplatin is one of the best and first metal-based chemotherapeutics [22,23]. It is active against a variety of solid cancers such as testicular, ovarian, bladder, lung, cervical, gastric, head and neck cancer, and some others [4,6]. However, due to severe side effects and acquired drug resistance, its clinical application can be limited. Therefore, much effort has been put into searching for new cisplatin-based anti-cancer complexes [6]. Recently, new pyrazole-platinum(II) complexes have been applied as promising anti-cancer agents leading to cell cycle arrest and apoptosis in MCF-7 and MDA-MB-231 breast cancer cells [7]. Moreover, combined therapies of cisplatin or its complexes with other drugs have been revealed to acquire better effectiveness, conquer drug resistance and lower undesirable toxicity [4]. As was mentioned in the Introduction section, tumor-associated MUC1 plays a crucial role in the development of many cancers. It has been reported to inhibit cell death and promote cell differentiation, proliferation, invasiveness, and metastasis [8]. Thus, inhibition of MUC1 activities could likely be an advantageous strategy in cancer therapy [24]. Moreover, its high expression, differential distribution pattern relative to normal cells, and altered architecture in malignant tissues made this molecule a top molecular target in anti-cancer treatment [13,14,[25][26][27]. Antibodies have been confirmed to be compelling additions to therapy for many types of cancer [28]. A number of anti-MUC1 monoclonal antibodies have been described in the literature, and some of them have been involved in pre-clinical studies [13,24,29,30]. In this study, we demonstrated the effects of combined action of an anti-MUC1 monoclonal antibody with cisplatin and two pyrazole-platinum complexes on apoptosis and cell death-related factors in CRL-1739 gastric cancer cells. We applied IgG class anti-MUC1 (BC2) monoclonal antibody reacting with a five amino acid epitope of the MUC1 core protein with low susceptibility for glycosylation. Hisatsune et al. [31] reported that MUC1 mucin on the cell surface could be internalized by anti-MUC1 antibody through the macropinocytic pathway. It has been also demonstrated that antibodies, by direct binding to their receptors, are able to inhibit their activity, resulting in the inhibition of signaling cascades that promote cell cycle and function [28]. In our research, we showed that 24 h incubation of AGS cancer cells with anti-MUC1 monoclonal antibody resulted in reduced expression of MUC1 gene as well as mucin expression in cell lysates and culture medium. Upon such results, we assumed that anti-MUC1 mAb potentially inhibited the activity and function of MUC1. Such a conclusion was supported by revealing an inhibitory effect of mAb on the MUC1 cytoplasmic tail. This part of MUC1 was reported to be involved in intracellular signal transduction through the interactions with several signaling molecules implicated in the cancer regulation by affecting the proliferation, apoptosis, and transcription of various genes [32,33]. Thus, we can speculate that revealed in our study, MUC1 mRNA inhibition triggered by anti-MUC1 action could occur via the MUC1 cytoplasmic tail. Combined administration of anti-MUC1 and two cisPt derivatives only slightly enhanced the inhibitory effect towards this part of MUC1. Surprisingly, cisplatin used alone and combined with mAb stimulated MUC1 cytoplasmic tail expression. Cisplatin complexes used in our study were PtPz4 [Pt 2 (pyrazole) 4 (berenil) 2 ]·4HCl·2H 2 O and PtPz6 [Pt 2 (N-methylpyrazole) 4 (berenil) 2 ]·4HCl·2H 2 O (Figure 1), the first one with an unsubstituted pyrazole ring, and the second with an ethyl group. These two complexes, as well as four others, with different substituents, were applied by Czarnomysy et al. [7] as anti-cancer agents in MCF-7 and MDA-MB-231 breast cancer cells. They revealed that the activity of the examined complexes depends on the type of pyrazole ligand. It is said that the general action of cisplatin and its complexes are based on interacting with genetic material what in consequence leads to apoptosis [2,3,7]. In our study, we revealed that PtPz6 administrated with anti-MUC1 the most strongly induced apoptosis, as compared with controls, cisPt, anti-MUC1 as well as with both cisPt complexes used alone and PtPz4 combined with mAb. Such a result can suggest that anti-MUC1 induced pro-apoptotic activity of cisPt complex with pyrazole substituted with an ethyl group. Evasion from apoptosis seems to be one of the most important hallmarks of cancer. NF-κB and Akt signaling pathways are among the key factors regulating the process of programmed cell death [24,34]. Thus, blocking of such factors activation is considered to be a highly rational direction in cancer therapy. It has been stated that NF-κB, which is generally regarded as a pro-survival agent, is one of the transcription factors which can be deregulated in gastric cancer. Many of the genes transcribed by NF-κB promote carcinogenesis [35]. However, the correlation between its activity and clinicopathological features is still not fully clarified. It has been also demonstrated that chemotherapy itself could evoke NF-κB activation in gastric cell lines leading to the achievement of chemoresistance [34]. Our results, in relation to NF-κB mRNA expression in AGS cancer cells, support the rationality of cisPt usage, as well as PtPz6 and anti-MUC1, as the factor was suppressed by such treatment. In human cancers, Akt is considered a significant mediator of the cell cycle, promoting cancer cell proliferation as well as contributing to resistance to apoptosis [36]. Akt action is based on the regulation of activity of several components participating in cell death, such as pro-apoptotic Bad, Bax, Bim, or caspase-9. Phosphorylation of Bad induced by Akt inhibits its heterodimerization with pro-survival Bcl-xL leading to restoration of Bcl-xL anti-apoptotic function [37]. Therefore, as this factor is usually highly activated in malignant cells, the blocking of its expression seems to be a rational target in cancer therapy. Moreover, numerous studies have clearly reported that activation of the Akt signaling pathway could regulate Bim expression. Following pro-apoptotic events, Bim translocates to the mitochondria, where it is essential to mediate the release of cytochrome c, which in turn activates caspase-9 [38]. Additionally, Bim upregulation has been reported to affect the chemotherapy response [39]. Therefore, the combined action of PtPz6 and anti-MUC1 showing inhibitory effect toward Akt mRNA revealed in our study seems to be an effective therapy, promoting activation of apoptosis. Such conclusion can be supported by our outcomes revealing that in the case of PtPz6 administrated with mAb, inhibition of Akt gene correlated with Bad and Bim mRNA stimulation as well as Bcl-xL mRNA inhibition. Bax is the next pro-apoptotic member regulating apoptosis. Its reduction plays a key role in tumorigenesis. It promotes cell death mainly through mitochondrial membrane permeabilization. On the contrary, pro-survival Bcl-2 impedes apoptosis by suppressing the activity of Bax [40,41]. There have been reports considering Bax/Bcl-2 ratio as a factor determining cell susceptibility to apoptosis. The authors suggested that a lower level of such a ratio may lead to resistance of cancer cells to apoptosis [42]. Our results do not support the presented idea. We revealed no significant stimulatory effect of applied treatment of Bax mRNA. Surprisingly PtPz6, cisPt, and anti-MUC1 monotherapy, as well as combined treatment of PtPz4 and cisPt with mAb, significantly stimulated Bax on protein level. We suggest that such lack of correlation between Bax mRNA and protein can be explained, e.g., by negative feedback-a high level of protein can downregulate mRNA expression. However, the inhibitory effect of monotherapy with cisPt and mAb as well as anti-MUC1 combined with PtPz4, PtPz6, and cisPt on Bcl-2 mRNA support rationality of applied therapy in anti-cancer treatment. Pro-apoptotic Bid has been reported as one more factor with a major role in gastric cancer development [43]. Our outcome, revealing stimulation of Bid mRNA expression by PtPz4 administrated with anti-MUC1 gives one more proof of rationality of anti-MUC1 application. Caspases are a family of proteases that play essential roles in the initiation (e.g., caspase-8,-9) and execution (e.g., caspase-3) of apoptosis [44]. Thus, it is generally stated that they promote cell death, and their loss is connected to cancer development. However, it was documented that caspases can also be involved in tumorigenesis through a non-apoptotic pathway by influencing proliferation, invasion, and migration [45,46]. Commonly, the initiatory caspase zymogens are transformed into the active proteases by dimerization-induced conformational changes leading to the formation of active, cleaved forms. On the contrary, executioner caspases require cleavage by initiatory caspases for their activation [47]. Upon the results of the presented work, we observed that combining anti-MUC1 with both cisPt derivatives induced apoptosis by enhancement the expression of caspase-9 and -3 mRNAs. Such effect was also observed for the action of cisPt and mAb towards casp-3, but not casp-9. Surprisingly caspase-9 mRNA expression did not correlate with caspase-9 protein. We can assume that caspase-9 underwent such posttranslational modifications, which led to the decreased expression of the protein. We may also suggest too low sensitivity of Western blotting, below that of RT-PCR. Nevertheless, the expression of both forms of caspase-3 protein, pro-and cleaved, increased with higher intensity after combined therapy. Our result, revealing increased caspase-3 expression after applied therapy in comparison to untreated control, points to leading cancer cells to apoptosis. Interestingly, Huang et al. [48] have reported a higher level of caspase-3 protein in untreated malignant, comparing to nonmalignant breast tissue. It was explained by the activation of caspase-3 in dying cancer cells what resulted in tumor cell repopulation and further growth. On many epithelial-origin cancer cells, MUC1 is enriched by specific tumor-associated carbohydrate antigens (TACAs), including Tn, T, sTn, and sT sugar structures. They are overexpressed on the surface of cancer cells compared to those on normal tissues [49]. It has been reported that the presence of such truncated glycan structures plays a key role in tumor initiation, progression, and metastasis as TACAs are able to interact with many factors participating in signaling pathways promoting cancer development (leading to the creation of a pro-tumor microenvironment, favoring tumor progression and metastasis) [50]. TACAs expression levels have been used as biomarkers of poor prognosis [51]. Moreover, they have been recently proposed as a specific "glycol-code" that could be considered as a novel immune checkpoint, offering new immunotherapeutic opportunities [49,50]. In our study, we assumed that anti-MUC1 based therapy could also block MUC1 associated TACAs expression. Thus, our results revealing significant inhibition of Tn, T antigens, and their sialylated forms, enhanced by combined therapy, strongly support the rationality of potential anti-MUC1 application in gastric cancer treatment. Conclusions Summarizing our results, we suggest the application of two pyrazole-berenil cisPt complexes combined with anti-MUC1 monoclonal antibody as a promising strategy in gastric cancer treatment, more effective than monotherapy. Generally, the anti-MUC1 addition to cisplatin derivatives enhanced pro-apoptotic effects towards Bax, Bad, Bcl-xL Bcl-2, caspases, and TACAs in comparison to untreated control and monotherapy. We realize that not all our results seem to be consistent. Some of them need to be elucidated in the next experiments, planned to be continued in the subject. We are also aware of a limitation of the presented study concerning the usage of only one cancer cell line. In the future, we are going to expand our research by applying at least one more gastric cancer cell line.	2021-07-03T06:17:01.631Z	2021-06-26T00:00:00.000	{ "year": 2021, "sha1": "d40f783569cc7965afdfac73d6d4a88f44d2a245", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1999-4923/13/7/968/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "e0c36c866e134594563336aad4390ac468f0fe91", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
245663254	pes2o/s2orc	v3-fos-license	Counting fluorescently labeled proteins in tissues in the spinning–disk microscope using single–molecule calibrations Quantification of molecular numbers and concentrations in living cells is critical for testing models of complex biological phenomena. Counting molecules in cells requires estimation of the fluorescence intensity of single molecules, which is generally limited to imaging near cell surfaces, in isolated cells, or where motions are diffusive. To circumvent this difficulty, we have devised a calibration technique for spinning–disk confocal microscopy, commonly used for imaging in tissues, that uses single–step bleaching kinetics to estimate the single–fluorophore intensity. To cross–check our calibrations, we compared the brightness of fluorophores in the SDC microscope to those in the total internal reflection and epifluorescence microscopes. We applied this calibration method to quantify the number of end–binding protein 1 (EB1)–eGFP in the comets of growing microtubule ends and to measure the cytoplasmic concentration of EB1–eGFP in sensory neurons in fly larvae. These measurements allowed us to estimate the dissociation constant of EB1–eGFP from the microtubules as well as the GTP–tubulin cap size. Our results show the unexplored potential of single–molecule imaging using spinning–disk confocal microscopy and provide a straightforward method to count the absolute number of fluorophores in tissues that can be applied to a wide range of biological systems and imaging techniques. The absolute quantification of single-molecule fluorescence is challenging, however, especially in tissues. Several quantification approaches have been taken, but all have drawbacks. For example, the intensity of fluorescently labeled molecules (such as GFPtagged proteins) can be calibrated by quantitative immunoblotting (Rusan et al., 2001;Wu and Pollard, 2005;Seetapun et al., 2012). While this strategy is suitable for cultured cell lines or unicellular organisms, it cannot be applied to tissues due to the difficulty of isolating single cell types for immunoanalysis. A second approach is to visualize single fluorophores and to use single-step bleaching to estimate the single-molecule fluorescence (Leake et al., 2006;Engel et al., 2009;Coffman et al., 2011). A limitation is that it generally requires a combination of strong illumination and low fluorophore density, for example total internal reflection fluorescence (TIRF) imaging near the cell surface (Leake et al., 2006;Ulbrich and Isacoff, 2007), and is therefore not suitable for imaging deep in tissues or structures containing a larger number of fluorophores. A third approach is fluorescence correlation spectroscopy (FCS) (Magde et al., 1972;Wachsmuth et al., 2015) or number and brightness (N&B) (Cutrale et al., 2019), which uses the fluctuations of the fluorescence intensity of the fluorophore-tagged biomolecules in a small volume to obtain quantitative information such as their abundance and mobility. However, these methods are designed for diffusing molecules and are not suitable for molecules undergoing directed motion. Thus, these techniques are either difficult to apply in tissues directly or restricted to the quantification of biomolecules that are dilute or diffusing. In this work we visualized single fluorophores in a spinning-disk confocal (SDC) microscope, a commonly used instrument for livecell imaging in tissues, and used single-step photobleaching to calibrate fluorescence intensity. While visualization of single fluorophores has been achieved routinely using TIRF, laser-scanning confocal, and epifluorescence microscopies, single-fluorophore intensities in the SDC microscope reported in earlier studies have been too low to use as calibration standards (Lawrimore et al., 2011). Here, we compared the fluorescence intensities and photobleaching kinetics in a commercial SDC microscope to those in TIRF and epifluorescence microscopes, to identify the theoretical and practical limitations on the sensitivity of the SDC microscope. The singlefluorophore intensity measured by SDC microscopy serves as a direct calibration standard for fluorophore counting. As a proof of principle, we quantified the number of end-binding protein 1 (EB1)-eGFP in the EB1 comets and the concentration of free EB1-eGFP in the dendrites of Drosophila class IV dendritic arborization (da) sensory neurons. Our results provide an estimation of the GTP-cap size and the binding affinity of EB1-eGFP, crucial information for characterizing the dynamic and biochemical properties of neuronal microtubules (MTs). Single fluorophores can be observed by SDC microscopy To quantify molecular fluorescence in a commercial SDC microscope (Nikon Ti Elipse, Yokagawa spinning disk, 50 μm pinhole, 100 mW diode lasers), we measured the intensities of single fluorophores. Stabilized MTs labeled with a low density of Alexa Fluor 488 were affixed to the surface of the coverslip by anti-tubulin antibodies. We first calibrated molecular fluorescence by TIRF microscopy ( Figure 1A, inset), a common method for single-molecule imaging. The intensities of the fluorescent puncta usually decreased in a single step ( Figure 1A), showing that the majority of puncta corresponded to single-Alexa Fluor 488 dye molecules, as expected. The average intensity of single Alexa Fluor 488 was measured from the histogram of step sizes ( Figure 1B), and the bleaching rate was determined by fitting the bleaching time histogram to an exponential ( Figure 1C). Next, we imaged the MTs by SDC microscopy. We observed fluorescent puncta that resembled the single fluorophores seen by TIRF ( Figure 1D, inset). The intensity traces of individual puncta showed clear single-step bleaching events, confirming that the fluorescence signals originated from single fluorophores ( Figure 1D). The fluorophore intensity and bleaching rate were measured from the step-size and bleaching time histograms, respectively (Figure 1, E and F). These results demonstrated that single-molecule imaging by SDC microscopy is possible. The black trace shows the intensity as a function of time of single puncta imaged using a Nikon SDC microscope with Yokogawa disk (circled in the inset image). The red dashed line corresponds to a single step. Single-step bleaching can be clearly observed. (E) Histogram of step sizes. (F) Histogram of bleaching times. The sample irradiance was 320 and 290 kW/m 2 in TIRF and SDC, respectively. The images were acquired with 100 and 500 ms exposure times in TIRF and SDC, respectively. ADU: analogue-to-digital unit. Comparison of excitation intensities in the SDC, TIRF, and epifluorescence microscopes The fluorescence intensities of single fluorophores were dimmer in the SDC microscope compared with the TIRF microscope even when the camera exposure time was five times longer (under similar excitation laser power; Figure 1, B and E). This is due to differences in excitation intensities and emission detection efficiencies. To investigate the excitation intensity differences, we imaged MTs labeled with a high density of Alexa Fluor 488 in the TIRF, epifluorescence, and SDC microscopes. We estimated the intensities of the local excitation fields by measuring the fluorescence bleaching under the same illumination irradiances (the laser power out of the objective divided by the illuminated area). The decay of the fluorescence intensities was well described by single exponentials (see examples in Figure 2A), and the bleaching rates (obtained from the exponential fit) increased linearly with average illumination irradiance ( Figure 2B), as expected because we are well below the fluorescence saturation intensity (see below). Under the same illumination irradiances, the bleaching rate constant was 3.8 ± 0.2 (mean ± SE unless otherwise noted) times larger in TIRF compared with epifluorescence microscopy (measured from the slopes of the linear regression; Figure 2B and Table 1). The roughly four-times-higher bleaching rate in TIRF over epifluorescence accords with the approximately fourfold enhancement of the evanescence field intensity predicted theoretically (derived in the Supplemental Information, and see Martin-Fernandez et al., 2013). This is the first quantification of enhancement of the evanescence field intensity that we are aware of. The bleaching rate constants measured from the highly labeled MTs (TIRF: 5.7 ± 1.0 s; SDC: 24.0 ± 1.1 s; Table 2; details in Materials and Methods) were in good agreement with the bleaching rate constants obtained from the single-molecule bleaching times (TIRF: 5.3 ± 1.1 s; SDC: 24 ± 3 s). This is further confirmation of the reliability of our single-molecule SDC observations. Under the same illumination irradiances, the bleaching rate constants of SDC and epifluorescence were similar (ratio of slopes = FIGURE 2: Bleaching rate measurements in different imaging methods. (A) Example bleaching curves under different illumination irradiances in the SDC microscope. The initial segment of each bleaching curve (until ∼10% of initial intensity) was fitted to a single-exponential decay (solid lines). (B) Bleaching rate constants derived from the exponential fits increased linearly with irradiance. Inset: zoom-in of the bottom three lines. Solid lines: linear regressions constrained to pass through the origin. Error bars: SD from three experimental measurements in each condition. 0.96 ± 0.04) ( Figure 2B and Table 1). This shows that the excitation field intensities are similar in the two imaging modes when the irradiance is the same, as expected. In SDC, each point in the sample is illuminated intermittently due to the rotation of the spinning disk; therefore, only a small percentage of the total area is illuminated at any one time. To assess the effect of the disk rotation, we measured the bleaching rate in illuminated regions when the disk was stationary (i.e., the stop-disk condition). When the disk was stationary, the bleaching rate was 19 ± 1 times larger than in the spinning-disk mode ( Figure 2B and Table 1). The instantaneous irradiance of each spot was thus about 20 times larger than the average irradiance, and only ∼5% of the field of view was illuminated by the excitation light when the disk was spinning. This result confirms that most of the illumination light is blocked by the spinning disk, transmitting only a small portion of excitation light at each point of the sample. In summary, the higher fluorescence intensities in the TIRF microscope are due in part to the fourfold enhancement of the intensity of the evanescence field that excites the fluorophores. The excitation intensities in epifluorescence and SDC are similar when the instantaneous illumination in the SDC is increased roughly 20-fold to compensate for the attenuation due to the small fraction of the spinning disk occupied by the pinholes. Comparison of emission efficiencies in the SDC, TIRF, and epifluorescence microscopes We next compared the fluorescence intensities of the highly labeled Alexa Fluor 488 MTs when imaged by the TIRF, epifluorescence, and SDC microscopes. Consistent with the single-molecule measurements, fluorescent MTs were the brightest by TIRF microscopy and dimmest by SDC microscopy ( Figure 3A) under similar illumination irradiances and identical camera exposure times. Like the bleaching rates, the fluorescence intensities increased linearly with increasing illumination irradiance ( Figure 3B), showing that the excitation light intensity is within the linear dynamic range. This is expected because the excitation intensity we used was much smaller than the predicted saturation intensity of Alexa Fluor 488 (4 GW/m 2 , about 10,000 times higher than the maximum irradiance we used). From the slopes of the linear regression, TIRF microscopy is 3.5 ± 0.2 times brighter than epifluorescence (Table 3), which agrees well with the 3.8 ± 0.2-fold increase in TIRF's excitation field measured from the bleaching rate (Table 1) and is expected from the enhancement of the evanescence field intensity in TIRF (mentioned above). However, the SDC fluorescence intensity was 3.6 ± 0.1-fold lower than that of epifluorescence ( Figure 3B and Table 3), even though the excitation intensities were similar (as judged by the similar bleaching rates stated above). The lower intensities of the SDC images compared with those of epifluorescence indicate that the SDC microscope suffers more light loss in the emission pathway. The loss is not due to the objective as the same objective was used. Some of the loss is expected to be due to the pinhole (∼20%; Supplemental Figure S1; Wilson, 1990); the rest of the loss is presumably due to the lenses, mirrors, and filters. In summary, we have accounted for the lower fluorescence intensity measured in SDC microscopy compared with that in wide-field (epifluorescence and TIRF) microscopy as a combination of different excitation intensities and emission efficiencies. As an overall check on our single-molecule fluorescence calibrations, we estimated the density of the Alexa Fluor 488 in the highly labeled MTs. The density was 59% (estimated by TIRF) and 54% (estimated by SDC); these values are similar to the labeling density estimated by absorbance using a spectrophotometer (43 ± 1%; mean ± SD) ( Table 2). Therefore, we have established a calibration standard for fluorophore counting by measuring the bleaching step size of single fluorophores using SDC microscopy. Establishing a consistent calibration standard for fluorophore counting with SDC microscopy To facilitate calibration of the single-molecule fluorescence in the SDC microscope, we used fluorescent beads as a reference standard. We imaged TetraSpeck beads (Thermo Fisher, cat no. T7279) in both the TIRF and SDC microscopes under the same imaging conditions as for the single-fluorophore experiments. The intensity ratios of TetraSpeck beads to single-Alexa Fluor 488 dyes measured by TIRF and SDC were 225 ± 72 (mean ± SE) and 227 ± 18, respectively (Table 2). By periodically measuring the intensity of TetraSpeck, we could test whether there has been a change in the excitation or emission intensities of the SDC microscope. When using different fluorophores, we adjusted the calibration based on the relative brightness compared with Alexa Fluor 488 at the illumination wavelengths, as well as considering the sample irradiance (laser intensity setting) and the camera exposure time. The accuracy of our calibration is primarily determined by the accuracy of the step size measured in the SDC, which had a coefficient of variance of 0.07 (SD/mean in three independent measurements). From the t distribution, the 95% confidence interval is ±0.07 × 2.92 ≅ 0.2 (2 degrees of freedom). Other sources of error include measurement of the irradiance (<5% SD/mean) and uncertainties of the fluorescence intensities inside the cell due to molecular interactions. The latter effects are difficult to judge. Overall, we expect our calibration to be accurate around 20% (SD/mean), which makes the 95% confidence range roughly a factor of two (between 0.6 and 1.4). Quantification of EB1-eGFP in Drosophila neurons MT EBs track growing MT ends in vitro and in vivo. The amount and the size of the EB binding zone on a growing MT tip can serve as an indicator of the GTP-tubulin cap size (Zanic et al., 2009;Maurer et al., 2011;Seetapun et al., 2012;Coombes et al., 2013;Strothman et al., 2019;Roostalu et al., 2020). Additionally, EBs can recruit various MT-associated proteins that may be critical for the dynamics and function of MT filaments (Akhmanova and Steinmetz, 2015). While the number and direction of EB comets are frequently used to estimate the number and polarity of growing MTs in neurons (Stone et al., 2008;Stewart et al., 2012;del Castillo et al., 2015), the absolute number and binding region of EB molecules have not been examined in detail. Quantifying the amount of EB proteins at a growing MT end, the size of the EB-binding region (i.e., the EB-cap size), and its binding affinity thus provide valuable insights into the dynamical properties of the MT cytoskeleton. To quantify EB comets in vivo, we expressed EB1-eGFP and the cell membrane marker CD4-tdTomato in class IV neurons of Drosophila larvae under cell-specific promoters (Rolls et al., 2007;Han et al., 2011). EB1 preferentially binds to growing MT ends and turns over rapidly; in this way it forms a comet-like distribution behind the polymerizing tip of the MT (Bieling et al., 2007;Dixit et al., 2009). Imaged by SDC microscopy, fluorescent comets were observed throughout the dendrites of these neurons (example in Figure 4A). To measure the shapes of the comets, we modeled the comets as exponential decays convolved with a Gaussian point-spread function using a Monte Carlo (MC) optimization procedure (red dashed curve in Figure 4B and see Supplemental Figures S3-S5 and Materials and Methods for details). By considering the fluorophore brightness (i.e., the product of extinction coefficient and quantum yield), emission spectra, illumination irradiance, and camera exposure time, we determined the intensity of a single eGFP using the bleaching step size of Alexa Fluor 488 as a calibration standard (see Materials and Methods for details). We found that each comet contains on average 84 ± 25 (mean ± SD, n = 6 larvae) EB1-eGFP monomers ( Figure 4C), with an exponential decay length λ of 190 ± 40 nm. If we assume that EB1 binds to GTP-tubulin in the lattice, then the GTP-cap has an average length of 190 nm (where the density decreases efold). Given that there are 300 tubulin dimers in 190 nm (assuming the dimer length is 8.2 nm and there are 13 protofilaments and that the GTP-tubulin density decays exponentially from 100% at the end) and EB1 has been suggested to exist as a dimer (Seetapun et al., 2012;Sen et al., 2013), we estimate that 14% of the GTP-tubulins have EB1-eGFP dimer bound (occupancy η = 42/300, assuming one EB1 dimer binds to one GTP-tubulin subunit and EB1-eGFP is much more abundant than the unlabeled EB1). The cytoplasmic EB1-eGFP monomer concentration was 0.68 ± 0.55 μM ( Figure 4D; mean ± SD, n = 6 larvae; see Materials and Methods for details), which is within the large range of physiological EB1 concentrations reported from various species and cell types (from ∼0.14 μM monomer in budding yeast to 2.1 μM monomer in HeLa cells) (see references in Supplemental Table S1). Combining these measurements, we can estimate the binding affinity between EB1-eGFP dimer and GTP-tubulin lattice by   (assuming that unlabeled EB1 and EB1-eGFP have the same affinities), where K D is the dissociation constant, η is the occupancy of EB1-eGFP (the number of EB1-eGFP dimers bound divided by the total GTP-tubulin sites in the lattice), and α is the fraction of EB1 monomers containing eGFP (i.e. the labeling density). If we assume α is equal to 1, K D = 1.9 ± 0.9 μM (mean ± SD, n = 6 larvae), which is similar to the dissociation constant estimated in human tissue culture cells (3.8 μM; Seetapun et al., 2012), but much larger than the one reported from in vitro measurements (22 nM;Maurer et al., 2014) (Supplemental Table S2). This estimation represents an upper bound of K D because we assumed that the overexpressed EB1-eGFP is much more abundant than the endogenous unlabeled EB1. Thus, our results suggest that these neurons expressed a few hundred nM of EB1-eGFP, with micromolar-level binding affinity to the GTP-tubulin lattice in dendrites. Previous in vitro studies have shown that the decay length of EB comets increases with the MT polymerization rate (Bieling et al., 2007;Strothman et al., 2019;Farmer et al., 2021). To investigate whether this correlation holds for dendritic MTs, we examined the relationship between the number of EB1-eGFP, the decay length of EB1-eGFP comet, and the comet speed. The comet velocity (0.094 ± 0.057 μm/s, mean ± SD from 68 comets in six larvae, similar to previously reported EB comets speed in Drosophila neurons; Ori-McKenney et al., 2012;Poe et al., 2017) showed positive correlations with both decay length and EB1-eGFP number (Pearson's r = 0.53 and 0.55, respectively, n = 68 comets), suggesting that faster growing MTs have larger GTP-caps and a larger number of bound EB1-eGFP ( Figure 5, A and B). These correlations are consistent with previous measurements in a cultured human cell line (Matov et al., 2010) and in the axons of cultured primary neurons from Drosophila (Hahn et al., 2021). Our measurements provide additional information, namely the absolute number and affinity of EB1-eGFP in tissues rather than cultured cells. The GTP-cap size of the dendritic MTs estimated here (∼23 layers of tubulin) was only about half of the GTP-cap size estimated in the cultured human LLCPK1 epithelial cells (∼55 layers) (Seetapun et al., 2012), potentially reflecting the slower average speed of the dendritic EB comets (94 nm/s in the dendrites, 157 nm/s in LLCPK1 cells; mean ± SE, see Supplemental Table S2 for a summary). Interestingly, the EB-comet lengths observed in the dendrites of Drosophila neurons (188 ± 14 nm with a polymerization rate of 94 ± 7 nm/s; mean ± SE) were considerably smaller than the comet lengths of reconstituted MTs with similar growth rates polymerized from purified mammalian brain tubulin (Maurer et al., 2014;Chaaban et al., 2018;Strothman et al., 2019) (summarized in Supplemental Table S2). Instead, the comet length is comparable to those on MTs polymerized from Caenorhabditis elegans tubulin; these MTs displayed higher catastrophe frequency than those from mammalian brain tubulin (Chaaban et al., 2018). While the GTP-cap size may not be the sole factor to determine the catastrophe frequency (Bowne-Anderson et al., 2015;Farmer et al., 2021), the shorter cap size may imply more frequent shrinkage events in these dendritic MTs. Live imaging of single MT filaments in the neuron of a whole organism remains a major challenge, leaving the direct measurement of neuronal MT dynamics difficult. Future investigation of the EB-cap size in mutants of MT polymerases, depolymerases, and MT stabilizers can potentially provide new insights into the relationship between GTP-cap size and MT stability in these neurons. CONCLUSION We introduced a strategy to quantify the fluorophore number in tissue using SDC microscopy. We imaged single fluorophores in the SDC microscope and used the single-fluorophore bleaching step size to calibrate fluorophore numbers. Applying this strategy, we quantified the number of EB1-eGFP in both puncta-like structures (i.e., EB1 comets) as well as its cytoplasmic concentration, allowing the estimation of binding affinity in vivo. This approach is built on the assumption that the fluorophore intensities inside the cells are similar to the ones measured in vitro, while the brightness of fluorophores may depend on the environments. More detailed intensity corrections can be performed based on previous reports comparing the photoproperties of several fluorescent proteins from both in vitro and in vivo systems (Chen et al., 2002;Heppert et al., 2016;Botman et al., 2019) or using proteins complexes with known copy numbers as internal standards (Thevathasan et al., 2019) to achieve better accuracy. A major advantage of the strategy introduced here is that the calibration process does not require additional microscopic or biochemical methods and can be performed directly using the identical microscopy setup as in the in vivo imaging experiments. The calibration procedure can be simplified without the needs of extra imaging standards (e.g., fluorescent beads or purified protein standards used for immunofluorescence), tissue fixation, or the isolation of specific cell types from the tissues. Additionally, our fluorescence intensity and bleaching rate measurements using TIRF, epifluorescence, and SDC microscopy provide a quantitative assessment of the minimal brightness required for visualizing single fluorophores using SDC microscopy, which is an important step toward single-molecule imaging in cellular systems. The quantitative imaging methods introduced here are broadly applicable to quantifying the number of target molecules in live cells within tissues, which are typically more challenging systems for other fluorophore-counting methods. Flow chamber and MT preparation Tubulin was purified from bovine brain as previously described (Castoldi and Popov, 2003). The preparation of imaging chambers and stabilized MTs followed the methods described in Gell et al. (2010) and Kuo and Howard (2021). All reagents were purchased from Sigma-Millipore except as otherwise noted. To affix the MTs onto the surface of the silanized coverslips, 25 μg/ml anti-tubulin antibody (clone SAP.4G5) solution was perfused in the flow chamber with 5 min incubation and washed by BRB80 buffer (80 mM PIPES-KOH, pH 6.9, 1 mM ethylene glycol-bis (β-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 1 mM MgCl 2 ). The channel was then passivated by incubation with 1% pluronic F127 solution followed by 2 mg/ml casein solution as previously described (Kuo et al., 2019). Alexa Fluor 488-labeled stabilized MTs were prepared by polymerizing Alexa Fluor 488-conjugated bovine tubulin in the presence of slowly hydrolyzable GTP analogue GMPCPP (Jena Bioscience) as previously described (Gell et al., 2010). For single-fluorophore imaging, Alexa Fluor 488 tubulin was mixed with unlabeled tubulin so that the final labeling density of the fluorophore was around 0.09%. Oxygen scavenger solution (40 mM glucose, 40 μg/ml glucose oxidase, 16 μg/ml catalase, 0.1 mg/ml casein, 1% β-mercaptoethanol in BRB80) was used for all fluorescent MT imaging experiments. Microscopy setup TIRF and epifluorescence imaging was performed on an inverted microscope (Nikon Ti Eclipse) with a 488 nm excitation laser and a 525/50 nm emission filter. For SDC imaging, an inverted Nikon TI microscope equipped with a confocal scanner unit (CSU-W1 disk; Yokogawa), which contains a four-band dichroic beamsplitter (Di01-T405/488/568/647; Semrock) and a 525/50 nm emission filter, was used. Both microscopes were operated by Nikon NIS element software. All images were collected by a 100×/1.45 NA oil objective (CFI Plan Apochromat Lambda; Nikon) with sCMOS cameras (Zyla 4.2 plus, Andor). Fly stocks We used Gal4 driver line ppk-Gal4 to drive the expression of UAS-EB1-eGFP (Stock 35512 from the Bloomington Drosophila Stock Center) to visualize the growing plus ends of MTs and the reporter line, ppk-CD4-tdTomato (Han et al., 2011), to observe the dendrite morphology of class IV dendritic arborization (da) neurons. Larva sample imaging Embryos were collected for 2 h on apple juice agar plates with a dollop of yeast paste and aged at 25°C in a moist chamber. The plates containing the first batch of embryos were discarded as the dendritic morphology of the da neurons was less consistent in those animals. Larvae were immobilized individually on agarose pads (thickness 0.3-0.5 mm) sandwiched between a slide and a coverslip. The imaging was done using a spinning-disk microscope: the Yokogawa CSU-W1 disk (pinhole size 50 μm) built on a fully automated Nikon TI inverted microscope with perfect focus system and an sCMOS camera (Zyla 4.2 plus sCMOS) and running Nikon Elements software. The EB1 comets in class IV da neurons were imaged with 80% of maximal power for the 488 nm laser and exposure time 200 ms. The tdTomato labeled membrane was imaged with 40% of the maximal power for the 561 nm laser simultaneously. Measurements of fluorescence intensity and bleaching kinetics The pixel intensities within 7 × 7 pixel 2 boxes (450 × 450 nm 2 ) centered at each tetraspeck bead or single fluorophore were summed to get the overall intensity. The background signal was estimated from regions away from fluorescent spots and further subtracted from the overall intensities. Fluorescence intensity traces for individual Alexa Fluor 488 dyes were obtained by calculating overall intensities over time. To analyze the single-fluorophore photobleaching data, we used the previously developed step detection algorithm that uses statistical tests based on the two-sample t test without assumed equal variance to identify steps (Chen et al., 2014). The detected steps smaller than a quarter of the average single bleaching step size were merged with neighboring steps to avoid unrealistic small steps. Steps that last longer than six frames and with signal-to-noise ratios larger than three were used for step size and single-step bleaching time estimations. Single-fluorophore intensity estimation The intensity ratio of single eGFP and Alexa Fluor 488 (I eGFP /I Alexa488 ) on SDC microscopy can be estimated based on the following equation: where r ε indicates the ratio of extinction coefficients at excitation wavelength (488 nm); r Φ is the ratio of quantum yields; T is the overall transmittance of the emission filter and dichroic beamsplitter; F is the relative emission intensity of the two fluorophores; r cam and r ex correspond to the ratios of camera exposure time and excitation irradiance between the imaging conditions from the in vivo experiments and in vitro single-molecule calibration in the SDC microscope, respectively. The extinction coefficients and quantum yields were obtained from Thermo Fisher and the Fluorescent Protein Database (FPbase). The emission spectra were obtained from the Chroma Spectral Viewer, and the transmittance profiles of the dichroic mirror and emission filters were based on the specifications from the manufacturers. The conversion factor from the spectral properties of fluorophores in our system = 0.65. Cytoplasmic EB1-eGFP concentration estimation In the axial dimension, diffraction limits the resolution to 2nλ/NA 2 , with n the refractive index of the medium between objective and sample, corresponding to a depth of field of ∼800 nm. The thickness of most branches studied in this paper fall within the estimated depth of field. Thus, we used the plateau of the MC fitting of the fluorescence signal ( Figure 4B, blue dashed line) from the midplane of the dendrite to calculate the cytoplasmic EB1-eGFP concentration. The cytoplasmic EB1-eGFP monomer concentration can be obtained by dividing the total number of local cytoplasmic EB1-eGFP monomers by the volume of the dendrites approximated by D L 4 2 π (where D is the full width at half maximum of the dendrite from the membrane marker and L corresponds to the length of the dendrite where total intensity was extracted). MC optimization method The overall intensity profile of the EB1-eGFP comets within 0.5 × 4-8 μm 2 rectangular box can be fitted by e Six 1000-step simulations were carried out for σ and λ detection. The obtained σ and λ are averages of results from six simulations. The total intensity of the comet was calculated by summing the total experimentally measured signal within the range of -2 σ to 2λ, with zero being the center of the comets.	2022-01-04T16:03:53.590Z	2022-01-02T00:00:00.000	{ "year": 2022, "sha1": "3a3aa2230ab4d691f2d974dcb2c94a7c30e1c737", "oa_license": "CCBYNCSA", "oa_url": "https://doi.org/10.1101/2022.01.02.474734", "oa_status": "GREEN", "pdf_src": "PubMedCentral", "pdf_hash": "a630ff353797884760e0e4d72daa219de051aded", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
233816542	pes2o/s2orc	v3-fos-license	Characterization of rambai (Baccaurea motleyana) genes putatively involved in sugar metabolism Rambai (Baccaurea motleyana) is one of the underutilized fruits native to Indonesia. Rambai has high antioxidant activities containing phenolic, flavonoid, and anthocyanin compounds, secondary metabolite compounds derived from sugar metabolism. The sugar metabolism involved several related genes. This research aimed to characterize genes putatively involved in sugar metabolism in Rambai. Six sugar gene families were identified from 37 077 contigs of the assembled-transcriptome database against to UniProt database using the BLASTX program. The six sugar-related genes were characterized involved nine contigs of sucrose-phosphate synthase (SPS), three contigs of sucrose-phosphatase (SPP), 14 contigs of sucrose synthase (SUS), 19 contigs of alkaline/neutral invertase (INV), one contig of cytosolic invertase (CINV) and five contigs of beta-fructofuranosidase (CWINV). This research aims to give a comprehensive study of the sugar metabolism mechanism in B. motleyana. The data also revealed the genes that encoded the enzymes that were putatively involved in sugar metabolism. Introduction The tropical rainforests of Southeast Asia, especially the Indo-Malay region, have diverse plant genetic resources, including wood species, medicinal plants, and fruit trees. Many fruits are considered rare or poorly known because they have not been commercially explored, such as menteng (Baccaurea racemosa), rambai (Baccaurea motleyana), and gandaria (Bouea macrophylla Griffith) [1]. The commercial value of plants is obtained from that function derived by its secondary metabolites as a source of food additives, pharmaceuticals, or flavor [2]. Rambai (Baccaurea motleyana) belongs to the family of Phyllanthaceae with menteng and tampoi. Rambai is a plant native to Sumatra, Kalimantan, and Java, then spread to Peninsular Malaysia, Sumatra, Kalimantan, Java, and Bali, and other countries as Thailand and the Philippines [3]. Rambai can grow in lowlands, tropical rain forests, open scrub, and riparian forests and often cultivated in home gardens. The fruit has a size of 2.5-4 cm, oval-shaped, the fruit's color is yellowish-green when unripe and yellowish-white when ripe [4]. In Kalimantan, rambai is one of the non-timber forest products (NTFPs), which are sold at a price of Rp.10,000 / kg and commonly consumed as a table fruit [3] [5]. Each 100 gram of B. motleyana arillode contains 64 kcal energy, 83.7 g water, 0.4 g protein, 0.4 g fat, 14.6 total carbohydrates, 0.1 g fiber, 0.2 g ash, and variants of mineral and vitamins such as Ca, P, Fe, Na, K, vitamin C, vitamin B, niacin, and vitamin K [6]. In addition to its nutritional and vitamin content, rambai fruit also contains secondary metabolites in phenols, flavonoids, and antioxidants. The phenol content of rambai in three different [8]. Furthermore, ethanol extract from the fruit skin showed antibacterial activity for Bacillus cereus, B. subtilis, Staphylococcus vulgaris, and Escherichia coli [9] and had a decreased blood sugar activity that caused hypoxides in mice [10]. Secondary metabolites known as the intermediate products of plant metabolism are usually described as small molecules with various functions such as cell structure, signaling, fuel, or even stimulate several enzymes responsible for having an inhibitory effect, defenses, stress response, and interactions with other organisms. Plants produce various organic compounds, most of which do not play a direct role in growth and development; these compounds are known as secondary metabolites [2]. The function of secondary metabolites in plants is closely related to plants' physiological processes in their environmental growth [11]. The expression of synthetic pathways of plant secondary metabolites can be affected by the supply of precursors, environmental and climatic conditions, and special treatment such as immobilization and biotransformation [12]. The precursor source can be derived from sugar metabolism, and this research aimed to characterize genes putatively involved in sugar metabolism in Rambai. This research aims to give a comprehensive study of the sugar metabolism mechanism in B. motleyana. The data also revealed the genes that encoded the enzymes that were putatively involved in sugar metabolism. Materials and methods The study was conducted in the experimental field of PT. Mekar Unggul Sari, Cileungsi Bogor. The NGS data were collected from DDBJ with accession DRA007358 and the assembled-contigs [13]. SwissProt database (UniProt) was used to annotate using the BLAST program to gene-related to sugar metabolism with the cut-off of 10 −5 . Gene primers of the Sucrose synthase (SUS), sucrose-phosphate synthase (SPS), Sucrose-phosphatase (SPP), Alkaline, neutral invertase (INV), Cytosolic invertase (CINV), and beta-fructofuranosidase (CWINV) are used from DNA sequences of rambai (B. motleyana), and genes actin is used as an internal reference gene. Results and discussion As the figure mentioned in Table 1, nine contigs of Sucrose-phosphate synthase show that this enzyme has an essential role in B. motleyana fruit metabolism. The enzyme of alkaline/ neutral invertase (INV) was established as the most counted contigs, as many as 19 contigs, followed by sucrose synthase (SUS), sucrose-phosphate synthase (SPS), Beta-fructofuranosidase (CWINV), sucrose-phosphate (SPP), and cytosolic invertase (CINV) respectively. [27]. Sucrose-phosphate synthase (SPS; E.C. 2.4.1.14) has a main role in sucrose biosynthesis. SPS in plant enzyme that regulated by reversible proteins and metabolites on both photosynthetic and non-photosynthetic tissues. This enzyme has the role of helping leaves respond to light and dark signals to produce the end product of photosynthetic [14]. Both SPS and SUS have a strong function on sucrose accumulation on some plants, such as Asian pear (Pyrus pyrifolia) and Chinese pear (Pyrus Bretschneider Rehd.). Meanwhile, SS and SPS activity increased during fruit maturation, and both of them play a strong role in sugar assimilation on fruits. A study reported that these enzyme activities were elevated four and seven times on the maturation stage, respectively [15]. In the aril of Litchi chinensis Sonn, the pattern of enzyme activities and gene expressions explained that hexose/sucrose ration was unlikely determined by SPS activity. This research also revealed that sucrose cleavage enzymes such as AI and SS more dominant than sucrose synthetic enzyme-like SPS in determining fruit's sugar content. In contrast to litchi, 'Yali' pear (Pyrus Bretschneider Rehd.) does not show an SPS and SS activities increasing during fruit development and fruit maturation [29]. The next enzyme established was Sucrose-phosphatase (SPP; EC 3.1.3.24), which counted as many as three contigs on rambai fruit flesh. SPP has a significant impact on the finalization of the sucrose synthesis pathway. Many studies revealed that most higher plants contain this enzyme on several types or isoform encoded by different genes. For instance, this enzyme is encoded by genes on chromosomes 1 (AtSPP1), 2 (AtSPP2), and 3 (AtSPP3a and AtSPP3b) on the dicotyl plant such as on Arabidopsis thaliana [17]. Another study reported that this enzyme was highly expressed on environmental stress such as cold, drought, and salinity. In extreme conditions, SPP has catalyzed sucrose-6F-phosphate (Suc6P) production that is used to produce an SPS enzyme that finally plays a role in hydrolyzing sucrose [16] irreversibly. A different gene might encode this enzyme on each plant. For example, this enzyme's presence on Pyrus patens was encoded by a four kDa protein, which had both of HAD phosphatase and C terminal domains of SPP [17]. The enzyme codes were related to the enzyme's function in the sucrose metabolism pathway, as revealed in Figure 2. Another enzyme established on rambai fruit flesh is Sucrose synthase (SUS/ EC 2.4.1.13), which has an essential role in sucrose assimilation during fruit ripening [18]. A study reported that this enzyme plays a dominant role in strawberry fruit ripening and will be inhibited by abscisic acid and sucrose at a specific concentration [19]. However, in some plants, SUS was responded to sucrose cleavage rather than sucrose synthesis [27]. The activity of SS (sucrose synthase) in 'Yali' fruit has a significant SS activity when at the early stage of fruit development but declined dramatically along with fruit development. This study also explained that SS activity was seasonal and leads toward sucrose cleavage, then sucrose synthase [29]. Meanwhile, in melon (Cucumis melo), the SUS role catalyzes sucrose synthesis from fructose and UDP-glucose or reversibly from sucrose to fructose and UDP-glucose. This reaction might occur after the production of sucrose-P finished by SPS [20]. The most dominant enzyme established at rambai is Alkaline-neutral-invertase (EC 3.2.1.26), with as many as 19 contigs. As many as the enzyme established, this enzyme is thought to have a strong impact on sugar metabolism. Many studies established that INV is an enzyme that has the function to catalyze the irreversible hydrolysis of sucrose into fructose and glucose. This enzyme could be determined by the optimum pH (acid or alkaline/neutral) and divided into two groups in higher plants. They are the alkaline/ neutral invertases and the acid invertases. Alkaline/neutral invertases belong to glucosidases with alkaline/neutral optimum pH ranged from 6.5-8.0 and predicted to be established on cytosol or other organelles. Meanwhile, acid invertases are called as fructofuranosidase, which have pH optimum (acidic) from 4.5-5.0. These enzymes are predicted to be located in the cell wall and vacuole [21]. Many studies have reported that most acid invertases are involved in several plant life cycle aspects such as photosynthetic assimilates partitioning, environmental condition response, cell enlargement, and plant growth and development [22][23] [24]. The next enzyme that putatively on rambai sugar metabolism is cytosolic invertase (CINV/ EC 3.2.1.26) appeared at one contig. Some studies showed that CINV genes are important for Lotus japonicas and Medicago sativa growth and development [30,31]. In Arabidopsis, the CINV/UGP vital important to cellulose synthesis. Another study reported that CINV activity was affecting carbon partitioning [32]. The last enzyme analyzed to have a role in rambai's sugar metabolism is beta-fructofuranosidase (CWINV/ 3.2.1.26) appeared at five contigs. Both of these enzymes play a strong role in producing glucose and fructose-the role of CINV and CWINV on sugar metabolism at plant cells described in Fig 3. Figure 3. Sugar metabolism at plant cell and the role of putative enzymes [25] Figure 3 described that CWINV from wall-cell was hydrolyzed the sucrose on apoplast to be glucose and fructose, then carried out to the cell by a monosaccharide transporter. Besides that, sucrose is delivered to other sink cells through plasmodesmata by sucrose transporter. The sucrose is then stored at the vacuole or hydrolyzed by vacuolar invertase (vINV). Furthermore, after the sucrose is stored in the cytosol, hydrolyzed by cytosolic invertase (CINV) becomes glucose and fructose. SUS can reverse this reaction to yield fructose and UDP-G [25].	2021-05-07T00:04:29.019Z	2021-03-01T00:00:00.000	{ "year": 2021, "sha1": "6fcda93b79daef144f784b7a723cc0e751e6fa7f", "oa_license": null, "oa_url": "https://doi.org/10.1088/1755-1315/694/1/012067", "oa_status": "GOLD", "pdf_src": "IOP", "pdf_hash": "c73ee00e658501cce2ca8409e2e77fccb6fab608", "s2fieldsofstudy": [ "Biology", "Agricultural And Food Sciences" ], "extfieldsofstudy": [ "Physics", "Biology" ] }
263804522	pes2o/s2orc	v3-fos-license	Construction of a predictive model for osteoporosis risk in men: using the IOF 1-min osteoporosis test Objective To construct a clinical prediction nomogram model using the 1-min IOF osteoporosis risk test as an evaluation tool for male osteoporosis. Methods The 1-min test results and the incidence of osteoporosis were collected from 354 patients in the osteoporotic clinic of our hospital. LASSO regression model and multi-factor logistic regression were used to analyze the risk factors of osteoporosis in patients, and the risk prediction model of osteoporosis was established. Verify with an additional 140 objects. Results We used logistic regression to construct a nomogram model. According to the model, the AUC value of the training set was 0.760 (0.704–0.817). The validation set has an AUC value of 0.806 (0.733–0.879). The test set AUC value is 0.714 (0.609–0.818). The calibration curve shows that its advantage is that the deviation correction curve of the nomogram model can maintain a good consistency with the ideal curve. In terms of clinical applicability, compared with the "total intervention" and "no intervention" schemes, the clinical net return rate of the nomogram model showed certain advantages. Conclusion Using the 1-min osteoporosis risk test provided by IOF, we built a male osteoporosis risk prediction model with good prediction effect, which can provide greater reference and help for clinicians. Introductions Osteoporosis is a chronic disease characterized by decreased bone density and deterioration of bone microstructure and is a global disease with high incidence [1]. Due to bone fragility and a higher risk of future fractures, osteoporosis is rapidly becoming a critical health issue today [2].At the same time, osteoporosis is often ignored and irreversible, once it occurs, it will bring multiple burdens such as life ability, spiritual needs and economic costs to patients [3].At present, the measurement of bone mineral density by dual-energy X-ray absorption method is currently recognized as the gold standard for the diagnosis of osteoporosis [4], but dual-energy X-ray method is expensive and should not be measured repeatedly in the short term [5].As a result, several clinical risk assessment tools have been developed to assess the risk of osteoporosis [6].However, these tools tend to favor women more, especially postmenopausal women, and ignore men or are less effective at predicting which men are [7,8].Although women have a higher risk of osteoporosis than men, the lifetime risk of a non-traumatic fracture after osteoporosis is estimated to be about 25% for a 60-year-old man [9,10].Men are also twice as likely as women to die in hospital after a hip fracture [11].The 1-year mortality rate after fracture was 31% for men and 17% for women [12].In addition, studies have shown that most older men with pre-existing fragility fractures do not know to take screening bone mineral density (BMD) tests or receive treatment [13].Therefore, a convenient, simple and highly controllable screening method for men is needed to reflect the risk of osteoporosis in the body and provide a preventive basis for the prevention of osteoporosis in men [14].The International Osteoporosis Foundation (IOF) offers a 1-min osteoporosis risk test, a simple and sensitive primary screening tool [15].The survey is an internationally recognized tool for raising awareness and consists of 19 questions.Compared with the traditional evaluation of osteoporosis, the IOF questionnaire is a more convenient, simple, controllable and authoritative method, which can effectively replace the dual-energy X-ray method for bone mineral density detection in the population, thus reflecting the risk of osteoporosis in the body and providing a basis for followup detection.And further evaluation by a primary care physician [8,16].Therefore, this paper aims to construct a clinical prediction model for elderly people using 1-min osteoporosis risk test as an evaluation tool for osteoporosis and verify its accuracy and clinical practicability. Survey object From January to June 2023, patients in the orthopedic osteoporosis clinic of Union Shenzhen Hospital of Huazhong University of Science and Technology (Nanshan Hospital) and Liwan Community Health Service Center of Shenzhen Nanshan Medical Group Headquarters were selected as the research objects to construct the prediction model.Inclusion criteria include (1) no secondary osteoporosis, (2) clear mind, no communication disorders, and (3) no long-term bed rest or no exposure to sunlight.Exclusion criteria include (1) patients with endocrine diseases such as type 2 diabetes mellitus; (2) patients with a history of chronic heart dysfunction, tumors, impaired liver and kidney function, and immune system diseases; (3) Patients with a history of surgery or disease that may cause gastrointestinal malabsorption; and (4) those who did not cooperate or did not complete the questionnaire. Research methods A 1-min test of osteoporosis risk was performed on all subjects, while basic information such as age, gender, and BMI were recorded.Bone mineral density was measured by dual-energy X-ray method, and the bone mineral density of lumbar spine, spine and radius was recorded, and the lowest bone mineral density was selected for recording. Diagnostic criteria The patients were diagnosed with osteoporosis according to the diagnostic criteria for Primary Osteoporosis (2022).Bone mineral density T value ≤ 2.5 indicates osteoporosis. Statistical methods R language (R.4.1.1)was used for data analysis.We then randomly divided the data set constructed by the patients from Union Shenzhen Hospital (Nanshan Hospital) of Huazhong University of Science and Technology into a training set and a validation set at a ratio of 5:5 to construct and verify the prediction model.After that, patients from Liwan Community Health Service Center of Shenzhen Nanshan Medical Group Headquarters were used as test sets to improve the reliability and robustness of the study results.Measurement data are expressed as Median (IQR) and counting data are expressed as n (%).Mann-Whitney U test, Pearson Chi-square test or Fisher exact probability method were used for comparison between groups, respectively.LASSO regression analysis is implemented through the "glmnet package," the "rms" package for the drawing of the nomogram and calibration curves, the "pROC" package for the ROC curve, and the Area under the ROC Curve (AUC) for evaluating the judgment of the nomogram.The internal verification of the nomogram model was realized by Bootstrap self-sampling 1000 times.Calibration curves are used to assess the predictive consistency of a nomogram.P < 0.05 was considered to be statistically significant. Results Of the 354 people who were eventually included in the training and test sets, 106 had osteoporosis.People with bone pine are older than those without osteoporosis.Having a parent diagnosed with osteoporosis or a hunchback in one of the parents; Smoking; Alcoholism; The proportion of patients with osteoporosis was greater in the problems such as rash fracture (P < 0.05).(See Table 1 for details). We randomly divided the study population into a training set (n = 177) and a validation set (n = 177) according to the osteoporosis rate at a ratio of 1:1.There was no significant difference in the 1-min test results between the two datasets (P > 0.05).(See Table 2 for details).Five characteristic variables selected from the LASSO regression model were included in the multifactor logistic regression, and the results showed that age (OR: 1.05; 95% CI 1.02-1.09);Are you underweight (OR: 3.32; 95% CI 1.51-7.39);After the age of 40, have you lost more than 3 cm in height(OR: 2.58; 95% CI 1.41-4.74);Do you currently, or have you ever, smoked cigarettes(OR: 1.93; 95% CI 1.13-3.33 was an independent risk factor for osteoporosis.(See Table 3 for details). A nomogram prediction model of osteoporosis was established based on logistic regression model.The model has a perfect score of 180, and when the score is more than 160, the risk of osteoporosis is 80%.(See Fig. 2 for details). Meanwhile, according to the model, the AUC value of the training set is 0.760 (0.704-0.817).The validation set has an AUC value of 0.806 (0.733-0.879).In addition, we calculated the AUC of the test results according to the judgment criteria of the IOF 1-min test questionnaire, and the result showed that the AUC value of 0.692 (0.612-0.773) was lower than our prediction model.(See Fig. 3 for details).The calibration curves of the nomogram prediction model were verified internally by bootstrap resampling 1000 times.The results showed that the nomogram model had good calibration degree and prediction consistency.(See Fig. 4 for details). According to the DCA curve, when the prediction probability threshold of the nomogram model is 0-0.75, the clinical net return rate of the nomogram model is greater than that of the "full intervention" and "no intervention" schemes, suggesting that the nomogram model has good clinical applicability.(See Fig. 5 for details). We then put our model to the final test using patients from another community health service center as a test set.A total of 140 patients were collected; detailed information is shown in Table 4. Inclusion of test set patients in our predictive model resulted in an AUC value of 0.714 (0.609-0.818), indicating that our model predicted well. Discussion We developed a clinical prediction model for male osteoporosis using the IOF 1-min Osteoporosis Test Questionnaire, which was validated and tested to show that the model has good efficacy and clinical applicability.This fills the gap that has long been lacking a reliable and easy-to-implement screening tool to identify people at high risk of osteoporosis in older men [17]. Osteoporosis screening and risk assessment enable clinicians to determine which populations require followup interventions to reduce their risk of complications and death [18].Consistent with other studies, our study found that age is the most important factor in osteoporosis, with older people at greater risk of osteoporosis [19].Secondly, family history, fracture history, and height loss increase the risk of osteoporosis, which is similar to many clinical observations, and some studies have shown that a variety of factors, including family history, history of fractures, smoking, excessive alcohol consumption, rheumatoid arthritis, etc., are risk factors for osteoporosis [20].Another risk factor associated with osteoporosis is BMI, which has been shown to be higher in heavier weight and slower in bone mass loss at the same height level [21]. Based on these risk factors in the osteoporosis risk 1-min test, we constructed a clinical risk prediction model after screening the characteristic variables using LASSO regression.The advantage of the nomogram used in this study is that its analysis results are more intuitive and effective.In developing the most appropriate model, we also included age due to the irreplaceability between age and the osteoporosis association.After that, a simple, inexpensive and effective preliminary screening tool was built to serve the risk factor prediction model in the later stage.In addition, rather than a 1-min risk test designed to help people become aware of their risk factors, our Fig. 2 A graph predicting osteoporosis risk model is designed to assess the risk of osteoporosis [15].The ROC of this model is 0.760 (0.704-0.817).This has a similar predictive effect compared to other predictive models, but it requires fewer problems and is easier to operate [7,22].Another advantage of our study is that we not only divided the original data set of our hospital into a training set and a test set to build and test the model, but also collected patients from other hospitals to verify the model.This makes our model more reliable. Machine learning-based computing methods are becoming increasingly prominent in healthcare applications.While traditional statistical methods rely on inferencing relationships between variables, machine learning is able to predict a patient's status based on other information about the patient [23].A review of 89 studies suggests that ML has the potential to be used to identify factors associated with the risk of osteoporosis, thereby predicting osteoporosis [24].There are a variety of ML methods used, such as SVM, ANN, and random forest.The best performing and most popular models are SVM and logistic regression [25].In addition, some deep learning models are widely used, and the best reported performance is nearly perfect.But more complex models require rich data sets to make modeling predictions useful; this is especially true of deep learning (neural network) models [26,27].A wide variety of features have been explored, and most studies related to the use of ML to diagnose osteoporosis have used imaging tests to constitute the algorithm's most important predictors, with X-rays, ultrasound, MRI imaging, and machine learning all applying the results to infer bone health.Compared with these more complex machine learning methods, although the prediction efficiency is slightly lower, our problem input is simpler, including only 6 variables such as age.At the same time, the way we use the bar chart also makes the prediction more intuitive [28,29].In terms of model stability, 1000 Bootstrap resampling shows that the model is stable and has good correction accuracy and prediction consistency.In terms of clinical applicability, compared with the "full intervention" and "no intervention" schemes, the prognosis of patients is better and the clinical benefits are higher, so it shows that the nomogram model used in this study has better clinical applicability.At the same time, in order to obtain the risk factors of osteoporosis patients, this study incorporated the clinical data of patients during hospitalization into the research model for analysis, which was easy to obtain.In summary, we use the 1-min test of osteoporosis risk provided by IOF to construct a risk prediction model with good prediction effect. Here's the strength of the study: First, it is a study specifically looking at the prediction of osteoporosis in men.Second, we conducted a cross-sectional study rather than a retrospective study.In addition, we not only verified the developed model, but also selected another hospital to verify the model.We also implemented strict inclusion and exclusion criteria to eliminate selection bias as much as possible.The limitations were that the age range we included did not include the entire age group and was not validated in more other geographies, and secondly, we did not take into account more basic information such as patient income and education level. Conclusions We developed a clinical predictive model for osteoporosis in men that was validated and tested to show good predictive outcomes. Compared with other studies, our predictive model can effectively predict osteoporosis using basic questionnaire information without using imaging data.It compensates for the disadvantages of time-consuming and expensive traditional tests, as well as other complex machine learning algorithms that require more information, and can easily and quickly predict osteoporosis.The study may have implications for developing a possible diagnosis of osteoporosis and could be valuable for doctors screening patients for osteoporosis in primary hospitals or community health centers. Fig. 1 Fig. 1 Texture feature selection using the Minimum Absolute shrink and selection operator (LASSO) binary logistic regression model.A The optimal penalty coefficient lambda (λ) was identified in the LASSO model, and 10 × cross-validation 140/90 was performed in the group.B The LASSO coefficient profiles of 21 features in the group were observed in the LASSO coefficient profiles as λ of the LASSO algorithm changed Fig. 3 Fig. 3 Receiver operating characteristic curve of the nomogram.AUC, area under curve Fig. 4 Fig. 4 Calibration curve of the prediction nomogram Table 1 Whether the risk of developing osteoporosis is compared with the 1-min testThe variable has a nonzero coefficient.The questions included five questions, including Are you underweight; Are you underweight (is your Body Mass Index less than 19 kg/m 2 ); After the age of 40, have you lost more than 3 cm in height; Do you currently, or have you ever, smoked cigarettes; Do you spend less than ten minutes per day outdoors (with part of your body exposed to sunlight), without taking vitamin D supplements (see Fig.1for details). VariablesLevelsNo osteoporosis (n = 248) Osteoporosis 1 (n = 106) PHave you ever taken corticosteroid tablets (cortisone, prednisone, etc.) for more than three consecutive months?Have you been diagnosed with an over-active thyroid, overactive parathyroid glands, type 1 diabetes or a nutritional/gastrointestinal disorder such as Crohn's or celiac disease? Table 2 A comparison of 1-min tests of osteoporosis risk in training sets and validation sets Table 3 Logistic regression analysis of risk factors for osteoporosis Table 4 Comparison of the one-minute test on the risk of osteoporosis among 140 subjects in the test set Do you fall frequently (more than once in the last year) or do you have a fear of falling because you are frail?	2023-10-11T14:08:39.899Z	2023-10-11T00:00:00.000	{ "year": 2023, "sha1": "14dae2433b1b9aefbffaa8e31faca68845c1f4f9", "oa_license": "CCBY", "oa_url": "https://josr-online.biomedcentral.com/counter/pdf/10.1186/s13018-023-04266-7", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "772603c97010e8cf6b89d5f36c8bce486a9d8130", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
71121968	pes2o/s2orc	v3-fos-license	PREVALENCE OF LATTICE DEGENERATION IN AXIAL MYOPIA Lattice degeneration is an important predisposing factor for retinal detachment. A cross sectional study of 221 patients and 405 eyes with axial myopia of 25mm (–3.oo diopters) or more, was conducted with an objective to find out prevalence of lattice degeneration of the retina among them. Of 221 patients, 48 (21.7%) had the lattice degeneration consisting of 35 uniocular (72.9%) and 13 binocular patients (27.1%). Of 139 males, 36 (25.89%) had lattice degeneration; of 82 females, 12 (14.63%) had lattice degeneration (RR=1.26; 95%CI=1.03-1.55; p=0.049). Axial length was from 25mm to 35.77mm (mean=27.63mm, SD 1.98). The greatest prevalence of lattice degeneration.9% (16 of 73 eyes) was found in eyes with axial length of 26mm to 26.99 mm (-6.0D to –8.97.0D), and the least incidence was 8.6% (8 of 93 eyes) in eyes with axial length 25mm to 25.99 mm (-3.0 D to –5.97.0 D). Over all prevalence of lattice degeneration was 15% (61 of 405 eyes) of eye(s) with axial length of 25 mm (-3.0 D) or more. In the age groups below 40 years, the prevalence of lattice degeneration was highest 85.24% (59 of 61 eyes). The lattice degeneration of retina is more prevalent in males of age less than 40 years with moderate axial myopia. INTRODUCTION Lattice degeneration of the retina is an important retinal abnormality, which is related to the retinal detachment (RD).The pathogenesis of lattice degeneration is still not well understand.However, the important mechanisms, which plays role in development of lattice degeneration are developmental, degenerative (abiotrophic) and ischemic (retinal and choroidal) process. 1 Typical lattice lesions are sharply defined circumferentially oriented areas of retinal thinning located anterior to the fundus equator. 2 A fishbone or crosshatched pattern of sclerotic, white retinal arterioles and venules within the base of the lattice degeneration are a frequent finding. 3owever, it can be present in association with other several clinical features like localized round, oval, or linear retinal thinning; pigmentation; whitish yellow surface flecks; round, oval, or linear red craters; small atrophic round holes; branching white lines, yellow atrophic spots and rarely tractional tears at the ends or posterior margins of the lesion, which are the frequent cause of retinal detachment. 4The estimated prevalence of lattice degeneration is 6% -8% of the general population 1,5 and the incidence is higher in individuals older than 10 years 6 and in myopic eyes. 7However it is not restricted to eyes with myopia as it was found in emmetropic cases (4.5%) 8 and in hyperopic or emmetropic eyes (25%) 8 Hyams and Newmann 8 reported a 15% incidence of lattice degeneration in 332 eyes with more than one diopter of myopia.Lattice degeneration is present in approximately 20% of patients with RD 9 and in fellow eyes of phakic RD. 10 The prevalence of lattice lesion was found to increase directly with increasing axial length in study by Karline and Curtin, 11 whereas reverse was true in the study by Celorio and Pruett. 12 OBJECTIVE Study the prevalence of lattice degeneration in axial myopia. PATIENTS AND METHODS This prospective study was carried out in retina clinic of Shree Rana Ambika Shah Eye Hospital in Bhairahawa, over a period of February 2003 to January 2004.All patients with axial myopia of 25mm and more, and the patients not mentioned in exclusion criteria, were included in this study.Cases excluded in this study were: eyes having poor visibility of fundus due to hazy ocular media (Corneal opacity, dense cataract, vitreous hemorrhage or opacities), myopia related with other ocular pathology (retinitis pigmentosa, retinopathy of prematurity, retinal detachment surgery), nonaxial myopia (corneal or lenticular), enophthalmos (scleral depression is poor) and recent intra-ocular operative cases (those who are not suitable for scleral depression due to risk of wound opening).All myopic cases were referred from general out patient door clinic to the retina clinic for evaluation.All included (consecutive) cases underwent anterior segment examination with slit lamp followed by fundus evaluation only in well-dilated eyes after counseling for good cooperation and scleral depression and with Goldman Three-mirror contact lens.Examination was performed without general anesthesia.Axial length was measured only with A-scan (contact method) for study and one highest reading was taken.Ultrasonography (USG B scan) was used in those cases having posterior staphyloma (to locate and measure it) followed by A-scan for more accurate reading.The parameters recorded after completing the examination were: demography, age, gender, axial length in mm (A-scan), presence of lattice degeneration and it's type, location, orientation and extent, presence of retinal breaks and retinal detachment.Above data was analyzed using general statistical tool. DISCUSSION Several studies have been done to show the prevalence of lattice degeneration in axial myopia and this study has similar results in many aspects.Over all in our study, the prevalence of lattice degeneration was 15% (61 of 405 eyes) of eye(s) with axial length of 25 mm (-3.0 D) or more, which was less than that of Celerio and Prutte 12 and Cambiaggi 15 et al ((24.1% and 20% respectively) in myopia of 6.0 D or more.The present study also showed that males were affected by lattice degeneration more than females (3:1 ratio) in contrast to Celorio and Prutte, where females were more commonly affected.Our finding of 42.62% (26 of 61 eyes) bilaterally affected was similar to the results of Celorio et all (45.8%),Karlin et al (40%), Byer 6 (34%) and Shiomi 16 (31.6%).Most of the lattice lesion (70.49%) in our study was found in temporal area and it was similar to Celorio (89.5%).Pigmented lattice was more common (82%) 17 Similar results of higher prevalence of lattice degeneration, was observed by others 18,19 too in myopia of axial length 25 mm to 27 mm (-3 to -10.0 D). In contrast to the results of Shiomis' and Celorio (Fig. 1), our study showed tendency of decreasing lattice with increasing axial length only in myopia of more than -15.0 D (Fig. 2).It can be explained, on the basis of Yuras 20 finding, that in high myopic eyes with posterior staphyloma, the lattice is significantly less than the entirely elongated eyes.The other factor is that lattice degeneration may be more closely linked genetically with mild (physiologic) than with severe (pathologic) myopia. Smith and associates 21 observed in 3065 consecutive postoperative eyes and found 6.3% prevalence of retinal detachment in moderate myopic -3 D to -7.50 D (25 to 26.5 mm axial length) and for severely myopic greater than -7.5 D, it was 4.8%.However, the incidence of retinal detachment is nearly double 22 after clear lens extraction for high myopia greater than -10 D. CONCLUSION Overall prevalence of lattice degeneration was 15% in the eye (s) with axial length of 25 mm or more and males of age less than 40 years with moderate axial myopia, were affected predominantly. Table III : Presence of Lattice According to the Gender Distribution in occurrence and it was predominantly (49.1%) found in our study too.Most frequently affected age groups by lattice (51 of 160 cases) in Celorio were 21 to 40	2019-03-07T14:08:45.543Z	2004-07-01T00:00:00.000	{ "year": 2004, "sha1": "8df57f0afe1fbef47f5b8d0a0d77c98fbf5bc4c1", "oa_license": "CCBY", "oa_url": "https://www.jnma.com.np/jnma/index.php/jnma/article/download/559/1058", "oa_status": "HYBRID", "pdf_src": "Anansi", "pdf_hash": "6ff4c56b6cb2b05802e13d47e35ad0e7c014040c", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
15367430	pes2o/s2orc	v3-fos-license	L1R, A27L, A33R and B5R vaccinia virus genes expressed by fowlpox recombinants as putative novel orthopoxvirus vaccines Background The traditional smallpox vaccine, administered by scarification, was discontinued in the general population from 1980, because of the absence of new smallpox cases. However, the development of an effective prophylactic vaccine against smallpox is still necessary, to protect from the threat of deliberate release of the variola virus for bioterrorism and from new zoonotic infections, and to improve the safety of the traditional vaccine. Preventive vaccination still remains the most effective control and new vectors have been developed to generate recombinant vaccines against smallpox that induce the same immunogenicity as the traditional one. As protective antibodies are mainly directed against the surface proteins of the two infectious forms of vaccinia, the intracellular mature virions and the extracellular virions, combined proteins from these viral forms can be used to better elicit a complete and protective immunity. Methods Four novel viral recombinants were constructed based on the fowlpox genetic background, which independently express the vaccinia virus L1 and A27 proteins present on the mature virions, and the A33 and B5 proteins present on the extracellular virions. The correct expression of the transgenes was determined by RT-PCR, Western blotting, and immunofluorescence. Results and conclusions Using immunoprecipitation and Western blotting, the ability of the proteins expressed by the four novel FPL1R, FPA27L, FPA33R and FPB5R recombinants to be recognized by VV-specific hyperimmune mouse sera was demonstrated. By neutralisation assays, recombinant virus particles released by infected chick embryo fibroblasts were shown not be recognised by hyperimmune sera. This thus demonstrates that the L1R, A27L, A33R and B5R gene products are not inserted into the new viral progeny. Fowlpox virus replicates only in avian species, but it is permissive for entry and transgene expression in mammalian cells, while being immunologically non–cross-reactive with vaccinia virus. These recombinants might therefore represent safer and more promising immunogens that can circumvent neutralisation by vector-generated immunity in smallpox-vaccine-experienced humans. Background Preventive vaccination still remains the most effective control against orthopoxvirus (OPXV) infections, as it can elicit neutralising antibodies against an incoming virus. The traditional smallpox vaccine was administered by scarification, but its use was discontinued in the general population from 1980, because of the absence of new smallpox cases. However, attenuated strains of vaccinia virus (VV) have been produced and tested in humans in attempts to develop strains with lower reactogenicity and fewer side effects. Indeed, in spite of the discontinuation of the smallpox vaccination programmes, the threat of deliberate release of variola virus for bioterrorism and the need for protection from new zoonotic infections still remain [1][2][3][4]. Also, the monkeypox virus (MPXV) resembles smallpox in the severity of its symptoms, and it might be a potential bioweapon if cowpox and MPXV can adapt to grow and spread in humans [5]. VV recombinants that express different transgenes have already been used in clinical studies for the prevention and immunotherapy of different infectious diseases. However, although extremely effective, VV raises safety concerns due to its high reactogenicity, its ability to spread to non-vaccinated subjects, and its moderate to severe side effects [6], especially in immune-compromised individuals [7]. Such side effects are not acceptable in a post-endemic smallpox era [6,8]. Clinical trials have also been performed with the attenuated VV-derived Lister clone LC16m8 and the modified vaccinia Ankara (MVA). LC16m8, which replicates in humans, showed protective efficacy in animal models, and was safely used for over 50,000 children in Japan in 1974 [9]. MVA has an extensive history of safety in humans, it is immunogenic and efficacious in both mice and non-human primates, where it can also protect against MPXV, and it is a leading candidate for an alternative smallpox vaccine [10][11][12][13]. However, as MVA replication in mammals is only partially abortive [14], and induces lower immunogenicity than the traditional smallpox vaccine [15], the search for new alternatives is still ongoing [16,17]. Failure of protection with MVA has been demonstrated in animals with CD4/CD8 combined immunodeficiency [6] and in Rhesus macaques infected with simian immunodeficiency virus with a very low cell count of the immune repertoire [18]. The VV antigens that protect against smallpox are not completely known. However, analysis of the immune responses after immunisation with traditional vaccines has shown that the neutralising antibodies are mainly directed against the surface proteins of the two infectious forms of OPXVs: the intracellular mature virions (MVs) and the extracellular virions (EVs). MVs are released after cell lysis, and they are responsible for the host-to-host spread because of their stability in the extracellular environment. In contrast, EVs are wrapped by an additional envelope, and they have an important role in cell-to-cell spread; i.e., in the dissemination within the host [19,20]. In particular, the L1 and A27 proteins on MVs and the A33 and B5 proteins on EVs are involved in the attachment, fusion and penetration of the virus into target cells [20,21]. Combined DNA-based vaccines expressing these four proteins are more protective than vaccines carrying individual immunogens [22,23], possibly because of the induction of synergistic antibodies that can act at the different infection phases, both at the initial exposure, and during viral dissemination [24,25]. Although other immunogens have been used to induce protective immunity, when these four genes have been delivered as DNA expression plasmids [22] or as virus-like replicons [17], they have already been shown to induce functional antibodies and to protect mice against VV parenteral/ intranasal challenge, and monkeys against MPXV intravenous challenge [26,27]. In the present study, four novel recombinants were constructed based on the fowlpox (FP) genetic background that express the VV L1R, A27L, A33R and B5R genes independently. Their correct expression was determined by RT-PCR, Western blotting and immunofluorescence, both in replication-permissive chick embryo fibroblasts (CEFs) and in non-permissive mammalian Vero and MRC-5 cells. The ability of the proteins expressed by the four novel recombinants to be recognized by VV-specific hyperimmune mouse sera was demonstrated by immunoprecipitation followed by Western blotting (IP/WB) using hyperimmune mouse serum. The recognition by VV-specific antibodies of a mixture of the four FP recombinants (4FPmix) was tested using a plaquereduction neutralisation assay. FP vectors are replication-restricted to avian species [28], but they are permissive for entry and transgene expression in mammalian cells, while being immunologically non-cross-reactive with VV. They therefore represent safer immunogens [29], as they can circumvent neutralisation by vector-generated immunity in smallpox-vaccineexperienced humans. The advanced replication cycle, long transgene expression, and balanced Th1/Th2 cytokine induction of FP-based recombinants [28] might elicit a more effective immune response. Construction of the recombinants Four FP-based recombinants expressing the VV MV L1 and A27 proteins and the EV A33 and B5 proteins were obtained by in-vitro homologous recombination [30][31][32], with minor modifications. FPwt was obtained by J. Taylor (Wadsworth Center, NY State Dept. of Health, Albany, NY) [29]. Molecular cloning of pFP A33R and preparation of the FP A33R recombinant have already been described [33]. For FP L1R, FP A27L and FP B5R , these genes were PCRamplified from the Lancy strain of VV DNA (Berna Biotech, courtesy of M. R. Capobianchi, L. Spallanzani National Institute for Infectious Diseases, Rome, Italy). They were separately inserted into the pcDNA3 and pBSII plasmids (L1R), the pBSII plasmid (A27L), or the pcDNA3 and pAFTd plasmids (B5R). Insertion into the pFP MCS recombinant plasmid was performed downstream of the VV H6 early/ late promoter [34], inside the 3-β -hydroxysteroid dehydrogenase 5-delta 4 isomerase gene interrupted by a multiple cloning site. The amplification of L1R was carried out using the forward V182 (5' GGG AAG CTT TTA AAT GGG TGC CGC AGC AAG CAT ACA 3') and reverse V183 (5' GGG CTC GAG ATT TTC AGT TTT GCA TAT CCG TGG TAG 3') primers. For A27L, the forward V354 (5' CCC GGG AAG CTT AAT GGA CGG AAC TCT TTT C 3') and reverse V353 (5' TTT TGG TAC CAT AAA AAT TAC TCA TAT GGG CGC CG 3') primers were used. For B5R, the forward V184 (5' GGG AAG CTT AAA AAT GAA AAC GAT TTC CGT TGT TAC 3') and reverse V185 (5' GGG CTC GAG ATA TTT ACG GTA GCA ATT TAT GGA ACT 3') primers were used. Amplifications were performed as described previously [35], using 2.5 mM MgCl 2 and 2 mM MgSO 4 , with annealing at 61°C for 30 s (L1R and A27L), or at 57°C for 30 s (B5R) and extension at 72°C for 45 s. The β-actin reference gene was amplified using the forward V84 (5' CTG ACT ACC TCA TGA AGA TCC T 3') and reverse V85 (5' GCT GAT CCA CAT CTG CTG GAA 3') primers, with 1 mM MgSO 4 and with annealing at 60°C for 30 s. The plasmid DNAs were purified and the genes were sequenced (Genenco, MMedical, Milan, Italy), to exclude any mutations arising from the PCR amplification. The plasmids are designated as pFP L1R (8,992 bp), pFP A27L (8,568 bp), pFP A33R (8,764 bp), and pFP B5R (9,173 bp). Recombinants were obtained by in-vitro homologous recombination in CEFs, using FPwt and the different pFP recombinants, with minor modifications. Recombinant plaques were identified by autoradiography after hybridisation with [ 32 P]-labelled specific probes, and then subjected to multiple cycles of plaque purification. One clone was selected for correct and high expression of each gene by Western blotting, using specific antibodies. The recombinant viruses were amplified in CEFs, purified on discontinuous sucrose density gradients, and titrated essentially as already described [36]. Briefly, the cells were harvested, ultracentrifuged at 30,000× g for 2 h at 4°C, and the pellets resuspended in 1 mM Tris, 150 mM NaCl, 1 mM EDTA, pH 7.4. The pellet then had 0.06% trypsin added, and was incubated for 5 min at 37°C, and the virus was released from the cells by sonication. The supernatant was overlaid onto a discontinuous 30% to 45% (w/w) sucrose gradient, in the same buffer. After ultracentrifugation at 38,000× g for 1 h, the viral band at the interface was recovered, diluted with 1 mM Tris-HCl, pH 9, and pelletted at 67,000× g for 1 h. The purified virus was resuspended in Ca ++ -and Mg ++ -free phosphate-buffered saline (PBS -), disaggregated by sonication, aliquoted, and frozen at -80°C until use. Western blotting To determine whether the L1, A27, A33, and B5 proteins were expressed by the recombinants at the same levels in the different cell lines, CEFs and Vero and MRC-5 cells were infected (10 PFU/cell) and examined by Western blotting, as already described [38]. The blotted nitrocellulose membranes were incubated overnight at 4°C with 1:100 dilutions of the primary antibody. Alternatively, two different specific antibodies (Beiresources, Manassas, VA, USA) were used for each gene: a mouse monoclonal antibody followed by goat anti-mouse horseradish-peroxidase-conjugated serum, and a rabbit polyclonal antibody followed by goat anti-rabbit horseradish-peroxidase-conjugated serum (1:2,000 dilution; DakoCytomation, Carpinteria, CA, USA). After a 1-h incubation and 2 h of washes, the proteins were revealed using the ECL system (EuroClone, Pero, Milan, Italy). Cells infected with FPwt were used as the negative control. Immunoprecipitation/ Western blotting analysis Conventional Western blotting was also performed after immunoprecipitation of cell lysates using specific antibodies (IP/WB). Vero cells were infected as for Western blotting, and immunoprecipitation was performed essentially as already described [35], with minor modifications. Sixteen hours p.i., the cells were harvested by resuspension in 1 ml lysis buffer (150 mM NaCl, 1 mM EDTA, 10 mM Tris-HCl, pH 7.4, 0.2 mg/ml PMSF, 1% NP40, 0.01% sodium azide) per Petri dish, and 0.6 TIU aprotinin (Sigma, St Louis, MO, USA). The lysate was clarified by centrifugation at 9,000× g for 20 min at 4°C, and immunoprecipitation was performed with 10 μl anti-IHD-J mouse hyperimmune serum, from our laboratory. The proteins were resolved using 15% SDS-PAGE, identified using a polyclonal antibody (Beiresources), and revealed using the ECL system. Immunofluorescence Protein expression by the recombinant viruses was also examined by immunofluorescence, which was carried out in CEFs and Vero and MRC-5 cells, essentially as described previously [35]. In particular, after infection at 37°C for 1 h with 1 or 3 PFU/cell (depending on the different cytopathogenicity of the recombinants), the cells were grown for 6 or 15-18 h before immunofluorescence. The cells were fixed either with only fresh 2% paraformaldehyde in PBSfor 10 min, for membrane immunofluorescence, or with paraformaldehyde followed by 100% cold acetone for 5 min at -20°C, for cytoplasmic immunofluorescence. The samples were incubated for 1 h with monoclonal or polyclonal antibodies (Beiresources) ( Table 1). FITC 1:100-diluted anti-mouse or anti-rabbit secondary antibodies (Cappel, MP Biomedicals, Inc., Aurora, OH, USA) were used. Cells infected with FPwt were the negative controls. The samples were viewed under a Zeiss Axioskop epifluorescence microscope. IHD-J and preparation of the four FP recombinants mixture The IHD-J strain of VV was kindly obtained from S. Dales (University of Western Ontario, London, Canada) [39] grown in Vero cells, and used to repeatedly infect Balb/c mice (1 × 10 5 PFU/mouse) via the airways, to obtain VV-specific hyperimmune serum. A mixture of the four FP recombinants (4FPmix) was also prepared in CEFs, by co-infection with the four FP viruses (3 PFU/ cell/each recombinant virus). The IHD-J and the 4FPmix viruses were amplified in Vero cells and CEFs, respectively, purified on discontinuous sucrose density gradients, and titrated as already described. Virus neutralisation assays Recognition of the 4FPmix by VV-specific mouse antibodies was tested using the virus neutralisation assay. VV IHD-J hyperimmune mouse serum was used to determine the inhibition of the infectivity of the 4FPmix. Neutralisation assays were performed by pre-incubating the 4FPmix with an equal volume of heat-inactivated hyperimmune serum, for 1 h at 37°C. Pre-immune serum was used as a negative control. The IHD-J virus was used in a parallel test with both pre-immune and hyperimmune serum. The sera were diluted starting from a 1:20 dilution in DMEM. The viral inoculum was adjusted to give approximately 10 2 PFU/Petri dish. Infection was allowed to proceed for 1 h at 37°C. The cells then had 5 ml DMEM with agarose LE 0.7% (SeaKem, FMC BioProducts, Rockland, ME, USA) added. The plaque numbers were counted on day 4 p.i., after adding an agarose layer containing 1.5% neutral red (Gibco). The neutralisation is expressed as the percentage of reduction of plaque numbers versus the control, where the viral inoculum was incubated with no serum. Transcript expression by the FP recombinants is similar in the different cell lines After RNA isolation from the infected CEFs and Vero and MRC-5 cells, the transcripts were detected by RT-PCR after an overnight incubation, as 779-bp, 363-bp, 584-bp and 980-bp fragments in all of the cell lines infected by the FP L1R , FP A27L , FP A33R and FP B5R recombinants, respectively (Figure 1, lanes A). As determined by densitometric analysis, similar levels of expression were observed in the different cells for each recombinant, except for FP B5R , that expressed the B5R mRNA 2.1 times more in human MRC-5 cells than in CEF and Vero cells. The amplification of human β-actin RNA is also shown as a 518-bp band (Figure 1, lanes A). As expected, the FPwt-infected cells used as a negative control did not show any specific band (Figure 1, lanes B). The recombinant proteins are expressed by all of the FP recombinants in all of the different cell lines, although at different levels The FP recombinants expression was verified by Western blotting on lysates of both replication-permissive CEFs and non-permissive Vero and MRC-5 cells, which were infected separately with FP L1R , FP A27L , FP A33R and FP B5R , or with FPwt. The polyclonal primary antibodies recognised specific 27-kDa, 12.7-kDa, 21-kDa and 37-kDa bands, which corresponded to the L1, A27, A33 and B5 proteins, respectively, in the infected CEFs and Vero and MRC-5 cells (Figure 2, lanes B). As determined by densitometric analysis, the expression levels for FP L1R were 4.2-fold and 3.9-fold higher in CEFs and Vero cells than in MRC-5 cells, those for FP A27L were 2.9-fold and 2.5-fold higher in CEFs and Vero cells than in MRC-5 cells, those for FP A33R were 2.6-fold and 3-fold higher in CEFs and MRC-5 cells than in Vero cells, and those for FP B5R were 2.1-fold and 3.1-fold higher in Vero and MRC-5 cells than in CEFs (Figure 2; lanes B). No specific bands were recognised when the cells were infected with FPwt ( Figure 2; lanes A). The corresponding monoclonal antibodies showed similar specificity, but lower binding activity (data not shown). Vaccinia hyperimmune mouse serum recognises FP-expressed heterologous proteins IP/WB was used to test whether the hyperimmune mouse sera could also recognise the FP-expressed proteins. After infection of the Vero cells with the FP L1R , FP A27L , FP A33R and FP B5R recombinants, the specific proteins were immunoprecipitated from cell lysates using sera from mice that had been repeatedly immunised with VV IDH-J ( Figure 3, lanes B). No specific bands were seen for cells infected with FPwt ( Figure 3, lanes A). Different localisation of the foreign proteins is detected by immunofluorescence To lines. Conversely, there was specific, greater, granular perinuclear and cytoplasmic localisation in the cells infected with FP A33R and FP B5R (Figure 4A, 3a-c, 4a-c), which was particularly evident in the MRC-5 cells infected with FP A33R . FPwt-infected cells were always negative ( Figure 4A, 5a-c). Immunofluorescence performed at 6 h p.i. gave similar results (data not shown). Membrane immunofluorescence was present only after infection with the FP A33R recombinant in MRC-5 cells ( Figure 4B, 1b vs. 1a). Polyclonal and monoclonal antibodies were used at different dilutions for the different recombinants ( Table 1). The 4FPmix is not neutralised by VV hyperimmune serum To determine whether the FP recombinant virions produced by the CEFs were carrying the structural products of the four transgenes on their surface, the 4FPmix, containing a mixture of all of the FP recombinants, was incubated with VV IHD-J hyperimmune serum. No reduction in plaque numbers was detected, as compared to samples where pre-immune serum was incubated with the IHD-J virus. High neutralising activity was seen in samples where IHD-J was pre-incubated with IHD-J hyperimmune serum ( Figure 5). Discussion Development of alternative replication-deficient vaccine candidates against smallpox and other zoonotic OPXV infections is still needed as a defence against new poxvirus outbreaks as well as to reduce the adverse reactions of traditional vaccines and the limited immune responses in VV-experienced individuals. Along with discontinuation of the smallpox vaccination campaign and the decline in herd immunity in most countries, an increase in MPXV infections has been described [40,41], which has mostly occurred in children, young adults, and immune-compromised patients [42]. MPXV is widely distributed in a variety of African rodents, and particularly in squirrels, which might be the reservoir [43], and it has the potential to be used as a bio-weapon [41,44,45]. Outbreaks of human MPXV infections were also reported in the Unites States in 2003, with 69 people infected following the importation of MPXVinfected rodents from West Africa. Also, in the Republic of Congo in 1996-1997, there were 92 cases and 3 deaths [46], which might have been related to the overall vanishing immunity against poxviruses throughout the world community. The available smallpox vaccines based on VV show relatively high rates of adverse side effects, and new strategies should therefore be devised to improve the safety of traditional smallpox vaccines. Replicationdeficient vaccines should reduce the risks for animal handlers that arise from accidental exposure to cowpoxvirus and other OPXV, limit the OPXV-infection of domestic animals, and should be useful for protective immunization in the event of intentional or accidental release of variola virus. In the present study, we have described the characterisation of four new FP recombinants, FP L1R , FP A27L , FP A33R and FP B5R , that express the VV L1R, A27L, A33R and B5R genes. We demonstrated that: (i) all of these recombinants can express the proteins correctly, although at different levels in different cell lines, as revealed by RT-PCR and Western blotting; (ii) MV and EV proteins show different subcellular localisations; (iii) recombinant FP virions (4FPmix) are resistant to neutralisation by VV-specific serum; and (iv) the four FP recombinants express functional foreign proteins that are immunoprecipitated by hyperimmune serum from IHD-J-immunised mice. Using Western blotting, heterologous VV-specific proteins were detected in cells infected by the four recombinants. Their expression in human and nonhuman primate cells was often higher than in CEFs, where the virus replicates. This can be ascribed to the cytocidal effect of the FP recombinants in avian cells, although it is not clear why this occurs only for the expression of specific genes. In particular, the higher transgene expression by FP B5R in human MRC-5 cells confirms the results already shown by RT-PCR. IP/WB was used to overcome the difficulty of revealing cellexpressed antigens by conventional Western blotting, especially when using mouse or human hyperimmune sera. This assay, which better avoids nonspecific binding to nitrocellulose and makes the native proteins detectable, showed that these proteins can be recognised by VV-specific hyperimmune mouse serum. By immunofluorescence, the heterologous proteins were mainly localised at the cytoplasmic level for all of the recombinants, and also at the membrane surface when MRC-5 cells were infected with FP A33R . Although no specific differences were observed among the cells lines, the expression of the proteins belonging to MV particles was always lower than the expression of the proteins belonging to EV particles. In particular, the intracellular expression of A33 and B5 was granular, perinuclear and cytoplasmic, which confirms previously reported data, and demonstrates their high expression in the juxtanuclear area and Golgi region [47]. Membrane fluorescence was only detectable after infection with FP A33R . This also confirms previous data showing the spontaneous translocation of the A33 protein to the plasma membrane, which, using a FP B5R recombinant, was only observed in very low amounts by immunoelectron microscopy, but not by immunfluorescence [47]. The very low fluorescence detected after FP L1R infection can either be ascribed to limited protein expression, which was not confirmed by the Western blotting, or to the low avidity of the antibody. This is also confirmed by other studies, where low expression was overcome by the use of the phorbol ester 12-O-tetradecanoylphorbol-13acetate (TPA) [48]. The absence of neutralising activity of IHD-J-specific hyperimmune mice sera against the virus particles released by CEFs infected with the four FP recombinants suggests that the L1, A27, A33 and B5 gene products may not be inserted into the envelope of the new viral progeny. This enhances the importance of these FP recombinants for their use in individuals who have already been vaccinated against smallpox, where these can be administered as a recall. As human peripheral blood monocytic cells and macrophages are permissive for penetration of the recombinants and expression of foreign proteins [28], antigen cross-presentation should occur, which should result in complete humoral and cellular immune responses. Conclusions To avoid lytic infection, ulceration and scab formation after dermal scarification with VV for smallpox prophylaxis, FP recombinants represent alternative safer immunogens. This arises from their natural host-range restricted replication to avian species [49], their correct transgene expression in mammalian cells, and their ability to elicit a complete immune response in vaccinated hosts [50]. Although these L1R, A27L, A33R and B5R VV-derived genes have already been successfully expressed using recombinant DNA [22,48], the use of avipox recombinants expressing these genes might provide better control against zoonotic OPXV infections, will offer advances in the fight against the threat of bioterrorism, and can also be used in VV-experienced individuals.	2017-06-23T18:53:05.285Z	2013-04-11T00:00:00.000	{ "year": 2013, "sha1": "8b91dbd4411631ca1754f02cc703981ea92a5fe6", "oa_license": "CCBY", "oa_url": "https://translational-medicine.biomedcentral.com/track/pdf/10.1186/1479-5876-11-95", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "cbf61db575888ae173173235865564da22da596e", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Medicine", "Biology" ] }
56241555	pes2o/s2orc	v3-fos-license	The Impacts of Brand Personality on Brand Loyalty : A Research on Automobile Brands in Turkey In this pilot study investigating the impacts of automobile brand personality perceptions of Turkish consumers on their attitudinal and behavioral intentions, the effects of the personality attributed to the brand by the consumers on their behavioral intentions regarding preferences, recommendations and willingness to pay higher prices for the brand are examined. The study consists of two main parts. Definitions of brand personality and brand loyalty are made in the literature review section where the conceptual framework is sought to be formed. Our hypothesis to test the attitudinal and behavioral effects of the automobile brand personality following the conceptual framework is analyzed using a sample of 368 participants. The research results indicate that automobile brands are perceived such as competent and excited, and the effects of these dimensions on both behavioral and attitudinal loyalty have been seen to be stronger than the effects on the other two dimensions, namely, conventionality and androgyny. In the conclusion of the study, the impacts are evaluated, and suggestions are given to business managers, marketing researchers, and marketing researchers. Introduction It is a well-known fact that brands are crucial for influencing consumers' loyalty and their purchasing behavior in the marketing process.Businesses benefit from the power of brands to establish a strong bond with their consumers.The consumer-side can obtain information about both the product and the company simply by looking at the brand.Brands reflect a personality and image towards the product they represent.These reflections in the consumer's mind form their purchasing behavior (Perreault et al., 2013). Differentiation of a brand in its product category emerges as an important marketing strategy since brands are similar in terms of price, quality, and distribution (Schneider & Bodur, 2009;Thomas & Sekar, 2008).In order to distinguish the brand, emotional factors should be brought forth.The fact that the abstract and emotional elements forming a brand are more influential for the consumers to position the product and the brand rather than the concrete and rational ones have led firms to focus on abstract and emotional elements (Eisend & Langer, 2007;Aaker, 1997).The use of personality traits in brand positioning appears as an important guide for increasing the consumer's preference, trust and loyalty, and also the consumer's intention to purchase (Aksoy & Özsomer, 2007).At the same time, a good brand personality is advantageous over the brand's competitors by influencing the final decision of the consumer (Sung & Kim, 2010;Swaminathan et al., 2009;Ramaseshan & Tsao, 2007;Freling et al., 2005;Kim et al., 2001). The brand personality, which forms one of the most difficult and complex parts of brand creation, is considered as a strategic instrument for brand management by practitioners and academics.Creating a strong brand relates to the design and execution of a genuine and effective brand personality (Keller, 2003).According to Aaker (1996), brand personality is created by the attribution of personality traits to a brand, which enables consumers to distinguish the various brands (Xue et al., 2007). Whether it is a product brand or an organization brand, the personality characteristics of the brand personality that the company has must be decided.Because, the meaning of a brand in the consumer's life is derived from the positioning of the brand and the image it creates and the image formed by the brand in the consumers' mind depends on the brand personality.Perhaps the most important consideration when creating a brand personality is whether or not there is a close relationship between the personality traits of the target consumers and the brand personality thought to be formed.Because, the brand purchasing behavior and brand loyalty highly depend on the extent to which brand personality is similar to the personality of the consumer's or someone of whom the consumer is a fan. Brand personality is a strategic tool that shapes the brand's communication with consumers.Brands can reflect and communicate themselves through their personalities in the minds of consumers.The brand personality becomes a key concept at this point to establish closer and more loyal relationships with the consumers or to be differentiated in the consumer's mind.Brand personality refers to the common style and attitude that the brand would use for transmitting its message.In this sense, brand personality is a very important factor in establishing an emotional bond.While creating the brand personality, it is necessary to consider the consumer's preference for the brand depending on the situation and conditions and to determine the appropriate image and emotional forms (archetypes). The brand personality tends to be more noticeable when it is based on an archetype, and it has a firmer place in the people's mind (Akın, 2011;Rojas-Mendez et al., 2004;Sen, 2002;Phau, 2001).The findings indicate that the consumers make purchasing decisions in accordance with the various information. The consumers, who have to make a decision without knowledge, would look for various clues in order to correct this situation since they would perceive themselves under risk due to uncertainty.At this stage, the brand and the meaning of brand for the consumer would play an important role in preferences of which they are reminiscent.Although there are numerous factors that affect the attitudes and intentions of the consumer, the personality appears as the most important factor that should be carefully considered by researchers (Akın, 2011).In other words, the personality expressed by the brand can be regarded as a convenient tool to be used as an indicator of whether or not it is "worth acting" in consumer psychology.Plummer (1985) and David Aaker (1996) stated that the brand personality is a significant influence on the competitive advantage and brand loyalty.Similarly, Ratchford and Vaughn (1989) found that personality factors have a decisive influence on attitudes and intentions in the formation of consumer behaviors toward the product and the brand (Shavitt, 1989).In the literature, there are findings claiming that the brand personality is influenced by such behavioral intentions as the consumer's increasing desire to purchase more, desire to pay higher prices, developing the brand preference and paying compliments to the brand (Akın, 2011;Kim et al., 2001;Hayes, 1999).Such studies in the literature constitute the focal points of this study in which the effects of consumers on the loyalty of brand perception are investigated. According to data obtained from the Automotive Distributors Association (ADA) in 2016, the automotive sector, which is one of the fastest growing and developing sectors in Turkey, has broken a new record with 983,720 vehicle sales.48 automotive firms operating in Turkey have broadened their product lines via product differentiation and product diversification.Automobiles are very important and strategic products in Turkish people's life.Purchasing automobiles is one of the biggest expenditures that the consumers make personally throughout their lives.Therefore, it is very important to make the right decision in the selection of automobiles.In Turkish society, the automobile is perceived as a reflection of modern life and free individual.Pointed out that Turkish consumers are comprised of people who pay attention to the quality and reliability of the products, who mostly consider the brand as the most important criterion in automobile purchasing decisions, who believe that the quality of the brand is related to the price of the product, and who make research on the brand in the process of purchasing an automobile.From this point of view, the market share and brand awareness of the automotive sector and the impact of brand personality on the brand loyalty will be examined.How consumers perceive the automotive brands they use or what they want to use in the future, what personalities they attribute to the brands, and how brand loyalty is measured are important issues that will provide the consumers with the opportunity to forecast future brand preferences.Therefore, in this context, research is decided to be carried out about the impacts of the personalities that the consumers attribute to the brand in the automotive sector on their loyalty. It is possible to claim that the positive, unique and distinctive brand features attributed to the brand have positive impacts on brand loyalty.When brand personality and consumer's character match, the consumer naturally chooses this brand to purchase (Li & Zhang, 2011).Much of the work in the area of consumer behavior is usually within the discipline of psychology, and a significant part of the research focuses on the relationship between personal characteristics and behavioral attitudes (Smith, 2012).Loyalty, switching brands, paying more, complaining and recommending tendencies be consumer behaviors rated within behavioral attitudes (Zeithaml, Berry, & Parasuraman, 1996).Lin (2010) found that positive brand personality perceptions influenced brand loyalty positively in the study on the relationship between brand personality and loyalty (Ekinci & Riley, 2003;Waller et al., 2006).In Lin's study, the "competence", "sophistication", "compatibility" and "clarity" dimensions of brand personality are shown to have a positive influence on attitude loyalty while "competence", "calmness" and "sophistication" dimensions have a positive effect on behavioral loyalty (Lada et al., 2014;D'Astous & Lévesque, 2003;Smit et al., 2002;Sung & Tinkham, 2005).Based on this information, the research hypothesis is as follows: H 1 : Brand personality has a positive impact on brand loyalty. Brand Personality and Measurement of Brand Personality Personality, as an important component of your identity, is also an important element of communication.Because a brand without personality cannot identify itself and it cannot remain in mind.Consumers are under the influence of similarities between their personalities and products when choosing among competing products (Özer, 2015;Rathnoyake, 2008).For this reason, each brand has to apply an image that reflects its style as "consistent" and "one-sided" as a reminder.This leads to an increase in the size of the relationship between the consumer and the brand, thereby increasing the consumer's preference rate for the brand. In this way, the consumer's confidence in the brand and the increase in loyalty also lead to a permanent differentiation that is not easy to be imitated by the rival brands (Aksoy & Özsomer, 2007, Diamantopoulos et al., 2005).In order to provide this relationship, it is necessary for the consumers to adopt a brand and personalize the brand to establish a relationship between the brand and themselves (Escalas & Bettman, 2005). Brand personality increases the personal meaning of the product for the consumer and ensures that the consumer is synonymous with the product.King (1970) and Plummer (1984) focused on this issue in their research conducted on consumer behavior.King (1970) emphasized that brand personality be an important element when the consumers choose between the two brands.It has been expressed that the consumers have chosen brands in a way similar to the selection of friends, that is to say, they attribute human characteristics to the brands (Cui et al., 2008;Mengxia, 2007).Plummer claimed that the consumers tend to pass judgment on a brand in defining it as if it was a living person with the help of questions such as "How would it look?Where would it live?What would it do?What kind of magazines would it read?What would it wear?What would it talk to and about what in a party?".Whether brands appeal to rational or emotional aspects, it is well known that the brand personality has a strong influence on the purchasing decisions of consumers.One of the important dimensions of the brand personality is that it allows the consumers to express the difference they want to present to others by using the branded products or by shopping at the related brand/company (Schneider & Bodur, 2009). The first study on the concept of brand personality in the literature is conducted by Gardner and Levy (1955).J. Aaker (1997, p. 347) defines brand personality as "a group of human characteristics associated with a brand" (Keller and Ailawadi, 2004).J. Aaker developed a generally accepted, valid and reliable scale (Rogos-Mendez, 2004;Nilson, 1999;Rojagopal, 2005), namely, "Big Five" (Thurstone, 1933;Goldberg, 1993), a scale pertaining big five theory adapted to marketing to be used in measurement of brand personality (Caprara et al., 2001;Bosnjak et al., 2007;Milas & Mlačić, 2007;Sweney & Brandon, 2006;Fennis & Pruyn, 2007;Rekom et al., 2006).As a result of Aaker's work, the scale with originally 114 characters has been transformed into a scale with 42 personality features, 15 sub-dimensions and 5 basic dimensions (Cui et al., 2008).The results of this study revealed that five dimensions used to measure brand personality in the United States are comprised of sincerity, excitement, competence, development/sophistication and robustness (Lin, 2010;Romaniuk, 2008).Several studies based on this scale have tested the validity and reliability of the constituent characters and dimensions (Azoulay & Kapferer, 2003;Morschett et al., 2008). Although Aaker described the brand personality at the individual brand level and it is said that the individual brands can be used in the measurement of the brand personality, the five-dimensional structure is thought to be more suitable for a structure composed of different brand and product categories.Researchers need to be very cautious if they believe that each has the same five-dimensional structure of the brand or that a simplified measure can be used to measure the personality of a single brand.What can be done in this case would be the re-observation and adaptation of Aaker's scale consisting of 42 variables and five dimensions (Cui et al., 2008, pp. 534-535).Thus, the study aims to measure automobile brand personality as a general category, not the brand personality of a single automobile brand.Aksoy and Özsomer (2007), as one of the studies conducted on validity and reliability of the brand personality scale for Turkey, have transformed the scale developed by Aaker into a structure consisting of 39 variables and four dimensions in accordance with the virtues and needs in Turkey.This study also utilizes the scale adopted by Aksoy and Özsomer for Turkey.Table 1 below is comprised of the dimensions and adjectives that constitute the scale.Besides the similarity with other studies on the subject in terms of dimensions, the included adjectives are unique to Turkey.Adjectives in Table 1 below are also used in the study. Brand Loyalty The increasing importance of the brand concept led to the examination of many concepts related to it (Çabuk & Orel, 2008). Brand loyalty is at the forefront of these concepts (Gounaris & Stathakopoulos, 2004).Every company wants to achieve the creation of loyal consumers to their brands.It will be possible to maintain the market share with the stability to be attained at this point.Firmalar aims to create a holistic consumer experience and rank all kinds of contact points with consumers in order to strengthen brand loyalty.Each form of interaction communicates with the user, and each detail such as image, icon, placement or button that the users face plays an important role in the connotations and judgments about the company.Gobe (2010) mentioned about the emotional economy in which people interact more with brands at this point.Emotional brands not only give support to who we are, but also physically provide us with the opportunities to transform into what we desire to become.Oliver (1997) defined loyalty as a continuous purchase of products and services made by consumers, and unchanged purchasing decisions of the consumers despite whatever reason due to any external factors (Kwong & Candinegara, 2014).Mowen (1998) defined brand loyalty as the customers' positive attitude towards a brand, their intimate feelings towards the brand and their desire to continue purchasing it in the future while described it as the strength of the consumer's confidence in the brand.Another definition of brand loyalty is "the tendency, desire, and action of intimate and non-random shopping in which the customer involves a particular company, vendor or goods & services within an environment where the customer has other alternatives". Brand loyalty is the measure of the consumer's commitment to the brand, and it forms the basis of the brand value (Supphellen & Grinhaug, 2003).Brand loyalty can, in general, be defined as "the positive attitude and behavioral response of the consumer to one or more brands in a product category over a period" (Engel et al., 1990, Rai & Medha, 2013).Given the fact that all these definitions are combined in the point of repurchase, preference, recommendation and engagement; the companies' need of loyal customers for the sustainable competition is better understood.Loyalty is seen as one of the key elements to achieve permanent and long-term success in terms of businesses. The formation of brand loyalty offers advantages in terms of businesses such as reducing marketing expenditures, being more dominant at retailer points, attracting new customers and allowing time for competing for activities.The brand loyalty, which causes positive communication among the consumers and reduces the consumers' resistance to the competitive strategies, contributes to the process that enables companies to reach more consumers (Dick & Basu, 1994). Brand loyalty is the loyalty of the consumer towards a brand which leads to re-purchasing of the brand, not only in the present period but also in the future (Kim, Lee, & Suh, 2015). Repetitive purchasing behavior and positive attitude towards a brand or company are considered as important indicators for the loyalty of consumers (Dick & Basu, 1994;Lee, Kim, & Kim, 2006).Sheth (1974) criticized the definition of loyalty as repetitive purchasing behavior by stating that different loyalty situations may be relevant for different customer profiles and different product groups, as far as the individual's ability to develop loyalty towards goods or services he/she has never bought.Brand loyalty is expressed as a function of psychological processes and a behavioral response (Jacoby & Kyner, 1973).Taking repetitive purchasing behavior into consideration as a measure of brand loyalty causes the consumers to ignore their feelings towards the brand, the reason why they constantly purchase the brand and whether or not they like it.It is very important to determine whether merely habits and conveniences or emotional ties to the brand cause the purchase.Consumers can sometimes keep purchasing the same brands because they are either low in price or do not have a better alternative.Therefore, concepts of 'brand loyalty' and 'repetitive purchasing' should be considered different from each other.The brand loyalty is used by the consumer to express that there is a real preference for the purchase of a brand, while repetitive buying behavior refers to the purchase of the same brand despite the absence of any emotional bond.Therefore, some researchers defined this as a brand loyalty if consumer's attitude towards a brand is more affirmative than competitive brands (Datta, 2003;Uzunkaya, 2016).Brand loyalty; providing a competitive advantage by creating an effective barrier against competitors, increases the company's ability to respond to competitive practices.Because of this, brand loyalty has been evaluated as a concept to be dealt with both behavioral and attitudinal dimensions (Kim, Lee, & Suh, 2015;Rai & Medha, 2013).These two dimensions, which are used in the measurement of loyalty, can be utilized separately in the measurement of consumer loyalty, as well as they are used together for the mixed loyalty studies in the literature. Behavioral loyalty points out a strong tendency to repurchase, while attitudinal loyalty expresses the emotional and psychological side of loyalty (Donnelly, 2009).In studies that deal with loyalty at the behavioral dimension, loyalty is treated as repetitive purchasing behavior only at certain time intervals.According to this approach, brand loyalty is defined as "the tendency to choose and buy a single brand consistently among many brands in the same product group", meaning that, loyalty is measured according to the repetition of purchasing (Sheth et al., 1999).Customers with repetitive purchasing behaviors need to be separated from the customers who prefer that brand to others and make purchasing in accordance with that preference. In order to make this distinction, the extent of consumer loyalty must be taken into consideration.The attitudinal approach in brand loyalty is determined by the presence of a positive attitude towards the brand in consumers.In the formation of these positive feelings, consumers' evaluations such as "satisfaction, commitment and integration" play an important role towards the brand.It can be said that the consumers have been motivated by these tendencies in the realization of purchasing behavior.In other words, the attitudinal approach focuses on the consumer's pure and genuine loyalty sensation which is beyond the actual causes that affect customers' choices of purchase, and how customers exactly perceive and appreciate the brand (Gounaris & Stathapoulos, 2004). Methodology This study aims mainly to determine the consumers' existing brand loyalties towards automobile brands and to predict how the consumers' brand preferences would be formed in the future.The automotive sector, which has developed rapidly in Turkey in recent years, has been dealt with in order to understand the purchasing behavior patterns of the consumers along with their attitudes and behaviors towards the brands. These studies also provide the enterprises with the opportunity to benefit from the correct management of marketing activities in the related sectors.In doing so, they tend to diversify products by constantly enriching their product lines.Since automobile purchases are among the largest amount of expenditures made by individual consumers, it is crucially important for the consumers to make the right decisions in automobile selection.With this study, it is aimed to measure the impacts of brand characteristics of automotive brands operating in Turkey on brand loyalty. Survey Design The main population of the study consists of the automobile owners living in Niğde.The survey is conducted on April-May 2015 using face-to-face questionnaire method with automobile owners.Questionnaires are drawn up from the related literature.Within this framework, firstly, four-dimensional scale with 39 variables adapted to Turkey by Aksoy and Özsomer (2007) is used to measure the brand personality.The scale developed by Şimşek and Noyan (2009), Back and Parks (2003), and Mano and Oliver (1993) is adapted for this study in order to determine brand loyalty components.Both scales are prepared with a 5-point Likert scale (1 -strongly disagree, 2 -disagree, 3 -undecided, 4 -agree, 5 -strongly agree).The questionnaire consists of three parts; brand personality scale is used in the first part, and brand loyalty scale is used in the second part.In the last part, the questions on both the demographic characteristics of the survey participants and the brands in the automotive sector are included. Sampling and Data Collection After the decision is made to carry out the research study on the automotive sector, the automobile owners in Niğde are determined as the target group.The survey data are collected by face-to-face questionnaire method on April-May 2015.For the application of the questionnaire, convenience sampling method is preferred among non-random sample methods.368 out of 400 questionnaires are found suitable for the analysis after the elimination of carelessly filled questionnaires with significant missing data.The survey availability rate is 92%, and this figure is considered to sufficiently represent the population.When the demographic distribution of the research sample is considered, it can be seen that the number of female automobile owners is 9%, and male owners appear to be of the dominant gender with a ratio of 91% in Turkey same as the rest of the world.44.1% of the research sample is composed of consumers under 35 years of age.33.3% of the participants has an average household income of 5000 TL or higher, and 51% of them are employed, both husband and wife.43.2% of the survey participants are employed in the public sector, 39.6% in the private sector and 17.2% are retired.The distribution of automobile brands owned by the participant automobile owners in Niğde province is as follows; 24.5% Tofaş-Fiat, 21.2% Renault, 14.6% Volkswagen, 13.1% Ford, 9.4% Opel and 17.2% other brands. Analysis and Findings The reliability of the two different measures used in the study is examined first, followed by the exploratory factor analysis for brand personality and brand loyalty scales.The reliability and validity test results of the scales used in the study are presented in Table 1 below.The reliability of the brand personality scale used in the study is calculated as .955.Reliability of the scale used in the research is evaluated by looking at the Cronbach Alpha coefficient which shows the internal consistency of the variables forming the scale.When the Cronbach Alpha coefficient is 0.60 or less, which vary between 0 and 1, the results of internal consistency reliability are unsatisfactory.In general, the lower limit of the Cronbach Alpha coefficient is assumed to be .70.As the number of variables increases, the correlation coefficient between these variables also increases, thus the reliability of the scale increases (Hair et al., 1998).From here, it can be said that the reliability of the efforts to measure brand personality is sufficient. The reliability of our scale used to measure brand loyalty is .831.The reliability of both scales is higher than.70 which is considered acceptable in social sciences (Nunnally, 1978).Furthermore, the variance values explained on both scales are calculated as .67 and .66.The definitions with low factor loads in the first analysis are excluded from the model as suggested by Hair et al. (2011;2014, p. 102).In conclusion; the factor load above the recommended limit of 0.50 for all expressions is obtained, and this is shown in Table 1.As a result of the exploratory factor analysis, the related variable is excluded from the variables that constitute the attitude loyalty measure since the factor load of the provision "I feel more attached to the automobile brand than other brands" is below .50.Additionally, in Aksoy and Özsomer's studies, the adjectives such as doing well, self-confident, original which are under the dimension of competence in the brand personality scale; and also amusing, cheerful, sympathetic, youthful, agile, passionate under the dimension of excitement are excluded from the original scale since their factor loads are below .50.It is observed that brand personalities of automobile brands are perceived as in quality, professional, global and reliable under the dimension of competence.When the dimension of excitement is examined, it is seen that automobiles are perceived as a libertarian, lively and active. It can be said that this is mainly due to the effect of similar themes being processed in automobile brand advertisements in recent times.It is surprising that Turkish consumers do not perceive automobile brands as traditional, but interestingly, they perceive automobiles, which are essentially high-priced products, to be moderate, modest and thrifty when the dimension of conventionality is considered.In the dimension of androgyny; masculinity and rebelliousness are on the forefront, indicating that males loom large as the dominant gender when it comes to the automobiles in Turkey, same as the rest of the world.The fact that the vast majority of participants being male also has an effect on this result. In the study, regression analysis is performed to measure the effects of brand personalities attributed to automobile brands by the consumers on their attitudinal and behavioral loyalties.When the analysis results are examined, the correlation coefficients of competence, excitement, conventionality and androgyny traits with attitudinal and behavioral brand loyalties are calculated as .764and .934,respectively (CI is checked along with VIF and Tolerance values since the correlation coefficient in behavioral loyalty is so high that it can point out multicollinearity problem.All three values indicate that there is no multicollinearity).Both correlation coefficients indicate a strong correlation between experience and loyalty.Our determination coefficients show that automobile brands with dimensions of competence, excitement, conventionality and androgyny have a power of explanation of the attitudinal and behavioral loyalties of the consumer at the ratios of .584and .873,respectively. It can be said that the first step of the hypothesis is fulfilled, meaning that brand personality has an impact on brand loyalty of the consumer.However, it is necessary to check the significance of this impact from ANOVA table.The ANOVA chart on both scales indicates that these impacts are significant.The standardized loads (β) and significance levels (p), which explain the impacts of the personality features attributed to automobile brands by the consumers on attitudinal and behavioral loyalties and/or the relationships between them, indicate that H 1 hypothesis claiming the existence of a significant interaction between brand experience and brand loyalty is to be accepted within all dimensions. One of the purposes of regression analysis is to make projections for the future.This requires the formulation of a mathematical regression model.The mathematical regression model is formulated as follows: (2) The generated mathematical regression model indicates that it is possible to predict the consumer's attitudinal and behavioral loyalties when any knowledge about the dimensions is acquired.When beta coefficients are examined, it is noted that the dimension of androgyny has a negative relationship with the dimensions explaining behavioral loyalty, and this dimension has a very small relationship coefficient with the attitudinal loyalty.It can be emphasized that it possibly due to most people being of the male gender regarding automobile ownership.Similarly, when beta coefficients are examined, the impacts of other dimensions of brand personality on behavioral loyalty are seen to be higher than on attitudinal loyalty.Consequently, it can be said that the brand personality is better at explaining the repetitive purchasing behavior of the consumer, while it somewhat fails to explain the attitudinal loyalty which expresses emotional and psychological loyalty. Conclusion Today, as the economy and competition conditions change, businesses have to change and evolve in order to continue their lives.Technological developments have led to intense competition to reduce or even completely abolish physical differences.It is now necessary to differentiate products from each other on the basis of the meanings to which they are attributed, not their physical properties.Ascribing various meanings to products begins with branding them.A successful brand differentiation can be achieved through personality development. Brand personality enables consumers to perceive the brand as friends by offering emotional benefits to the consumers.A well-established brand personality enhances brand choice and use, as well as strong emotional bonds to be built on it, would be creating brand trust and confidence. In order to keep up with the changing market conditions and to reach the consumers who constantly change their purchasing tendencies and expectations, it is obligatory for the enterprises to implement different strategies.They need to apply the appropriate branding strategies in their production process in order to be able to capture competitive advantage and bring out their difference.The brand is important both for the product to be positioned correctly and for the consumers to be able to ascribe meaning to these brands.Nowadays, when consumers purchase a product, the connotation that the brands make the consumers feel becomes important since they also purchase it besides the quality of the product. Products can be associated with personal characteristics and meanings ascribed to their brands by the consumers.For this reason, brand personality is important regarding improving the quality perception of the product.The brand personality, which is expressed as the human characteristics that the consumers attribute to the brand, provides many advantages for the companies.One of them is the formation of brand loyalty.Since a well-established perception of brand and brand personality can create an emotional connection between the consumer and the product, the formation of loyalty is inevitable.The concept of brand personality provides benefits to the brand and the company in a variety of ways.The most important benefit of brand personality is distinguishing brands from competitor brands. This decomposition would make it easier for the consumer in the decision-making phase to make a choice.At this stage, the meaning of the brand and the consumer's preferences with which they associate play an important role.The correct brand personality created within the framework of the needs of the target group can be easily recognized by the consumers, and consequently, the brand can be preferred.The preference of a brand leads to the concept of loyalty, which determines the brand's power and is the common goal of all brands.The objective of brand personality is to ensure that the brand is preferred by influencing the consumers' emotional decisions whenever they find it difficult to decide.Brand personality also provides the company with a sustainable competitive advantage.An original brand personality to be created against the competitor brands can create a competitive advantage by keeping them on the frontline in their positioning and promotional activities.The brands are personality indicators for the consumers by whom they are preferred. The brand is an important part of product policies.The fact that the brand personality can be measured can lead to the brand positioning and advertisement message formation of the brand's managers as well as the anticipation of brand expansion.The positioning of a brand personality based on competence and enthusiasm for global brands would make it possible for the brand to be demanded in different markets.Brand personality based on competence and excitement would allow the same messages to be successfully used in different markets. In this pilot study investigating the impacts of consumers' perception of automobile brand personalities on their attitudinal and behavioral intentions, detection of the impacts of personality attributed to the brands by the consumers on their preferences, recommendations, and behavioral intentions as if they are willing to pay higher prices for the brand is prioritized. Since it is known that the personality is mainly influenced by the cultural factors, the personality features adapted by Aksoy and Özsomer are utilized in measuring the personality attributed to the brands by the Turkish consumers.As a result of the factor analysis, it has been seen that automobile brand personality occurs in four dimensions.The first two dimensions emerged as competence and excitement. It is surprising that the automobile brands are perceived as competent in terms of the competence dimension and also as moderate and modest in terms the conventionality dimension.Also in the aspect of androgyny, it can be said that the automobile is predominantly masculine and rebellious. Reliability analysis is performed primarily in the study and then Cronbach alpha coefficient is found to increase to .70 in both brand personality dimension and brand loyalty dimension in general.The reliability of the study appears to be satisfactory, as long as the value matters.Also, it is seen that the scale and questions used in the questionnaire survey cover the research study as a result of the factor analysis. Regarding the research, as the result of regression analysis performed in order to determine whether or not brand personality has an impact on brand loyalty, it is found out that the dimensions of brand personality have explanatory power for both the behavioral and attitudinal loyalties at rates of .873and .584,respectively. Consumer perceptions of brand personality can also increase brand loyalty.The behavioral intentions of the consumers are considered together with their attitudinal and behavioral dimensions, and it is seen that the brand personality is very effective when the impact on intensions towards the automobile is focused on. In the automobile market where global brands are based on market share, both the behavioral and attitudinal intentions of the competence dimension are more influential than other dimensions in explaining the fact that global brands position themselves correctly. The impact of the dimension of the androgyny on both behavioral and attitudinal intentions seems to be rather limited.As a natural expectation, when it comes to technological and global products, it is preferable not to be perceived as traditional.Perception of automobile brands is thought to be effective in improving positive attitude and behavior.It is believed that continuing their activities to support this perception may be effective in improving the positive attitudes and behaviors towards the automobile brands and also in the formation of the brand loyalty.It is believed that new companies would increase their chances in competition by emphasizing similar features.As a result, knowing the perceptions of the brand personality, which is a significant influence on the brand preferences, would contribute to accurate and effective brand management decisions to be made by the companies.Positive perceptions towards the brands can be used as an element to create a significant competitive advantage over competitor brands contingent upon preference.Although a generalization would not be possible in Turkey due to the constraints of the pilot study, the data obtained and the results are considered to be useful for similar future studies. Table 1 . Dimensions that constitute brand personality in Turkey Table 2 . Exploratory factor analysis and reliability results	2018-12-18T04:30:39.669Z	2017-03-10T00:00:00.000	{ "year": 2017, "sha1": "e818db2907c45185dddcefdd73ca8bfd8b69b402", "oa_license": "CCBY", "oa_url": "https://www.ccsenet.org/journal/index.php/ijms/article/download/66962/36330", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "e818db2907c45185dddcefdd73ca8bfd8b69b402", "s2fieldsofstudy": [ "Business" ], "extfieldsofstudy": [ "Business" ] }
16552657	pes2o/s2orc	v3-fos-license	The therapeutic effects of SET/I2PP2A inhibitors on canine melanoma Canine melanoma is one of the most important diseases in small animal medicine. Protein phosphatase 2A (PP2A), a well conserved serine/threonine phosphatase, plays a critical role as a tumor suppressor. SET/I2PP2A is an endogenous inhibitor for PP2A, which directly binds to PP2A and suppresses its phosphatase activity. Elevated SET protein levels have been reported to exacerbate human tumor progression. The role of SET in canine melanoma, however, has not been understood. Here, we investigated the potential therapeutic role for SET inhibitors in canine melanoma. The expression of SET protein was observed in 6 canine melanoma cell lines. We used CMeC-1 cells (primary origin) and CMeC-2 cells (metastatic origin) to generate cell lines stably expressing SET-targeting shRNAs. Knockdown of SET expression in CMeC-2, but not in CMeC-1, leads to decreased cell proliferation, invasion and colony formation. Phosphorylation level of p70 S6 kinase was decreased by SET knockdown in CMeC-2, suggesting the involvement of mTOR (mammalian target of rapamycin)/p70 S6 kinase signaling. The SET inhibitors, OP449 and FTY720, more effectively killed CMeC-2 than CMeC-1. We observed PP2A activation in CMeC-2 treated with OP449 and FTY720. These results demonstrated the potential therapeutic application of SET inhibitors for canine melanoma. Melanoma accounts for 3% of all tumors and about 7% of malignant tumors in dogs, with the primary tumor sites being the oral cavity (56%), lips (23%), skin (11%) and digits (8%) [5,11]. The malignancy of canine melanoma varies dependent on the site of the primary lesion, with oral melanoma as the most common malignancy in dogs [1]. Radiation therapy is applied either alone or as an adjuvant therapy after surgery, but only a limited number of facilities can apply this treatment [2]. Although the literature reported that some canine melanomas respond to carboplatin, the ratio of complete response is quite low [7]. Therefore, the identification of novel therapeutic targets is required to improve treatment outcomes. Protein phosphatase 2A (PP2A) is a well conserved serine/threonine phosphatase and an important tumor suppressor [20]. PP2A inhibits the tumorigenic potential, such as disordered cell proliferation and cancer cell stemness, by suppressing tumor-promoting signals, such as ERK1/2, Akt/ PKB and c-Myc [16,20]. SET/I2PP2A is a potent physiological inhibitor of PP2A, and the expression of SET is increased in various human cancers including breast cancer, pancreatic cancer, colorectal cancer and chronic myeloid leukemia [8,9,14,18]. In human melanoma, SET knockdown (KD) suppressed tumor growth [8]. We previously reported that canine cells express several isoforms of the SET protein and showed that canine SETα protein has approximately 94% homology with human SETα [3]. OP449 is a peptide antagonist of SET, which directly binds to SET, leading to dissociation of PP2A from the inhibitor complex and allowing recovery of PP2A activity [4]. OP449 has been reported to decrease the tumorigenic potential of human B-cell chronic lymphocytic leukemia and breast cancer cells, and induce apoptosis [4,14]. Previously, we reported OP449 recovers PP2A activity and induces apoptosis in canine lymphoma cells [10]. FTY720 (Fingolimod) is a sphingosine analog used as an immunosuppressant in human multiple sclerosis patients. FTY720 is phosphorylated by sphingosine kinase 2, and phospho-FTY720 downregulates sphingosine-1-phosphatase receptor 1. The nonphosphorylated form of FTY720 has been reported to directly interact with SET and recover PP2A activity [17]. These reports suggested the efficacy of SET inhibitors for canine tumors, however, the role of SET in tumorigenic potential of canine melanoma has not been investigated. Here, we show that knockdown of SET expression in a canine melanoma cell line leads to decreased cell proliferation, invasion and colony formation. OP449 and FTY720 recovered PP2A activity and showed anti-tumor effects on canine melanoma. Our data demonstrate the potential clinical application of SET inhibitors for canine melanoma. Analysis of cell proliferation: 5.0 × 10 3 cells were seeded on 24-well plates, and after 3 days, Cell Counting Kit-8 (CCK8, Dojindo, Kumamoto, Japan.) was used to analyze cell proliferation according to the manufacturer's instruction. Matrigel invasion assay: To examine cell invasion, invasion assay was performed by using transwell chambers (Costar, Cambridge, MA, U.S.A.). After the polycarbonate membranes with an 8 µm pore were coated with 2% matrigel (BD Biosciences, San Jose, CA, U.S.A.), 1.5 × 10 4 cells of CMeC-1 or 4 × 10 4 cells of CMeC-2 in 200 µl media were added on the membrane (upper chamber). 880 µl of medium with 10% FBS was added on the lower chamber and cultured for 24 hr. The membranes were fixed with methanol for 15 min and stained with Giemsa solution (Muto Pure Chemicals, Tokyo, Japan). After the membranes were washed with distilled water, non-invaded cells were wiped with cotton-swab. The number of invaded cells through the membranes was randomly counted in 3 fields (× 400) under a light microscope (Eclipse TE2000, Nikon, Tokyo, Japan). Analysis of effects of SET inhibitors: 5.0 × 10 4 cells were seeded on 12-well plates and treated with OP449 for 1 day or FTY720 for 2 days. Cells were detached with 100 µl of trypsin-EDTA and trypan blue solution was added in an equal volume. Surviving cells were counted by microscopy. PP2A activity assay: PP2A activity assay was performed as previously described Statistical analysis: The results are expressed as means ± S.E. Student's t test was used for comparison between two groups. Groups more than 3 were compared using one-way analysis of variance, after which Fisher LSD test was used. For all analyses, a probability value of P<0.05 was considered statistically significant. RESULTS The protein expression levels of SET in 6 canine melanoma cell lines (CMeC-1, CMeC-2, KMeC, LMeC, CMGD2 and CMGD5) were examined by immunoblotting ( Fig. 1A and 1B). We observed relatively low level of SET protein in CMeC-2 and CMGD5. CMeC-1 is the primary tumor cell lines from skin origin, and CMeC-2 is the lung metastasis origin from CMeC-1 xenotransplantation. CMeC-2 expresses about 30% less SET protein than CMeC-1. We used these cells to generate cell lines stably expressing nontarget shRNA (shNT) and SET targeting shRNA (shSET). As shown in Fig. 1C, shSET effectively suppressed SET protein expression in both CMeC-1 and CMeC-2. We examined the effects of SET knockdown (KD) on tumorigenic potential of canine melanoma cell lines. Interestingly, SET KD inhibited cell proliferation (Fig. 2) and invasion (Fig. 3) of CMeC-2 (low SET expression), but not CMeC-1 (high SET expression), suggesting SET plays an important role in the tumorigenic potential of CMeC-2, and higher protein level of SET is not correlated with the effects of SET KD. Consistent with these data, the ability to form colonies with CMeC-2, but not CMeC-1, was suppressed by SET KD (Fig. 4). Immunoblotting was performed to clarify the cell signaling affected by SET KD (Fig. 5). We observed Thr412 phosphorylation level of p70 S6 kinase (corresponding to human Thr389) was suppressed by SET KD in CMeC-2, suggesting the involvement of mTOR (mammalian target of rapamycin)/p70 S6K signaling. Phosphorylation levels of ERK1/2, Akt/PKB and p65 NFκB, and protein levels of β-catenin and c-Myc were not different between shNT and shSET, indicating these signaling pathways are not affected by SET KD. Next, we examined the anti-tumor effects of SET inhibitors, OP449 and FTY720, on canine melanoma cell lines. OP449 dose-dependently decreased cell survival rate for both CMeC-1 and CMeC-2 (Fig. 6A). OP449 had more potent effects on CMeC-2. Consistent with this, FTY720 effectively killed CMeC-2, but did not affect cell survival rate of CMeC-1 (Fig. 6B). We further examined whether SET inhibitors restored PP2A activity. Phosphatase activity in extracts of cells treated with or without 2.5 µM of OP449 or 10 µM of FTY720 for 2 hr was assayed with a relatively specific phosphopeptide (K-R-pT-I-R-R) as a substrate. As shown in Fig. 7, OP449 and FTY720 increased PP2A activity about 1.5 fold in CMeC-2, but did not change PP2A activity in CMeC-1. These data suggest that SET inhibitors effectively antagonize canine SET and exert anti-tumor effects. DISCUSSION In this study, we observed SET KD decreased tumorigenic potential of SET-low expressing CMeC-2, but not SET-high expressing CMeC-1. Consistent with this observation, SET inhibitors more effectively killed CMeC-2 melanoma cells. Although increased SET expression and its correlation with adverse prognosis were observed in various human tumors [6, 8], our unexpected results suggest that a higher level of SET expression does not always mean a higher contribution for tumorigenic potential. The molecular mechanism of this complexity remains unknown, but genetic mutations or differences in post-translational modifications on SET may contribute to this mechanism. Species differences may also play a role in this result, because dogs specifically express a carboxy-terminal truncated version of SET, which can associate with PP2A [22], but may not associate as well with other tumor suppressors. We observed that SET KD decreased cell proliferation, invasion and colony formation of CMeC-2, indicating an important role of SET in the tumorigenic potential of CMeC-2. It has been reported that SET increases c-Myc protein levels [14], but we did not observe decreased c-Myc expression by SET KD in CMeC-2. ERK1/2, Akt/PKB, NFκB, Wnt/βcatenin and mTOR/p70 S6 kinase signaling are frequently upregulated in tumors, and PP2A has been reported to suppress these tumor promoting signals [16]. Among them, we observed decreased phosphorylation levels of p70 S6 kinase Thr412 (corresponding to human Thr389) by SET KD, which is the target of mTOR. Going further, it has been reported that PP2A directly associates with p70 S6 kinase, and knockdown of one of the regulatory subunits of PP2A enhances p70 S6 kinase Thr389 phosphorylation [12,21]. These results suggest that SET KD restored PP2A activity which suppressed p70 S6 kinase phosphorylation, leading to decreased tumorigenic potential of canine melanoma cells. We tested a potential therapeutic role for SET inhibitors, OP449 and FTY720, in canine melanoma. Previously, we reported that OP449 selectively induced apoptosis for SET high-expressing canine lymphoma cells [10]. However, in this study, both OP449 and FTY720 have more potent effects on SET low-expressing CMeC-2 cells. While protein levels of SET may be a biomarker for disease severity, these results do not support that SET levels predict the response of SETtargeting drugs in canine melanoma. Interestingly, effects of OP449 and FTY720 on CMeC-1 were different, i.e. 5 µM of OP449 killed almost 60% of CMeC-1 in 24 hr, but even treatment with 10 µM of FTY720 for 48 hr could not induce cell death, although it suppressed cell proliferation (data not shown). Because FTY720 is phosphorylated by sphingosine kinase 2, and phospho-FTY720 downregulates sphingosine-1-phosphatase receptor 1, a lack of potency, a lack of penetration into the cell or a PP2A independent mechanism may produce this inconsistency. Nonphosphorylated forms of FTY720, such as FTY720-Ome and FTY720-regioisomer, may provide effective tools to resolve this lack of effect. ACKNOWLEDGMENTS. This work was partly supported by a Grant-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science and Technology (Grant#25660227). The funding source had no role in the study design; collection, analysis or interpretation of data; in the writing of the manuscript; or the decision to submit the manuscript for publication.	2016-08-09T08:50:54.084Z	2015-06-11T00:00:00.000	{ "year": 2015, "sha1": "33d9519972c68553d0b480c32129ff990c228d50", "oa_license": "CCBYNCND", "oa_url": "https://www.jstage.jst.go.jp/article/jvms/77/11/77_15-0193/_pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "33d9519972c68553d0b480c32129ff990c228d50", "s2fieldsofstudy": [ "Biology" ], "extfieldsofstudy": [ "Biology", "Medicine" ] }
233944653	pes2o/s2orc	v3-fos-license	A likelihood ratio test for the homogeneity of between-study variance in network meta-analysis Background Network meta-analysis (NMA) is a statistical method used to combine results from several clinical trials and simultaneously compare multiple treatments using direct and indirect evidence. Statistical heterogeneity is a characteristic describing the variability in the intervention effects being evaluated in the different studies in network meta-analysis. One approach to dealing with statistical heterogeneity is to perform a random effects network meta-analysis that incorporates a between-study variance into the statistical model. A common assumption in the random effects model for network meta-analysis is the homogeneity of between-study variance across all interventions. However, there are applications of NMA where the single between-study assumption is potentially incorrect and instead the model should incorporate more than one between-study variances. Methods In this paper, we develop an approach to testing the homogeneity of between-study variance assumption based on a likelihood ratio test. A simulation study was conducted to assess the type I error and power of the proposed test. This method is then applied to a network meta-analysis of antibiotic treatments for Bovine respiratory disease (BRD). Results The type I error rate was well controlled in the Monte Carlo simulation. We found statistical evidence (p value = 0.052) against the homogeneous between-study variance assumption in the network meta-analysis BRD. The point estimate and confidence interval of relative effect sizes are strongly influenced by this assumption. Conclusions Since homogeneous between-study variance assumption is a strong assumption, it is crucial to test the validity of this assumption before conducting a network meta-analysis. Here we propose and validate a method for testing this single between-study variance assumption which is widely used for many NMA. Background Network meta-analysis (NMA) is an approach to combining evidence from multiple studies of multiple interventions and obtaining estimates of all possible intervention comparisons using indirect and direct evidence. Common approaches to network meta-analysis include a fixed effect model and a random effects model. The random effects model assumes that the true effect size can differ from study to study, because the effect size in each study is derived from a random distribution of effect sizes. Several assumptions about the data generating mechanism are made in network meta-analysis. Common to the fixed effect model and random effects model is the exchangeability assumption. The exchangeability assumption relates to the study populations and states that the randomized trials are similar on average, in all important factors other than the intervention comparison being made [1,2]. The exchangeability assumption implies the consistency condition is valid [3], i.e., the relative effect of A to B, can be derived from the difference using data from C compared to A and C compared to B for any treatments A, B, and C. A commonly used assumption unique to the random effects model is a single betweenstudy variation for all treatments [4]. Assuming that all effects sizes across all treatments have the same betweenstudy variation is a strong assumption. However, there are applications of NMA where the single between-study assumption is potentially incorrect and instead the model should incorporate more than one between-study variance estimate. A few approaches have been proposed to allow different between-study variation across treatment comparisons. Lu (2009) proposed a Bayesian approach to modeling between-study variance structures under the consistency assumption [5]. White (2012) proposed a partially structured heterogeneity model that allows for two between-study variances but did not have a practical reason for doing so [6]. Although these approaches have been proposed, the single between-study variation assumption remains widely used. In practice, there is a lack of guidance for when the homogeneous assumption should be challenged. The decision to assume one or more between-study variance should be informed primarily by the reviewers' knowledge of the data generating mechanism. However, the results from statistical testing, comparison of results of the NMA under both assumptions and the magnitude of variance estimates can also support any decisions made about the structure of between-study variance. Recently we conducted several network meta-analyses of interventions to prevent bovine respiratory disease in feedlot cattle, where the assumption of a single betweenstudy variance was questionable based on our knowledge of the biology of the disease and interventions included in the meta-analysis. Turner et al. [7] found heterogeneity might be related to the type of comparison and models with heterogeneous variances have been proposed with different informative priors under the Bayesian framework [8]. However, this is not applicable in frequentist framework. Additionally, limited work has been reported on testing the assumption of a single between-study variance across all treatment comparisons. Therefore, the objective of this project was not to model the betweenstudy variance structure, but to develop an approach to testing the homogeneity of between-study variance in a network meta-analysis based on the likelihood method. For network meta-analysis, several different methods of calculating the single between-study variance have been proposed [9][10][11]. However, we were unable to identify any commonly used approaches to testing this assumption compared to an alternative that two or more between-study variances exist based on a characteristic of the underlying studies. The sequence of the paper is as follows: • Section 2: The motivating example • Section 3: The model and proposed likelihood ratio test (LRT) • Section 4: The evaluation of the LRT using two methods • Section 5: Discussion of the evaluation and application. Motivating example The motivating example involved bovine respiratory disease, a multi-agent disease of cattle. Bovine respiratory disease (BRD) is the most economically important disease of feedlot cattle and therefore knowledge of the comparative efficacy of interventions to prevent, control and treat BRD is critically important. One common approach to preventing bovine respiratory disease is to administer antibiotics to all cattle at arrival at the feedlot. The aim of administering antibiotics at arrival is to preemptively treat animals with sub-clinical BRD and to prevent BRD in animals at risk. Trials conducted to assess how effective antibiotics are for this purpose, use the proportion of treated animals detected with BRD after a period of time, usually 28 days, as the outcome. The data available for assessing the comparative efficacy of antibiotics for this purpose included comparisons of antibiotic to antibiotic, and comparisons of an antibiotic to notreatment. For BRD prevention, the assumption of a single between-study effect for both types of comparisons is biologically questionable. It is known that some antibiotics are highly effective at treating and preventing BRD because the mechanism of action is very broad spectrum. An example of such a group of antibiotics is the macrolide group. Antibiotics in this group have consistent high quality evidence of low BRD risk after 28 days when administered at arrival [12,13]. This means that trials that compare a macrolide to a macrolide would be expected to have a comparative effect size near zero, if the effect size is measured as the log odds ratio (log OR). The between-study variation of macrolide to macrolide trials is therefore expected to be small. However, for trials that compare a board spectrum antibiotic, such as a macrolide, to a non-treated control, the expected variation in the effect size is much larger, because the risk of BRD in the 1st 28 days in cattle is highly variable in non-treated cattle. The data suggests that some groups of untreated cattle have close to zero animals detected with BRD after 28 days while other groups have 50% or more animals with BRD. The result of this naturally expected variation in BRD risk in the 1st 28 days of feedlotting in non-treated animals is a wider variation in the comparative effect sizes when active drugs such as macrolides are compared to non-treated groups. For example, if the macrolide is highly effective, we expect that the number of animals treated for BRD after 28 days will be close to zero regardless of the underlying risk of BRD in the group. However, the non treated group may have anywhere from zero to 100%. When these data are converted to a distribution of the comparative effect sizes (log OR), it is natural that more variation is expected between these active to no-treatment trials than the trials that are macrolide to macrolide. There are several other scenarios in BRD, where the assumption of a single between-study variance for all comparisons is questionable. For example, to prevent BRD in animals arriving at the feedlot, antibiotics or vaccines might be used. As with a no-treatment group, the response to vaccination is highly variable, yet the response to broad spectrum antibiotics like, macrolides is highly consistent. Therefore in a network of evidence that compared the efficacy of antibiotic and vaccines to prevent BRD, we would naturally expect the vaccine to vaccine comparisons to be more variable than board spectrum antibiotic to broad spectrum antibiotic comparison. It is these examples, that motivated the work described below. The likelihood for a random effects model of network meta-analysis under consistency assumption This section provides the basic model form used for formulating the likelihood ratio test. In the following, we consider T treatments that are compared in I studies each with n i arms. The set of treatments included in study i is given by T i . Let y i denote the estimates of relative effects for the ith study, y i = (y i,1 , ..., y i,n i −1 ) T and y = (y 1 , ..., y T ). The study specific treatment effects of study i are given by θ i where θ = (θ i , ..., θ I ). Then we have where i represents the vector of errors of study i. i is assumed to be normally distributed and independent across studies and its covariance is cov and is a scalar if study i only has two arms. The distribution of y is where S is a block diagonal matrix with each block S i , i = 1, ..., I. As the consistency assumption is made in the random effects model, all treatment effects are uniquely determined by T − 1 basic treatment comparisons with a common reference (usually a placebo). These basic parameters are denoted by the vector d. The relative effect size of all other possible treatment comparisons in the network are called functional parameters which can be obtained from the basic parameters. For example, if d 1,2 and d 1,3 are basic parameters in the network, then d 2,3 , a functional parameter, can be obtained by Let X denote the design matrix of size I × (T − 1). Each row of X corresponds to one study specific comparison and the columns represent the basic comparisons and . 1, 0, and -1 are the possible values in the design matrix. If one row of X only has one element of 1 and other elements are 0, then this study specific comparison is a basic comparison. If 1 and -1 occur in one row, then the relative effect parameter of the corresponding comparison is a functional parameter. For each study i, the design matrix is denoted by X i . Then, where δ i is the vector of between-study heterogeneity of study i. The random effects model usually assume δ i to be normally distributed. If study i only has two arms, then where the values of the diagonal elements of V i are τ 2 and off-diagonal values are τ 2 /2 [5,14]. The values of the off-diagonal elements are determined by the assumption that every source of direct evidence has the same between-study variance. The distribution of θ is where V is a block diagonal matrix with each block V i , i = 1, ..., I. The between-study heterogeneity is assumed to be independent of within-study errors. Hence, the marginal distribution of y is If we know τ 2 , then the maximum likelihood estimate of d iŝ Likelihood ratio test for the between-study variance parameter Here we discuss an approach to testing the assumption of a single τ 2 . Based on our motivating example, the between-study variance parameter for non-active to active treatment comparisons and active to active treatment comparisons are denoted by τ 2 n and τ 2 a respectively. The hypotheses to be tested are The log-likelihood function under the null hypothesis is Under the null hypothesis, the structure of V i is discussed in section 3. There are two potential data forms for V i under the alternative hypothesis. If study i only contains active treatments, then the values of diagonal elements of V i are τ 2 a and off-diagonal values are τ 2 a /2. If non-active controls are included in study i, then the diagonal values (variance) are τ 2 n and the off-diagonal values (co-variance) are τ 2 n − τ 2 a /2. For example, suppose study i is a three-arm trial that compares a non-active control (denoted by N) with two active treatments (denoted by A 1 , A 2 ). The between-study variance-covariance matrix for study i is To make the variance-covariance matrix semi-positive definite, the covariance should follow the following inequality: To meet this inequality the following constrains are placed on τ 2 n and τ 2 a : Here a three-arm trial is used to illustrate the covariance matrix structure and the constrains. Similar structures and the same constrain are applicable to trials with more than three arms. The likelihood ratio test (LRT) statistic is where the estimates of the parameters are the maximum likelihood estimates. The asymptotic distribution of this test statistic is χ 2 1 . Givenτ 2 , the maximum likelihood estimate ofd iŝ Real data implementation and simulation results The data used are from a network meta-analysis of antibiotic treatments for BRD in feedlot cattle [15]. The evidence network consists of 204 trial arms from 98 studies. Eight of the 98 trials have three arms. The total number of participants in all studies is 26,132, with the number of participants in a study ranging between 34 and 1726. Among the total 26,132 participants, 9467 had the event. There are 13 treatments in the network: non-active control (NAC), ceftiofur hydrochloride (CEFTH), ceftiofur bollus in pinna (CEFTP), ceftiofur sodium (CEFTS), danofloxacin (DANO), enrofloxacin (ENFO), florfenicol (FLOR), gamithromycin (GAMI), oxytetracycle (OXY) used at multiple doses, tildipirosin (TILD), tilmicosin (TILM), trimethoprim (TRIM), and tulathromycin (TULA). The outcome is the log odds ratio of the proportion of treated animals detected with BRD. A negative log OR means treatment benefit for the numerator treatment compared to the referent. The within-study variance is obtained using delta method. For example, in a 2-arm study with reported number of events r 1 and r 2 and sample sizes N 1 and N 2 , the within-study variance is calculated by 1/r 1 + 1/(N 1 − r 1 ) + 1/r 2 + 1/(N 2 − r 2 ). The number of pairwise comparisons is 106 in total with 66 non-active control to active treatments (N2A) comparisons and 40 active to active treatments (A2A) comparisons. The network plot is shown in Fig. 1. The size of the node is proportional to the number of arms and the thickness of the edges represents the total size of direct comparisons between each treatment pair. The number in the parentheses after a treatment abbreviation is the number of studies containing that treatment. The absence of a line means that there is no direct comparison in the evidence network. To evaluate the performance of the proposed LRT, two methods have been used. The first method is based on the Fig. 1 The network plot of the treatment arms for bovine respiratory disease in feedlot cattle. The size of the node represents the magnitude of the number of arms and the thickness of the edges represents the total size of direct comparisons between each treatment pair asymptotic distribution (χ 2 ) of the LRT statistic and the second method is established on the Monte Carlo simulation. Maximum likelihood estimation is applied to obtain the basic effect size parameters and τ 2 under the null and alternative hypothesis. We simulated 1000 data sets under the null hypothesis being true (a single between-study variance for all treatment comparisons) to assess the type I error rate and another 1000 data sets where the alternative hypothesis was true (two between-study variance, one for N2A and one for A2A) to evaluate the power given the significance level of 0.05. Under the null hypothesis, the simulated data y H 0 is generated from the real data y by whered H 0 is the maximum likelihood estimate givenτ 2 . Since the LRT statistic under the null hypothesis follows a chi square distribution when the sample size goes to infinity, we also assessed the type I and power for the scenario where the number of studies is five times the original to determine if the type I error can be well con-trolled when the sample size per comparison is larger. This increased-size dataset has the same network structure as the real data. For example, in the original network, there is no study comparing treatment TRIM with NAC, and this is also the case in the simulated network. Only one study compares TRIM with TILM as shown in Fig. 1, whereas for the increased-size data set there are five studies simulated for this comparison. Assessing type I error rate and power of the test based on the chi square distribution For each simulated dataset where the null hypothesis was true (a single between-study variance for all treatment comparisons), the maximum likelihood estimates were obtained and the LRT statistic calculated. The proportion of these 1000 LRTs that are beyond the 95% quantile of the χ 2 1 distribution is the estimated type I error rate. The power can be obtained by applying the same procedure on each simulated dataset where the alternative hypothesis is true (two between-study variance, one for N2A and one for A2A). Assessing type I error and power of the likelihood test based on the Monte Carlo simulation An alternative approach to the chi-square approach is a simulation based approach to testing. This procedure is as follows: 1 For each simulated dataset where the null hypothesis is true, the maximum likelihood estimates are obtained under both hypotheses and LRT is calculated, denoted by LRT b (b ∈ {1, ..., 1000}). For estimating power, the only change is to use each simulated dataset under the alternative hypothesis being true in the step 1. Results The values of τ 2 observed in the original BRD dataset are shown in Table 1. The p value of the likelihood ratio test based on the asymptotic distribution of the test statistic is 0.028 indicating a significant difference between the two heterogeneity parameters but the type I error rate inflates in this case. The simulation-based p value is very close to 0.05. Hence, making decision only relies on the cut-off of 0.05 for the p value of the LRT is not convincing. The heterogeneity parameters values estimated under two models are meaningfully different. The estimated between-study variance for the non-active control to active treatments comparison is four times larger than that for active to active treatments. This difference would have an impact on the confidence intervals of the relative effects of the comparisons in the network, especially for comparisons with fewer studies. Then the Wald 95% confidence interval ofτ 2 n −τ 2 a is calculated and given by (0.0282, 0.8469) which indicates a significant difference from 0. The effect of models with different heterogeneity parameters on the point estimates and confidence inter- Table 1 Estimates of τ 2 from the analysis of the a meta-analysis network for bovine respiratory disease treatments using maximum likelihood estimation vals of the relative effect sizes, are presented in Fig. 2. Figure 2 shows the 95% confidence intervals of the log odds ratios of the treatment pairs presented in the network plot under the models with one and two betweenstudy variance parameters. Treatment comparisons that involve only one study which has small study size tends to have wider confidence interval because of the large within-study variance. It can be seen in Fig. 2 that some confidence intervals change markedly in width under the different models. Some of the point estimates of the relative effect sizes shift because of the change of estimates in between-study variances which would vary the weight of direct and indirect comparisons. The estimate of τ 2 of N2A comparison in two τ 2 s model is greater than that in one τ 2 model and the τ 2 of A2A comparison is opposite. Therefore, the width of confidence intervals tends to be narrower for A2A comparisons in the two τ 2 model than in the one τ 2 model. Also, most of the point estimates of the effect sizes of N2A comparisons shift to the right under the two τ 2 model . It is not easy to predict the direction of the change of the point estimate of effect size or the width of the confidence interval in the two τ 2 model for each comparison since it is a mixed weight change of direct and indirect comparisons. The results of the study the likelihood ratio test performance in Table 2 shows the type I error rate and the power analysis results. The simulation based on the original data is labeled (60, 40) to indicated the number of studies. While the increased size data is labeled (330, 200). The increased-size data set have the same network structure as the real data. For the asymptotic distribution of the test statistics, the type I error is above 5%, i.e., 8.3%. Increasing the number of studies reduced the type I error drop to 5%, i.e., 4.4%. While in the Monte Carlo simulation-based evaluation, the type I errors are controlled in both settings. The power was suitable for all methods and datasets. By combining the results in Table 2 with those in Table 1, we can say there is a significant difference between the heterogeneity parameter of nonactive control to active treatments comparisons and of active to active treatments comparisons. In practice, if the p value of the LRT statistic is very close to the cut-off (i.e., 0.05 in this paper) like in this example, depending on p value only to make decision is not conclusive. Visual inspection of the results from the two models and how these results differ is helpful in reaching a conclusion. Conclusions We have proposed a likelihood ratio test for testing the homogeneity of the between-study variance parameter for the random effect network meta-analysis model. We illustrate this method with an example for testing the homogeneity between the non-active control to active treatments comparisons and of active to active treat- Fig. 2 The approximate 95% confidence intervals of the log odds ratios of the treatment comparisons presented in the network plot under the models with one and two heterogeneity parameters. The comparisons on the y-axis in blue are non-active control to active treatment comparisons. Those in black are active to active comparisons ments comparisons. Our example applied this likelihood ratio test in a network meta-analyses which contained a non-active control (or placebo or no-treatment) and our understanding of the biology of this example, raised concerns about the single between-study variance esti- mate. There are many other situations that this method can be applied, for example, the between-study heterogeneity for a pharmacological treatment vs surgery comparison might be different from that of a comparison of two pharmacological treatments. We also developed the variance-covariance matrix structure of the model with two heterogeneity variance parameters. In the motivating example, we applied the test and found the significant difference of the between-study variance of two types of comparisons. We have explored two ways to define the p value based on the same LRT statistic, one using the asymptotic χ 2 distribution and the other using a Monte Carlo simulated sampling distribution. In practice, we would recommend using the Monte Carlo p value, which has a better control of the type I error, especially when the number of studies is limited. The estimation method for the basic parameters and between-study variance is MLE. There are many literature comparing different methods of estimating the between-study variance parameter [16][17][18][19]. Different estimators may have different distributions and our method is based on the MLE. That is not to say MLE is the best estimator but we just propose a possibility that the between-study variance may not be the same across all comparisons and we use MLE and likelihood ratio test to show the single heterogeneity parameter assumption may not hold in some cases. Tests for this assumption using other estimators are possible extensions. Our likelihood ratio test is developed based on a model where the consistency condition is considered valid. If the consistency condition is not met, alternative models can be used to address inconsistency and the likelihood ratio test can be developed under the new model in an analogous fashion. Testing the homogeneity of between-study variance in network meta-analysis with inconsistency is an interesting topic that we leave as a possible future work.	2021-05-08T00:03:02.717Z	2021-02-24T00:00:00.000	{ "year": 2021, "sha1": "a7e13088d346c9ff82ee36bf04498ee1413fb61c", "oa_license": "CCBY", "oa_url": "https://systematicreviewsjournal.biomedcentral.com/track/pdf/10.1186/s13643-021-01859-3", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "416bc9434cb478b3714abfaabf0b8367b9f34d3c", "s2fieldsofstudy": [ "Mathematics" ], "extfieldsofstudy": [ "Medicine", "Mathematics" ] }
248863236	pes2o/s2orc	v3-fos-license	BodyMap: Learning Full-Body Dense Correspondence Map Dense correspondence between humans carries powerful semantic information that can be utilized to solve fundamental problems for full-body understanding such as in-the-wild surface matching, tracking and reconstruction. In this paper we present BodyMap, a new framework for obtaining high-definition full-body and continuous dense correspondence between in-the-wild images of clothed humans and the surface of a 3D template model. The correspondences cover fine details such as hands and hair, while capturing regions far from the body surface, such as loose clothing. Prior methods for estimating such dense surface correspondence i) cut a 3D body into parts which are unwrapped to a 2D UV space, producing discontinuities along part seams, or ii) use a single surface for representing the whole body, but none handled body details. Here, we introduce a novel network architecture with Vision Transformers that learn fine-level features on a continuous body surface. BodyMap outperforms prior work on various metrics and datasets, including DensePose-COCO by a large margin. Furthermore, we show various applications ranging from multi-layer dense cloth correspondence, neural rendering with novel-view synthesis and appearance swapping. Introduction Several fundamental problems related to human understanding in images can be addressed by labeling every pixel covering the human body with semantic information. This enables numerous applications including video analysis, image editing, texture generation and style transfer. From a single RGB image of a human, the literature has proposed methods to extract sparse information such as 2D body keypoints (e.g., face, hands, body joints), or 2D segmentation masks (e.g., for full body, clothes, hair or skin), and also 3D body pose and shape parameters defined by a template body model [6,8,14,43,47,60], while work on dense surface correspondence has further enabled pixel-*This work was conducted during an internship at Meta RL Research. Figure 1. We introduce BodyMap -a method that establishes accurate dense correspondences between a 2D image and the surface of a 3D clothed human with high precision. Our approach handles loose clothes, different hairstyles and various accessories, like hats and bags, providing crisp silhouettes, and works well in multi-person cases with occlusions. level understanding by establishing unique correspondences between 2D pixels covering the visible regions of the human body and 3D points on the surface of a body template. In the seminal work DensePose [35], correspondences are estimated between image pixels belonging to the human body and points in disjoint parts of a human body template located using UV coordinates, similar to local texture mapping. The method is trained on the large in-the-wild dataset DensePose-COCO and is robust to human pose variability, image resolution, diversity in clothing, and occlusions. However, it has some inherent limitations that im-pact methods that rely on it (e.g., for clothed-human applications) [1,27,62]. First, the discretization generated by dividing the body into disjoint parts produces clearly visible seams and discontinuities between them that are not optimal for learning models. Second, the DensePose estimates suffer from inaccuracy as reported in prior work [35,37,52], mainly due to the difficulty in acquiring ground-truth annotations for the task [2,30]. Follow-up methods have tackled some of its shortcomings and a few recent works addressed the discontinuity of UV maps [4,36,63]. HumanGPS [56] proposes to predict per-pixel embeddings using geodesic distances between corresponding points on the surface of a 3D human scan and does not produce an explicit mapping. None of the proposed approaches has established high-definition correspondences for areas with finer details such as hair and hands (with fingers), with generalization to clothed humans, especially with loose clothing. In this work we introduce a novel technique to establish high-definition full-body and continuous dense correspondence between images of clothed humans and the human body surface. Our method, which we term as BodyMap, takes as input an RGB image of a human and outputs accurate per-pixel continuous correspondence estimates for each foreground pixel (i.e. including the full body, with clothes and hair). We designed a transformer-based architecture that learns appearance-based and Continuous-Surface-Embeddings-based representations to infer accurate dense surface correspondence for the depicted human. Our variant of Vision Transformer [12] as a computational block of the encoder brings its advantageous properties for dense prediction tasks. The vector dimension is kept constant throughout all processing stages as well as global receptive field for every stage. With these properties, our network is well designed for dense correspondence prediction. Furthermore, we capitalize on the power of synthetic data. Since no real-world dataset provides ground-truth annotation at the quality we aim for (fingers, clothes, hair), we created a synthetic dataset of animated 3D clothed human scans. In that way, we obtained ground-truth dense correspondence for a large variety of humans with diverse clothing, in different poses and from different viewpoints. A differentiating factor of our framework is that it is not tied to a human body with topology constraints, and can handle layered representations such as humans with separate cloth geometries. To summarize, our key contributions are: • BodyMap is the first method to establish dense continuous correspondence for every foreground pixel of clothed humans, whether that is fingers, hair, or clothes that are displaced from the human body with high-precisionsomething that all prior works fail to achieve. • A novel transformer-based architecture designed specifically for this task that when trained in a multi-task learn-ing manner with per-pixel classification losses for each channel significantly outperforms prior works across several datasets and tasks. • We achieve state-of-the-art results on DensePose COCO by a large margin. We show our approach can be applied to real-world applications such as novel view synthesis. Our method can be extended to learn layered representations with clothed humans and predict per-geometry surface correspondences. Related Work Dense Surface Correspondences. One of the most widely used approaches in this topic is DensePose [35], where classification and regression branches were trained to obtain per-pixel body part and UV estimates. The body parts constitute the I channel which takes one of 25 values (including the background) and the UV estimates which are continuous numbers mapped to [0, 255]. However, its output is discretized resulting in seams between body parts. This problem is alleviated in Continuous Surface Embeddings (CSE) [36], which for each pixel learns a positional embedding of the corresponding vertex in the object mesh. In CSE correspondences are learned without being constrained on specific geometry types (e.g., humans), and show the effectiveness of their approach on other deformable object categories, like animal classes which was later extended by discovering correspondences between different object classes automatically [38]. HumanGPS [56] maps each pixel to a feature space, where the feature distances reflect geodesic distances among vertices of a 3D body model corresponding to every pixel. Similarly to CSE, for every image pixel they produce an embedding capable of differentiating visually similar parts and aligning different subjects into an unified feature space. Zeng et al. [63] introduced, a model-free 3D human mesh estimation framework, which explicitly establishes the dense correspondences between the mesh and the local image features in the UV space. They solve human body estimation problem relying on dense local features transferred to the UV space. Getting enough labeled data (especially non-synthetic) to learn dense correspondences is a challenging task. SimPose [68] proposed to alleviate the problem by using simulated multi-person datasets and a specific training strategy with multi-task objectives to learn dense UV coordinates. They obtain favourable results using only simulated human UV labels. The intricacy of getting dense and accurately annotated correspondences is further explored in UltraPose [61]. They provide a dense synthetic benchmark focusing on faces, containing around 1.3 billion corresponding points as well as data generation system based on novel decoupling 3D model. Architecture Designs for Dense Correspondences. There have been a couple of approaches in terms of network archi- Figure 2. BodyMap architecture. Given an RGB image we first obtain its CSE [36] estimates and feed both to their corresponding encoders. We utilize vision transformers specifically designed for this task to learn to extract accurate high-dimensional representations that are then fed to the BodyMap decoder that predicts per-pixel dense correspondences. tectures to extract dense human correspondences. In Dense-Pose [35] a Mask-RCNN [16] with Feature Pyramid Features [29] is utilized to obtain accurate image features. Sim-Pose [68] opted for a ResNet-101 backbone trained with losses adjusted to each of their tasks (e.g., human pose, segmentations, normals, UVs). Another simple yet effective choice employed by HumanGPS [56] is an Encoder-Decoder architecture such as U-Net [48]. Our investigation indicated that while one can achieve satisfactory results with the aforementioned approaches they are all unable to capture finer-level details in the depicted human as usually the extracted features are too coarse. To alleviate this we turned into transformer architectures due to their ability to learn these discriminative features necessary for either downstream computer vision tasks or reconstruction applications. Originating from Natural Language Processing, a Transformer architecture [58] has shown its effectiveness within a wide range of Computer Vision tasks: image recognition and classification [12], image retrieval [13], image generation [40] and image captioning [17]. We capitalize upon prior work on vision transformers for dense prediction tasks [45] (e.g., depth estimation) and introduce a new architecture explicitly design for predicting dense surface correspondences for humans. The Proposed Method: BodyMap The main goal of the proposed approach is to establish dense surface correspondence between a single RGB image and 3D body model. Our method takes as input a single RGB image, foreground mask and coarse correspondences retrieved using Continuous Surface Correspondences (CSE) [36]. CSE serves as a sufficient initialization which our method refines by providing more accurate estimates for the areas covering loose clothes, hair, fingers, etc. Thus, BodyMap provides per-pixel estimates for the foreground image resulting in much more accurate represen-tations and crisp silhouettes. The necessity of foreground mask stems not only from the foreground silhouette that we aim to complete with our estimates but also the image-level features that we prove to be essential in Section 4. Continuous Correspondences Continuous correspondences have significant advantages over their discretized counterparts. First, a continuous representation provides no seams between body parts. Second, it is conceptually simpler as there is no need to explicitly encode and later predict the body part. The benefits of utilizing a continuous representation for surface correspondences have already been discussed in a few prior works [36,56,63]. We follow a similar direction with [36] and design a continuous UV map that is then warped to body models in different poses, providing ground truth correspondences for our approach. The color scheme for correspondences used in the paper is unique color-wise: we chose different colors for every vertex of the parametric body model. Given the colored 3D body model we transform its surface into a 4K UV map, which is then utilized during rendering over a body model in a determined pose and from a desired view point. In that way, we obtain ground truth for the synthetic data used for training. Surface Embedding Transformers A classic architectural for a network predicting Dense Correspondences from an RGB image is an Encoder-Decoder (e.g., U-net). While simple convolutional backbones in the encoder can usually provide sufficient results, we observed that the right choice of the encoder architecture may significantly boost the whole pipeline performance. Compared to convolutions, transformer-based architectures do not suffer from limited receptive fields, resulting in more expressivity. Moreover, transformers avoid explicit downsampling of the input image embedding leading to more ac-curate and refined final representations. As illustrated in Fig. 2, we build upon the work of Ranftl et al. [45] for monocular depth estimation and introduce a simple yet novel transformer-based architecture designed explicitly for the task of predicting dense surface correspondences of humans. We transform the RGB image and its CSE estimate into tokens by extracting non-overlapping patches and then linearly projecting resulting flattened representations. Similarly to text-transformers, we add a specific token to the set, that aggregates the global knowledge about an image. The image and CSE embeddings are supplemented with positional embeddings and fed to separate vision transformer backbones with separate weights to retrieve dense features for each input. Later we refer to these blocks as appearance and correspondence transformers (Fig. 2). Positional encoding in Visual Transformers is essential to capture sequence ordering of input tokens instead of transforming the image into "bag-of-patches" omitting its relative order and global spatial consistency. The transformer outputs are fused forming an intermediate representation which is first resampled and then projected via residual convolutional units. It is then fed into the convolutional decoder where the representation is upsampled to generate a fine-grained correspondence prediction. Finally, the network outputs per-pixel RGB values that encode correspondences according to our coloring scheme discussed in the previous section. Supervision in the Image Space For each pixel p in the foreground image, we predict 3channeled (RGB) color p ∈ Z 3 which represents the correspondence (the colors in such a representation are unique which makes subsequent warping easy). Thus, we treat the whole problem as a multi-task classification problem where each task (predictions for the R, G and B channels) is trained with the same set of losses: Per-pixel classification loss L cls . For every color channel, we predict the per-pixel classification label l ∈ [0, 255]. BodyMap provides raw, unnormalized per-pixel scores for each of the classes in each of the three color channels and L cls measures the cross-entropy between the prediction and the ground truth. Since we noticed that it is quite challenging to predict correspondences of realistic gestures, we further define a loss weight for each pixel based on the body part segmentation. We set a higher weight for hands and head while a lower weight for the rest of the body to encourage fine-grained correspondence estimation. Silhouette loss L sil . We penalize the model for nonaccurate silhouette predictions by calculating the IoU between the predicted and ground truth foreground masks. Supervision in the 3D Geometry Geodesic loss L geo . While per-pixel cross-entropy classification losses supervise our predictions in 2D image space, we expand our supervision scheme to 3D by utilising geodesic distances on the surface of the body model. Geodesic losses have been instrumental in the literature for enforcing supervisions in the 3D space. We design a loss that pushes features between non-matching pixels apart, depending on the geodesic distance. We calculate geodesic distances between vertices predicted with correspondences for every foreground pixel and their ground truth counterparts. Theoretically, such a supervision eliminates imperfection of the proposed coloring scheme for correspondences: distant vertices may have resembling colors (green head and shoulders, blue arm and right thigh). Thus, the geodesic loss provides extra knowledge about the 3D geometry comparing distances between predicted vertices vs. ground truth ones. where V (I(x)) denotes the vertex corresponding to pixel location x in the image I and D g (·, ·) denotes the geodesic distance between two 3D points on the body surface. Regularization and Final Loss Consistency loss L con . We further add a regularization term to enforce the smoothness of the predictions in the neighboring regions. Specifically, we constrain the predictions from neighboring pixels to be geodesically close to each other, i.e., where p r is a randomly chosen pixel within the foreground silhouette, D g (p 1 , p 2 ) the geodesic distance between vertices corresponding to pixels p 1 and p 2 , σ geo is the normalizing constant for geodesic distances (maximum possible distance between points in the body model), σ col the normalizing constant for RGB colors, respectively. On each iteration we calculate this loss 100 times for different randomly chosen pixels later averaging the resulting values. Final loss. The final loss is a weighted sum of all the terms: where a loss weight λ corresponds to each loss term in order to balance them. Training Details The BodyMap network is first trained on synthetic data to learn surface correspondences for every foreground pixel. Given an RGB image, we obtain foreground mask and CSE estimates which serve as an initialization for the correspondences. However, if we were to test this model directly on DensePose-COCO that comprises multiple people, heavy occlusions and low-resolution images, then the results would be unsatisfactory. The annotations provided in this dataset are sparse and noisy with ∼100 pixel-SMPL vertex correspondences for each person in the image. To bridge this domain gap, we fine-tune our model on the training set of DensePose-COCO but with a key change that ended up having a significant impact. Given an image from this dataset, we generate pseudo ground-truth estimates on the fly by extrapolating both the available ground-truth annotations but also the CSE initialization such that they cover the whole estimated silhouette of the human. In that way we can fine-tune our model on real-data with denser supervision and utilize losses in both 2D and 3D spaces. To further boost the generalization capabilities of BodyMap, we introduce several augmentations. First, we do specific crops in order to get upper-body samples. Second, we generate frames with multiple synthetic people in it to simulate crowds and diminish the gap between synthetic and real data. Third, we do a standard set of augmentations, like rotations, slight hue and saturation changes. Experiments Datasets. Our proposed approach is trained mainly on synthetic data with the exception of the experiments reported on DensePose-COCO where we utilize the provided training set. We opted for the RenderPeople dataset [11] which has been used extensively in the literature [1,5,7,18,19,27,39,42,49,56,69] for various human reconstruction and generation tasks. We used 1000 scans which are watertight meshes wearing a variety of garments and in some cases holding objects such as mugs or bags. Since the scans are static we wanted to introduce additional pose variations and as a result we performed non-rigid registration, rigged them for animation and used a motion collection that provides 3D human animations from which we collect a set of 2, 446 3D animation sequences covering wide action categories of daily activities and sports. With a large set of scans and motions we randomly sample scan-motion pairs and render them with Blender Cycles from different views with uniform lighting to obtain the RGB sequences as well as the corresponding UV ground-truth. We perform a 90/10 train/test split based on identities. This large-scale dataset represents an effort to cover a wide range of motions, poses, and body shapes, captured from multiple views with people that can move towards the camera our even outside the frame and enables us to train our BodyMap network without making any explicit assumptions. At test-time BodyMap is evaluated quantitatively and qualitatively on both synthetic as well as real data ranging from COCO, fashion images (DeepFashion [32]) as well as a few 3dMD scans of real people captured with a full-body scanner. We used this solely for testing since we wanted to evaluate to what extent our approach can handle the domain gap between synthetic and real data. These real scans do not Average Precision (AP) and Recall (AR) on DensePose-COCO. AP and AR are calculated at a number of GPS thresholds ranging from 0.5 to 0.95. Our methods surpasses the state-of-the art methods DensePose [35] and CSE [36] include any objects but are noisier with complex facial expressions and enable us to stress-test whether our approach can handle such complex inputs. Baselines and Metrics. We consider two different ways of measuring the quality of correspondences evaluating both in 2D image space by comparing RGB values of the corresponding pixels and in 3D space by measuring geodesic distances between predicted and ground truth vertices. First, we calculate the accuracy of predictions in the 2D image space by calculating the percentage of pixels colored correctly within a specified threshold. Second, following the evaluation scheme of DensePose that is used widely in the literature [35,36,68] we measure average precision and recall over GPS scores. Geodesic point similarity (GPS) score is a correspondence matching score: where P j is the set of points annotated on person instance j, i p is the vertex estimated by a model at point p,î p is the ground-truth vertex p, and κ is a normalizing parameter. We calculate Average Precision (AP) and Average Recall (AR) metrics considering a vertex prediction as correct if the GPS score is higher than a threshold. Following the evaluation scheme established by prior work [35,36], GPS thresholds are ranging from 0.5 to 0.95. Additionally to metrics in 2D and 3D spaces, we evaluate the consistency over time of our predictions in order to estimate quantitatively the amount of flickering. We calculate percentage of positive correspondence matches between frames of the same video for visible vertices. Using the aforementioned metrics we compare BodyMap quantitatively to the previous works: Dense-Pose [35], CSE [36], SimPose [68] as well as several other baselines. However, calculating AP and AR metrics for HumanGPS is not possible due to the fact, that HumanGPS predicts only embeddings for every foreground pixel, that provide no information on UV coordinates or corresponding to pixels SMPL vertices. In their approach warping and appearance swapping is done by nearest neighbors search Quantitative results In Tables 2 and 3 we provide a quantitative comparison between BodyMap, DensePose, CSE and HumanGPS on the test set of the aforementioned synthetic dataset. In Tables 1 and 2 we do the same on DensePose-COCO dataset but also provide additional comparisons with prior work. Opposite to the synthetic dataset, for which we have ground truth correspondences for every foreground pixel, for DensePose-COCO we rely only on the available annotated points to calculate the metrics. BodyMap shows a substantial improvement over prior work across all the metrics for both our synthetic and DensePose-COCO datasets. The reasons behind this improvement stem from: i) specifically designed architecture that separates and takes the best out of RGB and CSE inputs; ii) training on well-designed and rendered synthetic data and later fine-tuning on specifically adapted DensePose-COCO with the additional tricks discussed in Sec. 3.6, which helps to bridge the synth2real domain gap; iii) the proposed training scheme that includes supervision both in the image space with per-pixel classification losses as well as the 3D space with geodesic losses. Temporal Consistency. In Table 4 we test how temporally consistent the dense correspondences of different methods are. We were motivated to run this experiment by observing how jittery DensePose predictions can be on videos. In terms of metrics, we estimate the percentage of positive correspondence matches between the current frame and a frame in the future with an interval in 1, 12, 120 on a synthetic sequence consisting of 18, 000 frames. BodyMap out- Ablation studies Different Architectures. We experiment with different backbones starting from a simple UNet with skip-connections and then progressing to more complex transformer-based solutions. In Table 5 we provide a comparison in terms of accuracy in the 2D space across all the architectures. An interesting finding is that a simple UNet architecture can get satisfactory results when trained with all the proposed supervisions described in Sec. 3.3 and 3.4. However, our proposed Vision Transformer (ViT) is capable of learning more accurate correspondences in challenging areas like neck, armpits, fingers and hair, making the predicted silhouette clear-cut and crisp. These differences are mostly visible in hard DensePose-COCO examples (with multiple people and occlusions), while on simple synthetic data cases UNet is performing nearly as good as ViT. We further experiment with the network design, feeding only RGB inputs to the net and omitting the Correspondence Transformer. While RGB-only method performs comparatively worse, it still outperforms existing approaches, e.g.DensePose, CSE or HumanGPS (Tables 1, 2). Different Losses. We also investigate the impact of the proposed losses in Table 6. While the best score is achieved with the whole set of proposed losses, per-pixel cross-entropy classification losses for color channels contribute the most. Silhouette loss makes the edges of final prediction more accurate and extra supervision in hands and head regions improves correspondences in these areas. Geodesic losses give tangible improvement only on DensePose-COCO, indicating that simple synthetic one- Qualitative Results In Figures 1, 3 we show correspondences for a few images from DeepFashion [32] which has lower quality inputs, RenderPeople, DensePose-COCO and finally images from real-people scans captured with a 3dMD system. The silhouettes of the inputs are well covered with our estimates, the hands and fingers are accurately captured and the face is well aligned. Loose clothes, even complicated cases like a long robe in the are well-handled. In Figure 3 we show qualitative comparisons between BodyMap and competitors: HumanGPS, DensePose and CSE on several examples from DensePose-COCO, Render-People and our synthetic dataset. While DensePose and CSE predictions are smooth and consistent, they do not cover the whole silhouette, totally omitting hair and loose clothes. HumanGPS handles silhouettes better, but still struggles with accurate correspondences in challenging scenarios with occlusions or produces blurry patches for back views (Line 4 in Figure 3). We also show in the supplementary, that HumanGPS predictions are not always temporally consistent, jumbling correspondences for right and left arms and legs while the person is rotating. Applications & Discussions Neural Re-rendering. One possible application is rerendering people from the source frame from another view point and/or in another pose. We introduce a model for neural re-rendering which aims at learning a function that given the complete texture map and the estimated BodyMap correspondences generates a photorealistic render in the image space. Before neural re-rendering it is needed to obtain a complete texture map, which we do in the following way. Given a source and a target view of a person we utilize the predicted BodyMap estimates and defined a warping function W that outputs high-quality neural re-renders at the target viewpoint. We represent W with a neural network that i) warps the input source RGB image to the UV space to obtain a partial texture, ii) learns to complete it to obtain a full texture estimate, and iii) warps it back to the image space using BodyMap and then uses a neural renderer that generates the final output render. Given a source and a target image our neural renderer generates overall higher-fidelity details than prior work as seen in Fig. 4(left), also in the face and hand regions, and does not suffer from color bleeding. In the supplementary material we describe in detail this application along with an architecture figure and also present an application to cloth swapping and motion retargeting. Layered Dense Correspondences. In all prior work dense human correspondences are estimated only for the body surface. That is because a body template (e.g., SMPL [33]) with UV information is available and sparse annotations for COCO exist to accomplish this task. However, when dealing with clothed humans (and especially in loose garments) estimating body correspondences in a single-layer as DensePose or our proposed BodyMap does, can be a challenging task. However, fine-grained clothes details like wrinkles and textile folds can be represented better with decoupling body and clothes correspondences to separate representations. In a first attempt to do so we present an application with a slight BodyMap variation predicting three separate representations for the unclothed body, upper clothes and lower clothes. We named this variation Layered-BodyMap. The architecture remains the same besides the three output heads instead of one. To generate ground truth data for such a task, we run cloth simulation for the two garments given various walking and hand-movement motions resulting in 12 sequences of people wearing 3D clothes from our collection. Opposite to BodyMap, where we use RGB together with CSE initialization as input, here we do not have any initialization for the clothes correspondences, and as a result we feed this network with RGB-only inputs, but condition the estimates on semantic segmentation masks. The predicted layered correspondences are accurate and cover the whole silhouette ( Fig. 4 (right)) which is a promising result that we believe future work will improve upon as more 3D garment libraries become available [3,50,57,67]. Limitations. Our approach relies on foreground human segmentation which makes it susceptible to the performance of that step. We tested different segmentation and matting approaches, [10,22,23,54], and opted for MMSegmentation due to its ability to preserve fine details like fingers and hairstyles. BodyMap was trained on high-resolution mostly full-body images which makes it susceptible to low-res inputs or when only the bottom of the body is visible. This is partly solved by imposing heavy augmentations but occlusions from objects remain a challenge. Moreover, due to the nature of the task most of the training data is synthetic which makes inference on real data challenging. We address that with the fine-tuning scheme described in Sec. 4.2. We show some failure cases in Figure 5, which mostly happen due to bad lightning or severe occlusions and provide additional examples in the supplementary. Conclusion We present a novel framework for establishing accurate dense correspondences between an image and the surface of a 3D clothed human. Our key contribution, BodyMap, is a transformer-based architecture that when trained with 2D and 3D supervisions significantly outperforms prior work. BodyMap addresses key limitations of current approaches, such as inability to handle loose clothes, body and garments being represented as a single surface, non-continuity of the correspondences for different body parts. We outperformed prior work by a large margin on synthetic as well as DensePose-COCO datasets and investigated the impact of each of our design selections. Finally, we provided examples of applications such as re-rendering in different poses and extend BodyMap to clothed humans with multiple layers of geometry with promising results. Supplementary Material In this supplementary material we provide information regarding the broader impact of our method (Sec. 6) additional details regarding fine-tuning our proposed BodyMap on real data (Sec. 7) and an in-depth discussion on applications to neural re-rendering and cloth swapping with several qualitative examples (Sec. 8). Additional qualitative evaluations and results are shown in the supplemental video. Figure 6. Sparse and dense ground truth for DensePose-COCO dataset. Sparse ground truth: annotation available in the dataset. Dense ground truth: pseudo ground-truth estimates are generated on the fly by extrapolating both the available ground-truth annotations but also the CSE initialization such that they cover the whole estimated silhouette of the human Broader Impact The positive impact of our technology is further discussed in the applications section, including novel view synthesis and appearance swapping. While every humanrelated technology may raise concerns of fraudulent activities, we should note that our approach does not reconstruct facial or any other features used for personality identification. Thus, it is not possible to identify a person using our method, which makes our technology safe. Fine-tuning BodyMap on real data While BodyMap is trained with mostly synthetic data, we further fine-tune it with DensePose-COCO dataset. Available DensePose-COCO annotations are extremely sparse (around 100 annotated pixels per person, also annotations are not present for small people in the background). Thus, we leverage a heuristically made dense pseudo ground truth generation scheme. Given an image from the dataset, we generate pseudo ground-truth estimates on the fly by extrapolating both the available ground-truth annotations but also the CSE initialization such that they cover the whole estimated silhouette of the human (Figure 6). In that way we can fine-tune our model on real-data with denser supervision and utilize losses in both 2D and 3D spaces. Table 7. Ablation study on the effectiveness of finetuning: metrics in 2D image space. We illustrate the accuracy in 2D space on DensePose-COCO (the percentage of pixels correctly matched within the established error window) before fine-tuning (no ft), and after fine-tuning with: (i) sparse annotations with 100 labeled points per person (ft 1), and (ii) dense annotations, obtained heuristically by interpolating annotations within silhouettes ft 2). We include the performance of DensePose and CSE for comparison. Table 8. Ablative studies on the effectiveness of finetuning: metrics in 3D space. We illustrate Average Precision (AP) and Average Recall (AR) over GPS scores on DensePose-COCO dataset before fine-tuning (no ft), and after fine-tuning with: (i) sparse annotations with 100 labeled points per person (ft 1), and (ii) dense annotations, received heuristically by interpolating annotations within silhouettes ft 2). We also include the performance of DensePose and CSE for comparison. We experimented with two ways of fine-tuning on real data: (1) using only available sparse annotations (sparse fine-tuning); (2) using the generated dense pseudo groundtruth estimates (dense fine-tuning). The results of the both schemes are presented in Tables 7 and 8, showing that dense fine-tuning allows to get the best results beating competitive methods such as HumanGPS, DensePose and CSE. Rendering people in arbitrary poses/views One of the straightforward applications of our proposed BodyMap framework is re-rendering people in novel poses/views. Previous attempts to do so include Texformer [59] -Transformer-based framework for 3D human texture estimation from a single image. They utilize pre-computed color encoding of the UV space obtained by mapping the 3D coordinates of a standard human body mesh to the UV space as a Query, feeding it to the Transformer. We also rely on specifically designed Transformerbased architecture to retrieve dense correspondences that can lately be used as a mapping function between 2D image space and 3D space. While we can do novel view ren- Figure 7. Novel view rendering with BodyMap. Given an RGB image we first obtain its CSE [36] estimates and feed both to BodyMap network to get refined per-pixel predictions covering the whole silhouette. With such correspondences we can obtain partial texture maps that are very well aligned with the complete texture map which are then utilized to train a texture completion network to fill in the missing information. At last, we use the BodyMap as a mapping function back to the image space where we train a neural re-rendering framework that generates the photorealistic renders. dering without any extra efforts simply warping the texture using BodyMap estimates as a mapping function, using additional neural renderer significantly increases the quality of the final render. Next we describe in the detail the proposed pipeline for generating novel views consisting of two main steps: texture completion network and neural rendering. The pipeline is also depicted in Figure 7. Texture Completion Network Given the BodyMap estimates and the foreground RGB image we define a warping function that maps each foreground pixel of the image space to the UV-map space using BodyMap estimates as the mapping function. Note that in the UV map space, every point on the mesh surface of a human body template is represented by its coordinates on this UV map. We then train a texture completion neural network that takes as input partial textures at 1024 × 1024 resolution and completes the missing information by producing the full texture. The fact that partial texture is so well aligned on the UV map with full texture enables us to utilize a U-Netlike architecture since the skip connections can transfer the aligned input from the encoder to the decoder layers without adding an additional overheard to the decoder. A challenge that arises when dealing with high-resolution inputs is that only a few samples can fit into the GPU memory during training. Instance normalization blocks have widely been used in such cases to avoid collecting batch statistics that can be inaccurate due to the small sample size [1,27], but our experimental investigation indicated that instance normalization produces completed textures with distorted colors in non-visible regions. To overcome this challenge we propose to utilize synchronized batch normalization which differs from previous methods in the way the statistics are computed over all training samples distributed on multiple devices. This enables us to learn more accurate batch statistics that can then be used at test-time with traditional batch normalization blocks. Thus, given pairs of partial and full textures ({P T , F T } ∼ T ) coming from the data distribution we train the texture completion network with the following losses: • Hinge version [28,64] of the adversarial loss, along with a multi-scale discriminator as used in Pix2PixHD [20]: • Perceptual loss. We utilize a pre-trained VGG [55] network and compute the L 1 loss between the completed texture estimate and the ground-truth texture map at the activations of five different layers of the network. Perceptual similarity losses help the network to generate fine-level details which are common in the textures of clothed humans. • Total variation loss. We add a total variation (TV) loss [34] with a small weight in order to encourage spatial smoothness in the generated textures and remove some artifacts that are quite common in image- Figure 8. Applications on virtual dressing and motion imitation: given a single RGB image and a variety of target poses our method can dress the target subjects with the input hoodie at different poses with the hands being potentially occluded while preserving fine-level details. Our approach also generates photorealistic renders of the input human given target images of other people with different poses and viewpoints. Our method has not been trained specifically for any of these tasks but it can still generalize and produce crisp results. to-image translation networks. The TV loss is formulated as follows: Neural Re-rendering We present an additional step to obtain photo-realistic human renders after texture completion. While we can assume access to 3D geometry for our synthetic data and perform rendering with tools such as Blender Cycles [9] or PyTorch3D [41,46], this is not the case for real-world examples. One approach to tackle this problem and obtain a 3D geometry would be to estimate the 3D human body pose [51] and shape from a single image by using any of the recent state-of-the-art methods [21,[24][25][26]44,60,65,66]. However, all these methods estimate the body under the clothing which is usually relatively slimmer and with pose inaccuracies (e.g., the body bends forward) due to the depth ambiguity which makes it unsuitable for rendering the texture of a clothed human on top. Thus, we propose a model for neural re-rendering which aims at learning a function X = R(BodyMap, F T ) that given the complete texture map F T and the estimated BodyMap generates a photorealistic render X in the image space. The advantage of our approach compared to prior work [53] is that BodyMap not only provides a proper silhouette in the image space that needs to be rerendered but also serves as a mapping function between the UV and the image space. This enables us to warp the texture map back to the image space using the BodyMap to obtain an initial estimate which is then fed to an encoder-decoder network that generates the final output render. The fact that BodyMap provides accurate per-pixel foreground estimates makes the warped image well aligned with a target output which simplifies the learning process of the neural renderer. Finally, since during the warping process some texture information can be lost due to warping inaccuracies [31], we pass F T through an encoder network to generate a lower dimensional tensor representation which is then fed via the bottleneck to the decoder of the neural render. We train this network with the same losses described that we used for texture completion network and in addition, we employ the following two losses: • Feature Matching Loss. We use a feature matching loss in the discriminator layers [20] to obtain highfrequency details such as wrinkles and cloth patterns which is defined as: L FM = ∑ 3 l=1 \|\|D l (x)−D l (G(x))\|\| 1 . • Reconstruction Loss in the face region. Using the segmentation mask of the face region which are then used to employ additional reconstruction losses in that area in order to force the network to estimate more photo-realistic faces and fix artifacts around the eyes and the mouth. Results Expanding Figure 4 from the main paper, we present several more results of re-rendered people in 9. It shows that even before re-rendering (column 6) novel views demonstrate a decent level of details including textile patterns, hairstyles and fingers. Neural re-rendering helps to get rid of occasional artifacts, smooth out the final result and indicate even more fine-grained details. Appearance swapping Additionally to generating people in novel views and/or poses, our approach allows to redress people providing renders of them in different clothes. Using BodyMap estimates as a mapping function, we show several examples of appearance swapping in Figure 8.	2022-05-19T06:47:33.136Z	2022-05-18T00:00:00.000	{ "year": 2022, "sha1": "a11678e0cc6a44317319b1318c6af2b35fad7a7b", "oa_license": null, "oa_url": null, "oa_status": null, "pdf_src": "Arxiv", "pdf_hash": "a11678e0cc6a44317319b1318c6af2b35fad7a7b", "s2fieldsofstudy": [ "Computer Science" ], "extfieldsofstudy": [ "Computer Science" ] }
32968803	pes2o/s2orc	v3-fos-license	The impact of innovation and the use of ICTs on human capital development in Spanish industry Purpose: The analysis of the characteristics of the industrial companies that invest in the training of their human capital in comparison to the organizations that do not train their employees. This characterization is focused especially on the innovative capacity of firms and their technological applications. It also analyzes the factors that determine the likelihood that a company invests in personnel training. Design/methodology/approach: Descriptive characterization of those companies that invest in personnel training, comparing them against those who do not invest in improving its human capital. Furthermore, factors influencing the decision that a company invest in training are analyzed through a logistic regression. Among them we included the intensity of technological usage and the innovation capacity. This research is based on data from the Survey of Business Strategies (ESEE) in 2009, conducted annually. The sample is approximately 1,800 companies, being representative of the Spanish manufacturing sector. Findings: Companies with a higher likelihood of investing in training of their workers are those with a larger number of workers, a higher level of qualification of employees, more stable labor relations, greater participation foreign capital, and also, have a higher level of technological uses, carries more technological partnerships with other organizations and has a more intense innovative activity. Introduction In the last few decades, human capital development has become a key factor in the competitiveness of companies (Monreal, 2004;Jeon & Kim, 2012) for two main reasons. Firstly, the large-scale incorporation of information and communication technologies (ICTs) into business activity means that new skills are required for work, as well as new ways of working and establishing relationships (Bresnahan, Brynjolfsson & Hitt, 2002).Secondly, an increase in the economic use of knowledge (Haelermans & Borghans, 2012) has led to a major change in the conditions and characteristics of the labour market (Carnoy, 2001). These two processes have brought about a general increase in the need for continuing vocational training (CVT).All training activities carried out by companies, employees or the corresponding representative organizations must be aimed at improving the professional skills and qualifications of working professionals, so that they can meet the needs of a changing, digitalized, global labour market (Castells, 2003).In fact, most authors (including Johnson, 1982;Piore & Sabel, 1984;Pfeffer, 1998) consider that the most efficient way of increasing company flexibility is through employee-centred organizational innovations that involve intensive training, boost productivity, and give workers a feeling of security and of belonging to a team (Carnoy, 2001).In short, increases in productivity depend not only on investment in technology, but also on organizational changes and skilled human resources, which are -250-Intangible Capital -http://dx.doi.org/10.3926/ic.423required to extract and benefit fully from the technological potential (Brynjolfsson & Hitt, 2003).Therefore, learning is important mainly because of the need for organizations to respond to constant, rapid changes in the surrounding environment (Coetzer & Perry, 2008).To train employees, a learning environment is required that enables them to advance professionally and to gain new general and specific skills and abilities (Fan & Wei, 2010).To achieve this, employees may train in an environment either outside or within an organization.The learning environment that is most closely associated with real work experience is training in the workplace (Fan & Wei, 2010). The report entitled Learning for Jobs, published by the OECD (2010), presents the four main advantages of workplace or internal training.Specifically, continuing education in the workplace provides a high quality learning environment that enables students-employees to gain practical, up-to-date knowledge under trainers or tutors who are familiar with the working method and the use of technologies.Secondly, workplace training facilitates a two-way flow of information between potential employers and employees, which makes the subsequent process of recruitment easier, less costly and more effective.Thirdly, employees who are trained in the workplace can make a productive contribution to the company.Finally, the fact that companies train their employees for their professional tasks shows the value of vocational training programmes in the labour market.This last point is particularly relevant in Spain, given the specific distribution of educational levels in the population.Specifically, Figure 1 In turn, the massive injection of knowledge into the economy has an impact on management, production and working methods.It makes employment relationships more flexible, either voluntarily or involuntarily, which forces companies to accept that process and product innovation is of great value to them, as is a continuous process of retraining.This constant involvement with knowledge, updating it, and continuous learning again favours the use of workplace training as an essential method to ensure a perfect symbiosis of work and training (Batalla-Busquets, Martínez-Argüelles & Vilaseca, 2010). In the current climate of job destruction and a lack of competitiveness of Spanish industry in increasingly globalized markets, this study aims to analyse and explore the potential of training as a tool to boost business success.In fact, it is increasingly clear that the main way to increase work productivity in Spain is to improve employees' skills and abilities, in other words, increase investment in the continuous development of human capital. Below, we describe the theoretical framework of the research, and present the sources of information used.After that, we describe the main results obtained in the study and, finally, we discuss the main conclusions and limitations. Theoretical framework The concept of human capital (Schultz, 1961) and its subsequent consolidation in the Theory of Human Capital (Becker, 1964) form a corpus of theory that explains the positive impact of investing in employee training.Basically, better qualified employees will have gained new skills and abilities that enable them to be more productive in their work.The increase in productivity is reflected in higher salaries (de Grip & Sauermann, 2013).At the same time as this formulation, the first empirical studies emerged on the factors that determine the existence of training processes in companies.The seminal work of Mincer and Polachek (1974) analysed the decision-making process in families from a gender perspective, and in terms of human capital investment.After this, the first scientific studies were published on the existence of a pattern explaining how investment in staff training is distributed (Duncan & Hoffman, 1979;Greenhalgh & Stewart, 1987). The results of the numerous empirical studies that use sources of business information are focused mainly on describing the attributes of companies that invest in developing the human capital of their professional team.At international level, some studies show that there is a positive relationship between the probability of receiving training and the size of the company, expressed in terms of the number of workers (Hashimoto, 1979;Oi, 1983;Holtmann & Idson, 1991;Barron, Black & Loewenstein, 1987;Lynch, 1994).Others highlight the academic experience of the employees themselves (for example, Altonji & Spletzer, 1991;Lynch & Black, 1995;Bishop, 1996;Harris, 1999), the type of contract (Oosterbeek, 1996), the salary level (Barron et al., 1987;Lynch, 1992;Bishop, 1994), or the unionization of employees (Arulampalam & Booth, 1998;Jonker & de Grip, 1999), among other corporate factors.In Spain, despite the lack of availability of data, which has limited the number of studies in this area (Peraita, 2000), we can find many studies with this focus, such as those of Abellán, Felgueroso and Lorences (1997), Crespo and Sanz (2000), Diéguez and Sinde (2004), Albert, García-Serrano and Hernanz (2005), Caparrós, Navarro and Rueda (2005), Vila and García-Mora (2005), Escardíbul, Oroval and Afcha (2007), García-Moreno, Guerras andRico (2007), Pineda Herrero (2007) and Castany (2008). The increasing interest in training is shown by the increase in public and private investment in improving the qualification of human capital (Peraita, 2000;García-Delgado, 2003;Pineda Herrero, 2007).In recent years, there has been a significant, sustained rise in the number of companies that provide training for their employees, and in the specific proportion of the cost of continuing training in total labour costs.This increase in investment reflects the importance given to the training and retraining of people as a key factor to increase work productivity (Monreal, 2004).In addition, many employees take training courses and cover the costs themselves. However, a high percentage of companies still do not consider employee training to be a key factor in maintaining and increasing competitiveness.This is particularly true in small and, to some extent, medium-sized companies, due to a lack of knowledge about public funds for promoting corporate training (Pineda Herrero, 2007). In terms of public investment, the funds allocated to continuing training have been increased and the funding structure has changed.For example, corporate training was boosted by the coming into force of Royal Decree 395/2007, in which the responsibility for employee training was attributed to the companies themselves.These public funds are managed by the Fundación Tripartita.This kind of training is a more flexible, effective way to respond to the diversity and frequent changes in employee training needs.Thus, in the 2006-2012 period, the percentage of companies who benefited from these funds rose from 5.8% to 31.1% (Fundación Tripartita, 2012). Nevertheless, a substantial proportion of the funds that the administration makes available to companies to fund employee training does not end up being used.In 2012, Spanish companies used only just over 30% of the total funds available for this kind of training (Fundación Tripartita, 2012).Therefore, the focus of interest of this study is to analyse the characteristics of companies that train employees and the determining factors in corporate training, to explain the gap between organizations that invest in training and those that do not. Data and methods To carry out this research, we used a cross-sectional sample for 2009 of 2,132 Spanish manufacturing companies from the Business Strategy Survey (ESEE) carried out by the Fundación Empresa Pública (FUNEP).This survey gathers unbalanced panel data for the 1990-2010 period on companies' strategic decisions and behaviour.Every four years, companies answer a full questionnaire (whose questions are supposed to remain the same each year).In the intermediate years, the companies answer a short questionnaire.As a result, complete information is available for 1990, 1994, 1998, 2002 and 2006.The ESEE survey population is companies with 10 or more employees dedicated to one of the activities corresponding to Divisions 10 to 33 of the CNAE-2009 classification, excluding Division 19 (activities related to oil refining and fuel handing, except nuclear fuel).The survey contains information about the general characteristics of companies, their employees, the organization, the business strategy, the equipment and uses of ICTs and innovation, among other aspects. Descriptive analysis To understand the situation under study, we first needed to characterize the Spanish industrial sector.The description was based on the results obtained by applying an analysis of variance (ANOVA), in which aspects related to employee training were central to the study.The characterization is divided into two blocks.The first includes variables that are validated in the scientific literature as determining factors in corporate training, such as company size (Holtmann & Idson, 1991;Black, Noel & Wang, 1999;Crespo & Sanz, 2000;Caparrós et al., 2005); the type of employment relationships (García-Espejo, 1999;Planas & Passard, 2000;Tugores & Alba-Ramírez, 2002;Caparrós, Navarro & Rueda, 2004;Castany, 2008), the sector (Harris, 1999;García Moreno et al., 2007;Turcotte, Léonard & Montmarquette, 2003), the foreign capital investment (Alba-Ramírez, 1994), level of training of employees (Peraita, 2000;García Moreno et al., 2007), among other variables.The second block provides a more detailed analysis of research and development activities, innovation, and technology uses, to validate the association between the existence of innovative activity in a company and corporate investment in human capital development, as stated by Alba-Ramírez (1994) or García Moreno et al. (2007) for the Spanish economy.If we add together all companies that stated they train their employees, 90.1% of companies considered that they had undertaken some training for their employees in the last financial year, and 9.9% did not allocate any resources to the development of their human capital. In this study, we found statistically significant differences between the two main kinds of companies that were analysed, depending on whether or not training was carried out in them (see Table 2). Companies in the Spanish industrial sector that do not invest in employee training are mainly micro-enterprises: 71% have fewer than 20 employees, and 91.6% fewer than 50 employees. The intensity of knowledge use is mainly low; specifically 86% of these companies are in sectors with low knowledge intensity.Foreign capital investment is practically non-existent, and the percentage of export companies is 31.6%;half that of companies that train employees.The proportion of temporary and permanent staff is similar to the average for Spanish industry.Frankly, the level of training of employees could be improved, as only 6.69% have a university qualification at intermediate or higher level.The database does not provide more detailed information about the level of non-university training of employees who do not have a university qualification (93.31%). Finally, the productivity of employees, measured using gross value added (GVA), is 57% in companies that provide training.GVA is calculated using the average number of total staff (ATS), which is calculated as the sum of the following concepts: Full-time permanent staff, 1/2 of part-time permanent staff (both concepts on 31 December) and the average number of temporary staff. Industrial companies that train their employees are larger: 52% have over 50 employees. Almost all of the medium-sized and large companies (with over 200 employees) train for their employees.A quarter of them carry out industrial activity in sectors with high knowledge intensity.To determine this figure, a dichotomous variable was used to indicate whether or not a company offered a knowledge-intensive service.The classification of companies by knowledge intensity was initially carried out according to OECD proposals (1999). The percentage of companies with foreign capital investment was 16.5%, and two thirds of companies that train employees were exporters.The level of qualification of employees was much higher in companies that offer training than in companies that do not: almost 14% of employees had a university qualification at intermediate or higher level.In our sample, the average productivity of companies that offer training was 50,000 euros per employee, whilst the average productivity per employee in companies that do not offer training was 29,000 euros.This difference of just over 20,000 euros is statistically significant. Below, we analyse the differences in research and development activities, technology planning, innovative activity, intensity of Internet use, and specific research and development jobs among companies that invest in training and those that do not make this investment.In the comparative analysis of research and development activities (see Table 3), we found that companies that train employees undertake more activities associated with research and development (38.3%).These activities are practically non-existent in companies that do not train employees.In addition, although technology collaborations do not take place in anywhere near the majority of companies, organizations that develop their human capital establish more collaborations with universities and research centres, and with suppliers and clients.These kind of collaborations are extremely rate in companies that do not train employees.Technology planning in Spanish companies, and more specifically in the industrial sector, is an area that is still in its infancy (see Table 4).Approximately a quarter of companies that provide training have a technology committee or assess perspectives of change or alternative technologies.In contrast, in organizations that do not invest in continuing vocational training, this planning is negligible: only 2% to 8% of companies in this group carry out some technology planning activities. The decrease in productivity in the Spanish industrial sector since the start of the twenty-first century is related to a reduction in corporate investment in technology activities and a decrease in the ability of companies to turn their investments into product and process innovation in particular (Huergo & Moreno, 2004).In fact, Table 5 shows that product and The differences between the two kinds of companies analysed in this study are statistically significant.Consequently, we decided to transform all of the variables related to product and process innovation into a dichotomous variable (INNOVEMPR) that is zero if the company does not innovate and one if the company undertakes some kind of product and/or process innovation.Figure 2 shows the difference between the two kinds of company and illustrates the statistically significant differences in the dichotomous variable between them.Individual hypothesis testing confirms the existence of significant differences.The Internet is widely used by companies in the Spanish industrial sector.However, the degree of penetration in the company's value chain is mostly low or non-existent.Almost 80% of companies have a website, but only 10% use this digital infrastructure as a sales platform.If we analyse the differences between companies that train employees and those that do not invest in training (see Table 6), we can see that most organizations that train employees have a website (81%), compared to 61% of companies that do not offer training.In addition, more sophisticated uses of the Internet are found in higher proportions in companies that train their employees. Hypothesis testing With An indicator of the intensity of Intensity use (USOSTECH) has been drawn up in which four levels of use are defined: • Non-existent: the company does not even have its own domain. • Low: the company has a website, but does not carry out any economic activity on the Internet. • Medium: the company has a website and purchases from suppliers on the Internet. • High: the company has an Internet domain and purchases and sells to other companies or end consumers on the Internet.To complete this descriptive analysis of research, development and innovation activity, we will briefly analyse the specific human capital dedicated to this kind of activity in organizations. Table 7 shows that research and development jobs in companies that do not offer training are practically non-existent.In contrast, companies that provide employee training do have this kind of staff.In some cases (5.4%) they recruit staff with experience in this area. Determining factors in corporate training In order to analyse the factors that determine the probability of a company investing in employee training, below we describe a discrete choice model with a dichotomous dependent variable.Specifically, we used a logistic regression model or logit.This model is a statistical technique that is suitable when the endogenous variable is categorical, and the exogenous variables are metric (Hair, Anderson, Tatham & Black, 2004), as in our case.It is a technique that has been widely used in similar studies, such as Greenhalgh and Stewart (1987), Barron et al. (1987), Abellán et al. (1997), García-Espejo (1999) and Batalla-Busquets et al. (2010). The logit model is expressed as follows: After the descriptive analysis of the variables, we examine the results of the econometric analysis that illustrates the effects of these variables on the probability that employees are trained.The logistic regression model described below includes factors that significantly influence the probability that a company assumes the costs of employee training, compared to companies that do not train their employees.The model is specified as follows: Where: Li As can be seen, this model is comprised of six statistically significant exogenous variables (total staff, productivity per employee, proportion of engineers and degree holders, level of Internet use, business innovation, and research and development activities).The degree of correlation between the variables, which was no higher than 40% in any of the cases, is shown in Table 8.The goodness of fit of the model is acceptable, as the overall predictive capability of the model is 90%, the Cox-Snell R 2 is 57.7% and the Nagelkerke R2 is 76.9% (see Table 9).All the variables have a positive effect on the probability of companies paying for training.In particular, there was a higher odds-ratio for the three variables related to the use of technologies and corporate innovation.In other words, companies that carry out research and development activities and/or use the Internet in a more sophisticated way and/or innovate in the product or process have a greater likelihood of investing in employee training.In fact, the use of increasingly specialized technologies requires very specific knowledge and skills that are not easy to find in the labour market.Consequently, training should help employees to gain these skills (Castany, 2008). Likewise, the size of the company has a positive influence on the probability of training, as described in the ample scientific literature.Among other reasons, this is because • larger companies have a greater capacity to substitute employees who are training, • they obtain economies of scale with the training, and • they have more opportunities to offer a more attractive career associated with the employee's qualification (Holtmann & Idson, 1991;Black et al., 1999;Crespo & Sanz, 2000;Caparrós et al., 2005;Castany, 2008). The gross value added per employee is positively correlated with the probability of training. This relation is due to the fact that companies that bring greater added value to the production process require staff with a higher level of training (Batalla-Busquets et al., 2010).Finally, the proportion of engineers and degree holders is also positively related to continuing vocational training.Good prior training of the employee is a guarantee of good returns on investment in training (Harris, 1999).In addition, better trained employees are more likely to take on greater responsibilities in the company, which are often closely linked to a training process (Peraita, 2000;García Moreno et al., 2007). Conclusions In a context such as the current one, in which job are constantly being lost and companies are folding due to a loss of international competitiveness among Spanish companies and the collapse of much of the domestic market, human capital development is seen as one of the few options that could change the course of the Spanish economy. The results obtained in this study show that the probability of companies investing in employee training depends on the total number of company employees, the productivity per employee, the level of qualification of employees, the level of Internet use, business innovation, and research and development activities.This study indicates that the Spanish industrial sector does not have ideal characteristics for investment in employee training, according to international literature, as many companies are small, their employees have a low level of qualifications, and there is little innovation and research activity. Consequently, the gap has not been closed between large competitive companies that use digital resource intensively, are innovative, and have highly qualified human capital, and the remaining small and medium-sized companies (Batalla-Busquets et al., 2010).This confirms a statement made by Crespo and Sanz (2000): the mass introduction of digital technologies into companies added to the lower probability of training for employees with a lower level of studies could lead to the practical exclusion from continuing vocational training of the least qualified employees, with the resulting loss of incentives for a company to train them. The Spanish industrial sector is reasonably well-equipped digitally.However, although most companies have a website, the use of this platform is negligible.When the business organization has not been globalized or digitalized, a good Internet infrastructure becomes a cost rather than a business opportunity.Therefore, an investment in employee training is required to maximize the Internet opportunities.In addition, as indicated by Huergo and Moreno (2004), the Spanish industry encounters difficulties in turning investment in innovation into tangible results that can be used to improve process and products, and consequently increase the competitiveness of the industrial sector. The main limitation of this study is sample bias, as most of the sample is comprised of large companies and there is only a small percentage of small and medium-sized companies, when these represent almost 85% of the Spanish industrial sector.A second limitation of the study is in the concept of company training that does not involve an associated cost.In this case, it is difficult to discern the quality, intensity and returns on this training. Figure 1.Level of education in the adult population (25-64 years).Spain and the EU-21.2011 Figure 2 . Figure 2. Innovation in the product and/or process.Training vs. no training Figure 3 . Figure 3. Intensity of Internet use.Training vs. no training In this model, Pi takes the value of one if the company invests in training, and zero if the company does not invest in training.Therefore, if we apply these values to the logit model, the result is:Li = 1n(1) = 1n(+∞) = +∞ If the company invests in training Li = 1n(0) = 1n(-∞) = -∞ If the company does not invest in trainingBelow, is a descriptive analysis to characterize the Spanish industrial sector according to investment/no investment in improving the training of employees. is the logistic function associated with the probability of a company paying for employee training (Pi = 1) or the lack of company training (Pi = 0); PERTOTi is the total number of staff in the company; PRTPi is the productivity per employee in GVA; PIL0609i is the proportion of engineers and graduates in the period 2006-2009; USOSTECHi is an indicator of the use of the Internet; INNOVEMPRi is a binary variable that indicates whether a company has innovated in the product and/or process, and AIDi is a dichotomous variable that indicates whether the company has carried out research and development activities. Table 1 presents the frequencies in the survey sample, which was divided into four business categories: no training, training by the company, some training outsourced, all training outsourced. Table 1 . Frequencies in the sample of companies(ESEE data) Table 3 . Intangible Capital -http://dx.doi.org/10.3926/ic.423process innovations only occur in a small percentage of companies that train employees, and are non-existent in companies that do not invest in training.Just as worrying as these data is the lack of planning and assessment of innovation results.Research and development activities and technology collaboration Companies that train employees compared to those that do not invest in training Table 5 . Innovative activity.Companies that train employees compared to those that do not invest in training Table 6 . Internet use.Companies that train employees compared to those that do not invest in training Table 7 . Specific staff for research and development activities.Companies that train employees compared to those that do not invest in training Table 9 . Model of the probability that companies pay for training.Logit model; dependent variable: training paid for by the company or no training	2017-09-09T15:28:00.974Z	2015-06-15T00:00:00.000	{ "year": 2015, "sha1": "b9cf7d60b50b4297296fa80133035e4a70b67eb6", "oa_license": "CCBY", "oa_url": "https://www.intangiblecapital.org/index.php/ic/article/download/423/468", "oa_status": "GOLD", "pdf_src": "ScienceParseMerged", "pdf_hash": "b9cf7d60b50b4297296fa80133035e4a70b67eb6", "s2fieldsofstudy": [ "Business", "Computer Science", "Economics" ], "extfieldsofstudy": [ "Economics" ] }
49686335	pes2o/s2orc	v3-fos-license	Benign Acute Childhood Myositis: Perplexing Complication after Acute Viral Pharyngitis Benign acute childhood myositis (BACM) is a rare transient condition that occurs in children during the early convalescent phase of a viral upper respiratory infection. BACM is self-limiting and characterized by sudden-onset bilateral calf pain that leads to difficulty in walking. We report a case of a 5-year-old boy with BACM who presented with acute-onset bilateral calf pain after a resolved episode of viral pharyngitis and subsequently refused to walk. With conservative treatment, the patient recovered completely after approximately 1 week. Although perplexing and challenging for clinicians unfamiliar with BACM, awareness of this rare clinical condition is essential to preventing unnecessary investigations and reassuring the patient and parents of its excellent prognosis. INTRODUCTION Benign acute childhood myositis (BACM) is a rare self-limiting disease that primarily affects young school-aged boys and characteristically follows a viral upper respiratory infection (URI). It is also characterized by sudden-onset calf pain leading to a refusal to walk due to motor difficulty. 1) This refusal to walk can be perplexing for both clinicians and parents and commonly leads to an extensive work-up. Since many clinicians are unfamiliar with BACM, it is often misdiagnosed and interpreted as a more severe and complex disease. 2,3) The clinical manifestation is usually preceded by an initial phase of a viral URI, most commonly associated with influenza viruses. 3) The illness typically lasts for a brief period of time, usually up to 1 week. 1) Clinicians often treat many cases of patients with viral URI; however, BACM, a rare complication of viral URI, is still not well recognized. In this report, we present a case of BACM occurring in a 5-year-old boy in the early convalescence phase of a viral URI. We also review the literature regarding BACM and emphasize why knowledge of BACM is essential to preventing unnecessary invasive diagnostic procedures and therapeutic interventions, allowing clinicians to reassure patients and their parents about its excellent prognosis. However, the C-reactive protein level was elevated, confirming the presence of an infection (2.7 mg/dL; normal range, 0 to 0.5 mg/dL). A urinalysis showed normal results; no blood or proteins were detected. A nasopharyngeal aspirate (NPA) was taken to test for influenza A and B; human parainfluenza virus types 1, 2, and 3; respiratory syncytial virus; adenovirus; and human metapneumovirus (routine panel) by polymerase chain reaction. The NPA results were positive for influenza A but no other viruses. Patients with flu do not ordinarily require hospital admission. However, the parent requested admission due to the fever and anorexia. On admission, the patient received symptomatic treatment with simple analgesics and intravenous fluids to prevent dehydration. Antiviral therapy was not considered because the patient had no risk factors such as asthma, cardiac disorders, or other conditions. During the 2-day hospitalization, the patient's symptoms resolved completely including the injected pharynx and his condition rapidly improved, allowing us to consider hospital discharge. On the morning of the third day after admission, the patient was unable to get out of bed and unexpectedly refused to walk. The caregiver attempted to make the patient stand, but he was unable to bear weight on his legs and complained of bilateral calf pain. On examination, the patient was afebrile with normal vital signs and normal consciousness. Furthermore, there was no evidence of trauma to the lower limbs. A neurological examination revealed normal muscle power and tone; the tendon reflexes and sensation in his lower limbs were also normal. However, the patient could not walk. A detailed examination failed to reveal any abnormalities other than calf tenderness, and no erythema or edema was noted. At rest, the patient kept his feet in slight plantar flexion since passive dorsiflexion at the ankles caused sharp pain. A laboratory investigation revealed a significantly increased creatine kinase (CK) level (2,690 U/L; normal range, 5 to 217 U/L) and normal leukocyte count (7.84×10 9 /L; normal range, 6 to 15×10 9 /L) and C-reactive protein level (0.3 mg/dL; normal range, 0 to 0.5 mg/dL). Complete blood count, electrolytes, blood urea nitrogen, creatinine, prothrombin time, partial thromboplastin time, and other routine blood parameters were also within normal limits. Urinalysis was normal with no detected myoglobin. After consulting the pediatric neurology department, we extended his admission for observation and conservative management. The patient was clinically diagnosed with BACM. Two days after its onset, the calf pain spontaneously resolved, and the patient was able to walk and run without support. CK levels gradually decreased, and a repeated urinalysis was negative for myoglobin. The patient was discharged 5 days after admission (2 days after BACM symptom onset). A repeat CK measurement 1 week later revealed normal levels, and the patient made a complete recovery with no residual impairments or neurological sequelae (Table 1). DISCUSSION The rare clinical entity BACM was first reported as "myalgia cruris epidemica" by Lundberg 1) in 1957. He described 74 school-aged children with severe calf myalgia following a URI. During an influenza epidemic, BACM was distinct from the diffuse myalgias that are common during the course of influenza infection. 1) In most cases, BACM is self-limiting, with the myositis developing once the patient's viral URI is already resolving. 1) Although it most commonly occurs after an influenza B infection, BACM has also been associated with influenza A, parainfluenza, Coxsackie virus, herpes simplex virus, Epstein-Barr virus, and adenovirus infections. 4) Data from animal studies suggest that BACM is possibly related to direct invasion of the muscle tissue by the infecting virus and its subsequent inability to replicate within the myocytes. 5) The initial infection results in muscle fiber necrosis, which is sufficient to elevate CK levels; although no viruses could be detected in biopsy Normal ranges are given in parentheses. specimens, this suggests that the virus is unable to replicate in this tissue. 5) However, in some cases, influenza virus antigens were detected in muscle biopsies from patients, suggesting that these viruses can directly infect the muscle. 6) Thus, uncertainty remains as to whether the myositis is caused by direct viral action or immune-mediated mechanisms. 7) Most patients who develop BACM are children, although the condition has been described in adolescents. 8) Boys are affected more frequently than girls. 7) The clinical picture is usually preceded by influenza, with symptoms of URI such as fever, malaise, cough, sore throat, and rhinorrhea. 2) Patients present with sudden-onset intense bilateral calf pain; characteristically, tenderness and difficulty walking or a refusal to walk after a period of rest, such as upon awakening, occur. Serum CK level is consistently elevated in almost all patients and returns to normal within 1 month of symptom resolution. 7) All clinical symptoms spontaneously and rapidly resolve within a week ( Table 2). The differential diagnosis in pediatric populations for failure to weight-bear includes Guillain-Barré syndrome, rhabdomyolysis, dermatomyositis, polymyositis, and muscular dystrophy ( CONFLICT OF INTEREST No potential conflict of interest relevant to this article was reported. Table 2. Key features associated with benign acute childhood myositis [1][2][3][4][5][6][7] Key features Sudden refusal to walk or gait-related abnormalities on awakening due to bilateral calf pain after resolving viral upper respiratory infection Boys between 3 and 14 years of age are most commonly affected Normal neurological exam (deep tendon reflex; motor and sensory examination) Elevated serum creatine kinase level and normal urinalysis findings Patient's feet in slight plantar flexion; passive dorsiflexion at the ankles elicits pain With conservative management, all clinical symptoms are spontaneously and rapidly recovered within 1 week	2018-07-16T23:43:50.036Z	2018-07-04T00:00:00.000	{ "year": 2018, "sha1": "f0585e33e6b5a4ab4710944fd4b880ad05c791e9", "oa_license": "CCBYNC", "oa_url": "https://www.kjfm.or.kr/upload/pdf/kjfm-17-0113.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "f0585e33e6b5a4ab4710944fd4b880ad05c791e9", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
16493406	pes2o/s2orc	v3-fos-license	Serum S100B Protein as an Outcome Prediction Tool in Emergency Department Patients with Traumatic Brain Injury SUMMARY Objectives Traumatic brain injury is a common cause of death and disability worldwide. Early recognition of patients with brain cellular damage allows for early rehabilitation and patient outcome improvement. Methods In this prospective study, the clinical conditions of patients with mild to moderate traumatic brain injury (TBI) were assessed, and patient serum S100B levels were measured. Patients were followed up one month later and evaluated for level of consciousness, presence or absence of post-traumatic headache, and daily activity performance (using the Barthel scale). Student's t-test and the chi-square test were used for data analysis, which was performed using SPSS software. Results The mean serum S100B value was significantly lower for patients with minor TBI than for patients with moderate TBI (23.1±14.2 ng/dl and 134.0±245.0 ng/dl, respectively). Patients with normal CT scans also had statistically significantly lower serum S100B levels than patients with abnormal CT findings. The mean S100B value was statistically significantly higher for patients with suspected diffused axonal injury (632.18±516.1 ng/dl) than for patients with other abnormal CT findings (p=0.000): 24.97±22.9 ng/dl in patients with normal CT results; 41.56±25.7 ng/dl in patients with skull bone fracture; 57.38 ±28.9 ng/dl in patients with intracranial hemorrhage; and 76.23±38.3 ng/dl in patients with fracture plus intracranial hemorrhage). Conclusions Serum S100B levels increase in patients with minor to moderate TBIs, especially in those with diffused axonal injury. However, serum S100B values cannot accurately predict one-month neuropsychological outcomes and performance. Introduction Traumatic brain injury (TBI) is a common cause of death and disability worldwide. TBI is a public health priority because it is associated with extensive physical, psychological and social impacts and a high economic burden. [1] Some studies have demonstrated that more than 10-40% of patients with TBI are still disabled 6-12 months after trauma, including those with mild TBI and unremarkable neuroimaging findings. Although early recognition and proper management of patients with TBI may result in better rehabilitation and substantial outcome improvement, assessing different cellular and clinical aspects and effects of TBI is still less than optimal. [2][3][4] S100B, a calcium binding protein highly expressed in astroglial cells of the brain and released in cerebrospinal fluid (CSF) and blood, can be measured by available immunoassay kits. Different studies have evaluated S100B as a biomarker for different brain injuries, such as stroke [5,6] , bacterial meningitis [7] , carbon monoxide poisoning [8] and TBI [9][10][11][12] . Some recent studies have also highlighted the complex release pattern of S100B and its potential role in brain tissue repair processes [13][14][15][16][17] This prospective study evaluates the diagnostic and prognostic roles of serum S100B protein in emergency department (ED) patients with minor to moderate TBI. Materials and Methods Patients were enrolled conveniently between March and May 2012 at two teaching hospitals with a total annual census of 80,000 adult patients. The institutional ethics committee (Faculty of Medicine, Iran University of Medical Sciences) approved this prospective study, and informed consent was obtained from all patients. Participants Patients at least 18 years old with a clinical diagnosis of acute mild to moderate TBI were enrolled. Patients with a history of isolated head trauma and Glasgow Coma Scale (GCS) score between 9 and 15 who presented in the ED within the first six hours of their head injury were considered to have mild to moderate TBI. All clinical assessments, including GCS calculations, were performed by a research assistant who was a physician. The research assistant was blinded to other assessments results. Patients with the following were excluded: severe TBI (GCS≤8); hemodynamic instability; body temperature greater than 38.5°C; concurrent trauma to any other organs; concurrent primary and secondary brain injury, including refractory severe hypoxia (arterial oxygen saturation <92% while receiving 100% oxygen), post-traumatic seizure, and skull bone fracture; and any other identified or suspected differential diagnosis for the patient's decreased level of consciousness, including alcohol abuse, drug abuse, substance abuse, drug toxicity, hypo/hyperglycemia, hypo/hypernatremia, endocrine disorder, or infection. Patients who did not undergo a head CT scan were also excluded. Intervention S100B assay: A blood sample was drawn from the peripheral veins within the first six hours of ED admission. The time of blood sample collection was recorded. Samples were centrifuged, and the serum was refrigerated at -20°C until analyzed. Neuroimaging: Ten millimeter thick slices obtained using a GE VCT Lightspeed 64 multi-slice detector were interpreted by a board certified radiologist and confirmed by another consultant radiologist who was blinded to the first interpretation. Both radiologists were blinded to the clinical conditions and S100B results of the patients. All pathologic findings, including skull bone fracture and any type of intracranial hemorrhage (e.g. brain contusion, subdural/epidural intracranial hematoma), were reported as positive computed tomography findings. Follow up: The patients were called by two blinded research assistant one month later. During follow-up, patients were evaluated for level of consciousness, presence or absence of post-traumatic headaches, and daily activity performance (using the Barthel scale) to determine if any significant intracranial complications had occurred (.i.e. complications requiring further neuroimaging). Lost to follow-up (n=19) -Wasted blood samples (11) -Refused to participate (6) -Failed to reach by telephone (2) Measurements Initial TBI severity was assessed using the GCS. Patients with GCS scores between 9 and 15 were considered to have mild to moderate TBI. To measure S100B serum levels, the human S100 ELISA kit (BioVendor -laboratorni medicina a.s., Brno, Czech Republic) was used. The lowest detection limit of the test is about 15 pg/ml. Serum S100B levels were measured in ng/dl. The Barthel scale is an ordinal 10-variable scale used to measure patient performance on daily activities and to predict the likelihood a patient will be able to live at home independently. The Barthel scale has high inter-rater and test re-test reliability, as well as, high correlations with other measures of physical disability. The ten Barthel scale variables are: presence/absence of fecal incontinence; presence/absence of urinary incontinence; and help needed with grooming, toilet use, feeding, transfers, walking, dressing, climbing stairs, and bathing. Each variable is given a score (between 0 and 3). These scores are summed to determine the total score (out of 20). The higher the Barthel score, the less assistance the patient is likely to need with daily activities after discharge from the hospital. For example, when a person can perform about 50% of their daily tasks and activities independently, then their Barthel score will be 10 out of 20. [18][19][20] Patient outcome measures were level of consciousness, residual headache, and Barthel score one month after trauma. Data Analysis The Student's t-test was used to compare the mean values of quantitative variables, and the Chi square test was used to compare qualitative variables. All data analyses were performed with SPSS version 13.5 (SPSS, Inc., Chicago, IL). Results One hundred eighty-seven patients were assessed for eligibility, and 78 patients were excluded from the study: six patients had hemodynamic instability; 18 patients had concurrent trauma to other organs; 34 patients had concurrent brain injuries; and 20 patients had other causes of decreased level of consciousness. Venous blood samples were obtained from 109 patients with minor to moderate TBI who had undergone CT as a part of their routine diagnostic evaluations. Eleven samples were wasted due to various errors between initial preparation and analysis. A total of 98 patients with mild to moderate TBI and available serum S100B results were followed. During the telephone follow-up one month post-trauma, six patients refused to continue participating in the study, and two additional cases were unreachable by telephone. Follow-up interviews were performed for 90 patients, all of whom completed the study. No patients had died in the month between injury and follow-up, and all patients had GCS scores of 15. Serum S100B levels were higher in patients with lower Barthel scores, but the difference was not statistically significant (p=0.06). Discussion The S100B protein has a half-life of two hours and can be measured both in CSF and in the blood. Although some studies have shown that S100B protein levels increase after extra-cranial injuries in the absence of brain injury, [21] many other studies have introduced S100B protein as a highly sensitive and specific biomarker of CNS injuries. [13][14][15][16][17] S100B has been suggested as a triage tool for identifying patients who need neuroimaging and as a diagnostic tool for early recognition of patients with possible brain tissue injury and timely administration of medication (e.g. benzodiazepines to reduce post-concussion syndrome risk after mild TBI). S100B has also been suggested as a prognostic tool to identify atrisk patients and to begin rehabilitation activities as soon as possible, especially for patients who do not need neurosurgical interventions. [22][23][24][25][26] The present study found that although serum S100B increas-es in minor to moderate traumatic brain injuries (especially in cases of DAI), it cannot accurately predict one-month outcomes. These results are compatible with some other studies which have emphasized the complicated release pattern of S100B. These past studies have highlighted the role of blood-brain barrier integrity and disruption in S100B release into the serum, the poor correlation between serum and CSF S100B levels, and the possible reparative roles of S100B that may improve outcomes in patients with acute brain injuries. These studies also mention that the relationship between S100B values and likely outcomes in patients with TBI is not necessarily a causative relationship. [27] A study of a large cohort of patients showed some association between high serum S100B level and poor outcome in patients with brain injury, but not significant enough to support use as an outcome prediction tool. [28] Similarly, a review by Townend showed that, although patients with high serum S100B levels at initial evaluation may be at higher risk for disability after TBI, no association between serum S100B levels and the neuro-psychological performance of injured patients has been established. [2] Metting et al. studied 94 patients with mild TBI and demonstrated that S100B is not related to outcome or imaging results. [29] Some newer studies have proposed that serum S100B level might be used for predicting the probability of brain death in patients with TBI. [30] Conclusion The current study showed that serum S100B levels increase with minor to moderate TBIs, especially in patients with suspected DAI. However, serum S100B cannot accurately predict one-month neuropsychological outcomes and performance. Limitations The present study has some limitations. The study was conducted at two teaching hospitals, and the human S-100 ELISA kits may not be available at other smaller hospitals. Only patients who had undergone brain CT were enrolled; patients who had not undergone CT or who refused to undergo neuroimaging were not included. The sample size was small, and similar studies with larger sample sizes would be preferable. The study did not focus on any cutoff S100B level to categorize at-risk patients, though it might be helpful to determine a cutoff diagnostic serum S100B value.	2018-04-03T05:05:21.171Z	2014-12-01T00:00:00.000	{ "year": 2014, "sha1": "00f893cbc22600111ea816db13dfd3e86b9f5f69", "oa_license": "CCBYNCND", "oa_url": "https://doi.org/10.5505/1304.7361.2014.74317", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "ad80ef586a54c7175b371eb91c8db25bb93f0ea5", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
221865336	pes2o/s2orc	v3-fos-license	Neutrophil-to-lymphocyte Ratio can Predict Outcome in Extensive-stage Small Cell Lung Cancer Abstract Background The neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and lymphocyte-to-monocyte ratio (LMR) were analyzed in various carcinomas and their potential prognostic significance was determined. The objective of present study was to determine the correlation between these parameters and the survival of patients with small cell lung cancer (SCLC), since very few studies have been published on this type of carcinoma. Patients and methods One hundred and forty patients diagnosed with SCLC at University Hospital Center Zagreb, between 2012 and 2016 were retrospectively analyzed. Extensive-stage disease (ED) was verified in 80 patients and limited-stage disease (LD) in 60 patients. We analyzed the potential prognostic significance of various laboratory parameters, including NLR, PLR, and LMR, measured before the start of treatment. Results Disease extension, response to therapy, chest irradiation and prophylactic cranial irradiation (PCI), as well as hemoglobin, monocyte count, C-reactive protein (CRP), and lactate dehydrogenase (LDH) showed a prognostic significance in all patients. When we analyzed the patients separately, depending on the disease extension, we found that only skin metastases as well as LDH and NLR values, regardless of the cut-off value, had a prognostic significance in ED. Meanwhile, the ECOG performance status, chest irradiation, PCI, and hemoglobin and creatinine values had a prognostic significance in LD. Conclusions NLR calculated before the start of the treatment had a prognostic significance for ED, while PLR and LMR had no prognostic significance in any of the analyzed groups of patients. Introduction Lung cancer is still one of the most malignant diseases nowadays. It is the most commonly occurring cancer in men and the second most commonly occurring cancer in women according to the latest data by the International Agency for Research on Cancer (IACR). 1 At the same time, lung cancer is the leading cause of cancer death among both men and women. For the purposes of comparison, breast cancer in women occurs three times more of-ten than lung cancer, while the mortality is almost equal. Moreover, prostate cancer and lung cancer have almost the same incidence in men, but the lung cancer mortality rate is four times higher than the prostate cancer mortality rate. 1,2 Small cell lung cancer (SCLC) is the most aggressive subtype of lung cancer. Nowadays, small cell lung cancer makes up about 15% of all lung cancers and occurs almost only in smokers. The incidence of this lung cancer subtype has decreased in the last few decades, but primarily in developed countries. 3,4 There are no global data on SCLC prevalence. In Croatia, there are no separate data on SCLC either, and the available epidemiological data relate to lung cancer as an entity. In the last twenty years, a slight reduction in the share of SCLC in relation to the total number of lung cancer patients has been observed at our institution, which is the largest thoracic oncology center in the country. According to literature there are differences in survival rates for various tumors, including small cell lung cancer, depending on ethnic origin. 5 Therefore, the results of epidemiological and clinical studies in one geographic area are not applicable to some other geographic areas. The main characteristics of small cell lung cancer are its rapid growth and early spread to distal body parts. This is the reason why in most cases this carcinoma is diagnosed late, when metastatic disease has already developed. 6 Surgical treatment is therefore rarely possible, but in the last few years it has been recommended for certain patients with early-stage disease. 7 Before the introduction of platinum-based antineoplastic drugs for the treatment of malignant disease, the median survival of patients diagnosed with small cell lung cancer was two to three months. 8,9 The survival rate has increased four to five times with chemotherapy, but for most patients with extensive-stage disease it does not exceed ten months. In fact, this tumor is extremely chemosensitive and usually responds to chemotherapy very well. However, it recurs very rapidly and most patients die after a relapse. Despite numerous clinical trials, progress in the treatment of small cell lung cancer has been modest. However, as treatment of limited disease (LD) became more successful with the introduction of thoracic radiotherapy and prophylactic cranial irradiation (PCI), concurrent chemoradiotherapy has been a standard in the treatment of LD for a long time now. 6 The optimal radiation therapy protocol has remained controversial until this day, although it has been established that there are no differences in either survival or toxicity between hyperfractionated and normofractionated radiotherapy. 10,11 The application of consolidation radiotherapy in selected patients with extensive-stage disease (ED) and a good initial response to chemotherapy have partly contributed to the improved survival rate, but application has been very inconsistent. 12,13 Immunotherapy has resulted in significant progress in the treatment of numerous malignant diseases, including nonsmall cell lung cancer (NSCLC). Expectations for the treatment of small cell lung cancer were high as well. For the time being, adding checkpoint inhibitors to first-line chemotherapy in ED has resulted in a slight increase of overall survival and progression-free survival, but the results are far from expected. [14][15][16] It is well known that infection and deregulated inflammatory response are associated with the occurrence and progression of almost all chronic diseases, including cancers. 17 In the last few decades, a great number of researches investigating the role of different inflammatory markers in cancer development and outcome have been published. [18][19][20] Usually the investigated inflammatory markers include C-reactive protein (CRP), lactate dehydrogenase (LDH), erythrocyte sedimentation rate, platelet (Pc) and neutrophil counts. [21][22][23] In most cases, it has been found that elevated levels of these parameters are associated with poorer outcome of various cancers, including small cell lung cancer. 24,25 On the other hand, the lymphocyte count reflects the immunological status of a host, thus a low lymphocyte count is a predictor of poorer outcome. 26 The prognostic value of combinations of these and other parameters has also been extensively investigated. Among them, the neutrophil-to-lymphocyte ratio in various chronic diseases, including numerous malignant diseases, has been investigated the most. [27][28][29] In this study, we have investigated CRP, LDH, Pc, hemoglobin (Hb), creatinine, neutrophil-tolymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and lymphocyte-to-monocyte ratio (LMR) and their impact on the outcome of patients with SCLC. To the best of our knowledge, this is the first study carried out exclusively on a European population which investigated the prognostic significance of all three mentioned ratios in patients with limited-stage and extensive-stage small cell lung cancer. [29][30][31] Patients and methods Patients For research purposes, we analyzed the medical records of 438 patients diagnosed with small cell lung cancer admitted to the University Hospital Center, Department for Lung Diseases Jordanovac between 2012 and 2016. We included only patients whose disease was verified by histopathological analysis and who had undergone first-line chemotherapy or chemoradiotherapy. Some additional criteria needed to be met in order to be included in the research: documented laboratory test results with the investigated parameters measured up to three weeks before the first chemotherapy, as well as data on performance status, follow up, and outcome. The following patient categories were excluded from further research: surgically treated patients, patients with combined small cell lung carcinoma, patients with one or more synchronous tumors, patients who received no therapy, patients without the required medical records, and patients lost to follow-up. After exclusion of the mentioned groups, 140 patients remained who met all the required inclusion and exclusion criteria for further investigation. Out of the total number of patients, 80 were diagnosed with extensive-stage disease and 60 with limited-stage disease. The patients' performance status was measured before the start of the treatment and defined according to the Eastern Cooperative Oncology Group Performance Status (ECOG) scale. 32 Regarding the ECOG status, the patients were divided into two groups: good ECOG status (0-1) and poor ECOG status (2)(3). All patients underwent a thoracic and abdominal computed tomography (CT) scan before the start of the treatment. Skeletal scintigraphy was done only in cases with a clinical indication, because it was not routinely performed at our Department. The same applied to brain CT scanning. Disease extension was defined according to the staging system established by the International Association for the Study of Lung Cancer (IASLC) in 1989, which divides SCLC into two stages, "limited-stage disease" and "extensive-stage disease". 33 The patients underwent follow-up chest X-ray scans after every two chemotherapy cycles. A follow-up CT scan was performed after the treatment was completed, especially in cases of initial limited-stage disease. Regression of a primary tumor and metastasis or stable disease was marked as response to therapy what was in fact disease control after initial therapy, whereas progression of the disease was marked as no-response. Response to therapy was assessed radiologically and clinically (e.g., if a patient had subcutaneous metastases or palpable lymph nodes in a region which had not been examined by CT). In our institution, patients usually receive 4-6 cycles of the first-line platinum-doublet chemotherapy. Patients who received a minimum of two and a maximum of six cycles of the mentioned chemotherapy, with or without radiotherapy, were included in the study. A concomitant or sequential radiotherapy protocol was carried out, primarily in patients with limited-stage disease or as palliative treatment in patients with extensive-stage disease and a good response to chemotherapy. Prophylactic cranial irradiation was mainly performed in patients with limited-stage disease. Data collection and ethical consideration Data were collected by using the electronic information database, based on good clinical practice and complying with international standards including the Helsinki Declaration on Patient Safety. We obtained approval for data collection and analysis by the Ethics Committee of our institution. Since this was a retrospective study, informed consent was not required. Demographic, laboratory, cytological, histopathological, clinical, and treatment data were collected on the patients included in the study. Laboratory test results obtained shortly before the start of treatment, that is, a maximum of three weeks before the first chemotherapy, were included in the study. Among all the hematological results, the following parameters were analyzed: leukocyte count, lymphocyte count, neutrophil count, monocyte count, platelets, hemoglobin, CRP, creatinine, and LDH. The neutrophil-to-lymphocyte ratio was calculated by dividing the total neutrophil count by the total lymphocyte count. The platelet-to-lymphocyte and lymphocyte-to-monocyte ratios were calculated in the same way. Overall survival (OS) was defined as the length of time from the date of diagnosis to death from any cause, or the last follow-up for patients who were still alive. Progression-free survival (PFS) was defined as the length of time from diagnosis to progression or death, depending on what happened first. Statistical analysis For the analysis of demographic and clinical data, we used descriptive and inferential statistical methods. Parameters are indicated as sum and percentage, arithmetic mean +/-standard deviation, or as interquartile range limits with the median as a measure of the central tendency. Differences among the ranked parameters, i.e., the investigated values, were calculated by using the Mann-Whitney U test. Differences among categorical data were tested by using the Chi-square test with Fisher's exact test for smaller samples. Intercorrelation among the variables was tested by using Spearman's rank correlation coefficient varying within the closed interval -1 ≤ r ≤ +1. For survival analysis, the Kaplan-Meier estimator was used, and the Log-rank test (Mantel-Cox) was used as a test of significance. The or for clinically relevant parameters. All P values were two-tailed. The level of significance was set at Alpha = 0.05. Statistical analysis was performed by using IBM SPSS Statistics for Windows, Version 21.0 (IBM SPSS Inc, Chicago, IL, USA). Cut-off values suggested by the literature were used for testing the potential prognostic value of the investigated ratios, since the ROC curves of the investigated ratios did not have a statistical significance. All ratios were tested regarding two cut-off values. The cut-off values for NLR were 4 and 5, those for PLR were 150 and 250, and those for LMR were 2.64 and 4.19. [34][35][36][37][38][39] Patient characteristics Patient characteristics regarding the disease stage are shown in Table 1. Out of 438 patients diagnosed with small cell lung cancer or mixed neuroendocrine carcinoma between 2012 and 2016, 140 met the inclusion and exclusion criteria and were included in the study. Of those 140 patients, 80 were diagnosed with extensive-stage disease and 60 with limited-stage disease. The mean patient age was 63.1 years with a mean deviation of 9.2 years (42-87 years of age). Slightly more males than females were involved in the study (89 or 63.6%). The majority of the patients were smokers (95.7%), of good performance status, 0-1 according to the ECOG scale (82.9%). Only 14 patients (10%) received less than 4 chemotherapy cycles. Fortyfive patients (32%) underwent radiotherapy, most of whom were in the limited-stage disease group. Only twelve patients underwent PCI (8.6%), again significantly more in the limited-stage disease group. Disease control was observed in 119 patients (85%). After two years, 125 patients (89.2%) died. Fifteen out of the total number of patients included in the analysis (10.7%) survived for more than 2 years, and all of them belonged to the limited-stage disease group. According to the statistical analysis, disease control, PFS, OS, and outcome were significantly better in the limited-stage disease group. Of the laboratory parameters, a significant statistical difference regarding the disease stage was only observed for CRP and LDH. The mean NLR and PLR values were higher in the extensive-stage disease group of patients, while the mean LMR value was higher in the limited-stage disease group, but the difference was not statistically significant. In the extensive-stage disease group, a statistically significant difference of LMR values regarding patient According to the Kaplan-Meier estimator, survival analysis of all 140 patients showed a statistically significant difference in the overall survival regarding disease extension, radiotherapy to the primary tumor, prophylactic brain irradiation and disease control. Therefore, patients with limited-stage disease, patients with disease control, irradiated patients and patients who underwent PCI had a better survival. Of the laboratory parameters, a statistically significant difference in the overall survival was observed regarding the hemoglobin, CRP, LDH, and boundary monocyte values, whereas a statistically significant difference in the overall survival regarding the ECOG status, NLR, PLR, and LMR was not observed ( Table 2). Separate testing showed a statistically significant difference in overall survival in patients with extensive-stage disease, considering the presence of skin metastases and laboratory parameters including LDH and NLR, regardless of the cut-off values. Therefore, a better overall survival was observed in the patients who did not have skin metastases and had lower LDH and NLR values (Table 3). No positive correlation between overall survival and ECOG status, number of metastatic sites, and disease control was observed in the subjects with metastatic disease. A statistically significant difference in overall survival, regarding the ECOG status, radiotherapy of the primary tumor, prophylactic cranial irradiation, and laboratory values such as hemoglobin and creatinine levels, was determined in the limited-stage disease group of patients ( Table 4). As we have already mentioned, Cox regression was used for determining possible multiple interactions among the variables. Thus, all statistically significant parameters from the Kaplan-Meier analysis were included in the multiple regression model. In this model LDH became the most significant prognostic factor in extensive-stage disease, while the ECOG performance status became the Table 5. Discussion Numerous prognostic factors were investigated in various cancer types in order to find the factor which would most accurately define the patient groups that could benefit from a certain therapy and consequently expect a better survival. 39 The established fact about the important role inflammation plays in the process of carcinogenesis has led to research into the prognostic significance of various inflammatory markers. In the past decade numerous papers have been published on such research in relation to non-small cell lung cancer 29,31 , but, very few studies of this kind have been done for small cell lung cancer. The present study was conducted with the intention to determine potential prognostic parameters of survival in a European population of patients diagnosed with SCLC. Survival parameters were identified for the whole population of patients, as well as separately for patients with extensive-stage and those with limited-stage disease, in order to determine differences between these two groups. As disease extension and performance status are generally among the most investigated prognostic parameters, they were verified as the most important for SCLC as well. 23 Our study also showed that disease extension was a significant prognostic factor, and certainly the most significant predictor of longer survival. On the other hand, performance status showed a prognostic value only for the limited-stage disease patient group, which can be explained by the fact that it was possibly assessed more accurately in this patient group. As a matter of fact, performance status assessment is a subjective method and in retrospective studies there is always a possibility that the criteria for certain patients varied. Unlike in other neoplasms, age did not have a prognostic significance in most of the studies regarding SCLC, which was confirmed in our study, too. 23 Neither gender nor smoking status had a prognostic significance, but, it is noteworthy that the number of non-smokers in the study was negligible. Of all the variables, radiotherapy, PCI and disease control had a survival impact in the whole research patient group. When we separated the patients with extensive-stage from those with limited-stage disease, radiotherapy and PCI retained a survival impact in the patients with limited-stage disease, as we expected. However, disease control showed prognostic value neither in LD nor in ED. In the last few decade various laboratory parameters regarding prognostic value have been investigated. Their ratios have also been investigated recently. Some studies verified a prognostic significance of hemoglobin, leukocyte count, CRP, LDH, and serum sodium concentration in SCLC. 23,25,40 The prognostic significance of hemoglobin and LDH was confirmed in our patients, along with a lower significance of CRP and monocyte count as prognostic factors. When we excluded disease extension from the analysis, LDH retained a prognostic significance in the ED group, while hemoglobin retained a prognostic significance in the LD group of patients. Besides, creatinine level occurred as an independent prognostic factor for survival in the LD group of patients, but again only in the extremely small number of patients with increased creatinine levels. Although the combinations of various laboratory indicators, including NLR, PLR, and LMR, have already been examined as prognostic factors in SCLC, a relatively small number of studies have been published regarding this type of cancer. Most of the published papers investigating the predictive significance of these parameters in patients with lung cancer address non-small cell lung carcinoma. [29][30][31] Consulting the literature in English until May, 2020, we found a total of twenty studies, seven of which had been published in 2019, which investigated one or more of these three ratios in patients with small cell lung cancer. It is interesting to note that most of the studies relate to the Asian population. For example, the prog- 35 NLR, as the most researched ratio, was the subject of investigation in seventeen studies, of which only three were European. 25,35,42 There are only two studies investigating the prognostic role of NLR and/or PLR exclusively in the ED group of patients. 34,43 To our knowledge, to date neither of these two parameters have been investigated on a European population in cases of extended SCLC. As race has been determined as a significant prognostic factor in SCLC patients, in the sense that being Caucasian represents a favorable independent prognostic factor, we were interested in whether our results would differ from the ones obtained elsewhere so far. 5 It is important to mention that the results of the former studies are inconsistent, that is, some studies showed a statistically significant correlation between the NLR and PLR ratios and overall survival of the patients, while others did not yield a statistical significance. In fact, some studies didn't investigate these ratios in correlation with survival at all. [44][45][46] The only prospective study conducted in the USA on more than 900 patients verified that NLR was a prognostic parameter for OS only in the extensive-stage disease group of patients, which is consistent with our results. 47 The same study established that PLR was a prognostic parameter for OS only in limited-stage disease, which was different from our results. There are no prospective studies for LMR. Most retrospective studies which investigated NLR established its prognostic value, regardless of whether it was investigated in LD, ED, or simultaneously in both patient groups. Among twelve retrospective studies investigating PLR, only three showed a prognostic significance of this parameter. [48][49][50] Out of the two studies investigating LMR, only one showed a prognostic significance of this parameter. 38 In the prospective study mentioned above, among other things it was established that NLR and PLR were statistically significantly greater in patients with extended disease. 47 In our study, the mean values of NLR and PLR were also higher in ED patients, while LMR was higher in LD, although the difference was not statistically significant. On the other hand, we found statistically significant differences in LMR values in correlation with patient age in the ED group, i.e., higher LMR values in the younger age group of these patients. In spite of the fact that some of our results were consistent with those from the only prospective study, our study had numerous limitations. In every study where data are collected from available records, there is a possibility that some of it may not be reliable, particularly data undergoing subjective assessment. As mentioned earlier, performance status is one of such parameters, thus making it more difficult for analysis in retrospective studies. A similar situation may arise in the assessment of peripheral lymph node regression during patient follow up and evaluation of the response to treatment. Furthermore, in the determination of disease extent, especially in concomitant chemoradiotherapy candidates, assessment based only on clinical examination, bronchoscopy, and CT is not sufficient. Since this type of carcinoma is characterized by rapid spread, complete staging should be done prior to treatment, including brain CT and bone scintigraphy. This is the standard procedure at our institution today, but was not always possible in the past for technical reasons. The relatively small number of subjects enrolled in the study was also a limitation. However, two published studies enrolled approximately the same number of patients. 51,52 Also, some of the published studies were conducted in even smaller groups of participants. 44,51,53 Although some studies had a large number of patients, they didn't analyze patients separately considering disease extension. 54 It is important to note that the number of patients enrolled in the study was probably not adequate for the analysis of certain variables. Namely, only a very small number of patients with skin metastases and increased creatinine participated in the study, as well as very few patients with a low performance status. This presents a problem for many studies, since low-performance status patients are usually not candidates for differential treatment and are rarely included in clinical studies. The same applies for kidney failure patients. On the other hand, since the skin is an uncommon metastatic site, such patients are rare. Considering the confidence interval, it is clear that according to this study skin metastases are not a favorable indicator of survival. On the contrary, creatinine can be considered a favorable indicator of survival despite the small number of patients. As far as the investigated treatment procedures and their prognostic values are concerned, there are certain limitations as well. In the group of all patients, statistically significant differences were found for survival in relation to PCI and thoracic irradiation. However, when the patients were analyzed separately in relation to the extent of the disease, those differences disappeared in the ED group. This is due to the fact that disease extent is one of the most important prognostic factors for SCLC, which was established in 2003 in a prospective study involving 436 patients. 23 Therefore, these two patient groups should always be investigated separately, because the differences in their prognoses entail different modes and aims of treatment. In our study, PCI remained prognostically valuable in the LD patient group, but with an insufficient number of subjects for the result to be considered reliable. This treatment procedure has always been controversial, presenting an issue for confrontation and opposing research. 55 The prognostic value of PCI was certainly not the primary aim of our study. In spite of its limitations, we believe that our study will contribute to the elucidation of small cell lung cancer, as well as stimulate further research on this type of carcinoma, which has somehow always remained in the margins of lung cancer research. Conclusions The objective of this study was to determine a potential prognostic value of the neutrophil-tolymphocyte, platelet-to-lymphocyte, and lymphocyte-to-monocyte ratios in patients diagnosed with extensive-stage and limited-stage small cell lung cancer. To the best of our knowledge, this is the first study carried out on a European population which analyzed all three of the mentioned ratios. According to the study, NLR could be a good prognostic marker in patients with extensive-stage SCLC. Further prospective studies are definitely needed for this type of cancer.	2020-09-24T13:06:18.296Z	2020-09-22T00:00:00.000	{ "year": 2020, "sha1": "651686c74b77f14592e08db94199aad00cc5d830", "oa_license": "CCBYNCND", "oa_url": "https://doi.org/10.2478/raon-2020-0054", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "5635823aa5031b43102a13bc5be7a5ea87dd24c5", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
254720708	pes2o/s2orc	v3-fos-license	Research on composition analysis and identification of ancient glass products related modeling : In this paper, in order to solve the problem that ancient glass would be weathered by the environmental influence during the burial process, and its chemical elements would change accordingly, based on Chi-square test and Spearman correlation coefficient analysis, the relationship between the types of glass relics and chemical components was analyzed, and the corresponding relationship was obtained.Then, the glass relics were classified and predicted by decision tree and random forest algorithm, and the classification and prediction results were given. Introduction Glass is a testament to the friendly exchanges between China and the West along the ancient Silk Road.Ancient Chinese glass absorbed Western technology and was locally manufactured, similar in appearance to imported products, but different in chemical composition.The main component of glass is silicon dioxide.Because the choice of flux is different, the other main chemical composition of glass is also different.The known state-owned heritage glass is divided into lead-barium glass and potassium glass.In the long history, many precious ancient Chinese glass products were buried in the earth or exposed to the air, and were more or less weathered. The composition analysis of these cultural relics is conducive to the analysis of ancient Chinese glass products technology, which has a high research value. Aiming at the research problems of ancient glass products, this paper analyzes the classification rules of high-potassium glass and lead-barium glass. Based on the classification problems, the principal components are used to subclassify the glass, and the rationality and sensitivity of the classification results are tested. (1) Data preprocessing: the known data are merged, and the number, type, surface weathering and chemical composition content of the cultural relics are put into a table. Glassware classification model (2) the selection of parameters and indexes before data analysis: use of the classification of decision tree learning function of SPSSPRO software "comprehensive" table data is analyzed [1], the nominal variable selection type glass, glass types are also studied at the same time, quantitative variable selection the chemical composition, chemical composition is as the glass type parameters.Due to the large number of chemical components in the table and the small change in the content of some chemical components, it does not have high reference value, so we chose to abandon the chemical components with small reference value and reduce the index.In principle of index selection, we should select the chemical components with high content and big difference between the two types of glass as indicators.Combined with the solution results of the research background, silica and potassium oxide have a greater contribution to the weathering of high-potassium glass.Barium oxide and lead oxide have a greater contribution to the weathering of PB-barium glass.Finally, we selected silicon dioxide, potassium oxide, lead oxide, barium oxide, four chemical components as indicators to analyze the classification rules of glass types [2]. (3) Data analysis: Taking glass type as the research object, silica, potassium oxide, lead oxide, barium oxide, four chemical components as indicators to analyze the classification rules of glass types [3]. (4) Using the four chemical components as indicators, the analysis results are as follows: the decision tree classification learning training sample 46, in which the lead oxide content lower than 5.355% is classified as high-potassium glass, and the lead oxide content higher than 5.355% is classified as lead-barium glass.The structure of decision tree is shown in figure 1 [4]. Table 1 [5]. Figure 2. (6) Further thinking and improvement of the model: Although the accuracy of the above results reached 100%, only one of the four indicators selected was included in the consideration scope. In view of the small number of training samples of the model, the results of the model may be accidental, so we analyzed the data for several times after shuffling and recombination, and found that the results were still the same.In order to eliminate the possible contingency and find new classification rules to make the results more accurate, the lead oxide index was deleted and the remaining three indexes were used as parameters to analyze the classification rules [6]. The following results were obtained by using the three indexes to analyze the classification rule: barium oxide content more than 3.06% or barium oxide content less than 3.06% and silica content less than 47.645% were classified as lead barium glass, and lead oxide content less than 3.06% and silica content more than 47.645% were classified as high-potassium glass.The improved decision tree structure is shown in figure3 [7]. (7) Evaluation of the new model: the test result accuracy of the test set based on decision tree classification learning reached 100%, and silica accounted for 23% of the classification weight, barium oxide accounted for 76.8%, and potassium oxide accounted for 0%.The accuracy of the results is reliable [8]. Table 2 Evaluation results of the new model Accuracy Recall rate Accuracy rate F1 Traning Set The characteristic importance of each indicator of the model is shown in Figure 4. (8) Analysis results of classification rules: The content of lead oxide lower than 5.355% is classified as high potassium glass, and the content of lead oxide higher than 5.355% is classified as lead barium glass.The content of barium oxide more than 3.06% or barium oxide less than 3.06% and silica less than 47.645% is classified as lead-barium glass, and the content of lead oxide less than 3.06% and silica more than 47.645% is classified as high-potassium glass [9]. Solving the classification model of glass products For high potassium, the process of solving the sub-classification of lead-barium glass is as follows: (1) Data preprocessing: the data in the "comprehensive" table were divided into two categories: high-potassium glass and lead-barium glass, which were respectively saved into the new table and named as "high-potassium" and "lead-barium". (2) Cluster index -selection of chemical composition [10] Considerations before selection: for the same type of glass, the content of each chemical composition is different and has differences. For example, the content of silica in high-potassium glass is high, and the content of sodium oxide is very low, or even there is no sodium oxide.In the process of clustering iteration, since the sodium oxide content of most high-potassium glasses is close to 0 or even 0, the sodium oxide content in the clustering center will always be near 0, which affects the clustering accuracy.Therefore, in the selection of clustering indicators, we will choose to discard such chemical components as sodium oxide. Selection of chemical components: we first cluster all chemical components as indicators, set the cluster number to 2, and get the clustering results. Then we eliminate the chemical components that are significantly different from the divided categories, and get the indicators for the final clustering. The difference results of each index and classified category of high-potassium cultural relics in the first clustering are shown in Table 3. When the P value is greater than 0.05, it can be considered that the component is significantly different from the classified category, which will affect the clustering accuracy.It should be omitted.Therefore, the final selection of high potassium index is silica, potassium oxide, calcium oxide, alumina, iron oxide. Table 3 Difference results of various indexes and classification of high-potassium cultural relics clustering Table 4 shows the difference results of indicators and classification of the first clustering of leadbarium cultural relics. Similar to high-potassium cultural relics, the final indicators selected for leadbarium cultural relics are silica, sodium oxide, calcium oxide, alumina, lead oxide, phosphorus pentoxide, and strontium oxide. Table 4 Difference results of various indexes and classification types of lead and barium cultural relics clustering (3) Results of cluster analysis High-potassium cluster analysis results: As shown in Table 5, there were significant differences in each index and classification category, and the clustering results were highly reliable. Table 6. According to the component contents of the cluster center coordinates, the first cluster is named as the high silica content class, and the second cluster is defined as the low silica content class. The corresponding cultural relics of the high potassium category after clustering are shown in Table 7. Table 7 Cluster center coordinates The results of lead and barium cluster analysis are shown in Table 8.The clustering results showed that there were significant differences in each index and classification category, and the clustering results were highly reliable. Table 9. 223 After clustering, the first type was named as low silicon and high lead class according to the component content of the cluster center coordinate, and the second type was defined as high silicon and low lead class.The corresponding species of each cultural relic of lead and barium species after clustering are shown in Table 10. Due to space limitation, only part of the data are listed in this paper. Analysis of sensitivity and rationality of subclassification results (1) The results of two cluster analyses showed that there were significant differences between the selected indicators and the classified categories, so the classified categories were strictly classified according to the content of chemical components of cultural relics. (2) The data tables after subclassification were saved as "high potassium subclass Division" and "lead and barium subclass Division", and the data in the tables were trained by decision tree.The species after clustering were selected as the research objects, and the chemical components in the clustering step were selected as indicators.In other words, silica, potassium oxide, calcium oxide, alumina and iron oxide were selected as indicators for high-potassium classes.Lead barium selected silica, sodium oxide, calcium oxide, alumina, lead oxide, phosphorus pentoxide, strontium oxide as indicators.The decision tree classification analysis (this step is the same as the principle of analyzing the classification law of cultural relics) was carried out to obtain the accuracy of the model. If the accuracy is high, the model is reasonable. The precision rate of high potassium subclass classification is shown in Table 11. The accuracy of lead and barium subclasses is shown in Table 12. In the subclassification rationality analysis step, the data table after subclass division is put into the decision tree classification learning model for training, and the trained model has a high accuracy.Therefore, we can use this model to analyze the sensitivity.Firstly, we modify the chemical composition of some cultural relics, and take the cultural relics with modified chemical composition as new cultural relics, and put them into the trained model for prediction. The categories before and after modification are compared. If the categories before and after modification are the same, the model is stable and the classification results are stable. High potassium cultural relics mainly modified the composition of silica, potassium oxide and calcium oxide.Comparison of chemical composition and species of high-potassium cultural relics before and after modification: (There is no column of cultural heritage label in the prediction result, because the number of cultural heritage number will affect the clustering result.However, the unmodified chemical composition content comparison can be used to judge the one-to-one correspondence between the two tables of cultural relics), as shown in Table 13. Table 13 Results of high potassium cultural relics The prediction results of the model are shown in Table 14. The compositions of silica, lead oxide, and barium oxide were mainly modified in lead-barium relics.The chemical composition and species comparison of lead-barium cultural relics before and after modification are shown in Table 15. Table 15 Result Table of lead and barium cultural relics The prediction results of the model are shown in Table 16. Table 16 Model prediction results Table Prediction Results Y The first probability of prediction results The second probability of prediction results SiO2 Na2O CaO Al2O3 PbO P2O5 SrO After two comparisons, the subclasses of the same cultural relics remained unchanged before and after modifying the chemical composition content, indicating the stability of the model. Conclusion In order to solve the problem that ancient glass would be weathered by environmental influence during burial and its chemical elements would change accordingly, this paper analyzed the relationship between the types of glass relics and their chemical components based on chi-square test and Spearman correlation coefficient analysis, and obtained the corresponding relationship.Then, decision tree and random forest algorithm are used to classify and predict the glass relics, and the classification and prediction results are given.	2022-12-16T16:09:52.693Z	2022-12-04T00:00:00.000	{ "year": 2022, "sha1": "ba585ffe32b0f0c75e43a25d2c0e3a9441a623f8", "oa_license": "CCBYNC", "oa_url": "https://drpress.org/ojs/index.php/HSET/article/download/3158/3045", "oa_status": "HYBRID", "pdf_src": "Anansi", "pdf_hash": "ab47a4e861c9aa6c6a5a950bbc8070a39642ffe3", "s2fieldsofstudy": [ "Materials Science" ], "extfieldsofstudy": [] }
235504047	pes2o/s2orc	v3-fos-license	Discovery of New Theory Analysis of Equilibrium Point Population Versus Food on Theory of Thomas Robert Malthus and Its Development ( Case Study of Indonesia ) The future of humans on this tiny planet earth has entered a grim beginning in the midst of rapidly growing technological progress. How not, human life whose population is growing fast is not comparable to food as a source of life that grows slowly. This study aims to calculate the cross point or equilibrium point between population and food using the population series and food series of Thomas Robert Malthus original and the results of its development by Matius Irsan Kasau. The data and methods used are types of secondary data sourced from the Indonesian Central Statistics Agency (BPS), which is processed by a mathematical method that compares the population with food in each series. The results of research with Indonesian data in 2010 showed that for the original Malthus theory with the number of children on average per couple of 4 people, the distance between generations 25 years, 75 years of life expectancy, population 237 million, food 66.5 million tons obtained equilibrium point occurred in 2085, namely in the third generation. As for Malthus's theory of development results with an average number of children of 2.6 people, a distance between 23 years generation, and 69 years of life expectancy, the equilibrium point was obtained in 2171, namely the seventh generation of the current generation. I. INTRODUCTION This More than 200 years ago, precisely in 1798 a British pastor and mathematician from the University of Cambridge named Thomas Robert Malthus stated two very famous natural laws of population and food, namely (Asha,1988), (Kasau, compared to the growth of food production which actually grows very slowly, then there will be a cross over point which is the break-even point between the population and food. This break-even point is called equilibrium point, which is the point where the food crisis starts where the population begins to experience mass hunger and death. Matius Irsan Kasau, ST. Aminah Dinayati Ghani If it is considered at a certain time the population is people and the food production capacity at that time is kg, then the two rows can be written one generation each, and three generations live together ( In case of: then the food is still abundant then the food breaks even with the population, the beginning of the food crisis then food is not enough anymore, there is mass hunger and death. The number 120 is the Mo food equivalence rate with the Go population, which means that one population needs 10 kg of food to be able to live for one month or 120 kg of food (rice) to live for one year. Therefore the unit of food Mo must be in kilograms where 1 ton = 1000 kg. Development of the Malthus Theory The original Malthus Theory has a number of drawbacks as follows Kasau, Jilid 1, 2019), (Kasau, 2012), (Kasau,2009) : 1. The number of series numbers is rigid, so is the time range of population folding 25 years. 2. There is no correlation between the population series and the food series. Both series grow on their own from the tribe to the next tribe separately. 3. Less formulated clearly so implementing it is very vague. These shortcomings can be corrected, among others, by taking into account the influence of population and food growth which is a function of: pair factor " " defined as half of the average number of children per couple, distance between generations T and life expectancy. For = 3T population growth and food can be expressed (Kasau, Jilid 1 and Jilid 2, 2018): Population 1 and 3 generations: ; (2.6a) ; (2.6b) 1 and 3 generation food: ; ; (2.7a) ; (2.7b) Therefore the situation of food availability 1 and 3 generations has a tribe with the equation: ; (2.8a) ; (2.8b) Note that for the average number of children per couple is 4 people or pair factor a = 2, then the terms of the original Malthus theory 1 and 3 generations live together, but for the average number of children per couple or pair factor other Malthus theories The original is no longer valid so that the theory of the results of its development can be broadly varied and flexible. Table 1 shows the ratio of food to population in each generation for 5 different pair factors, namely: 1; 1.3; 1.5; 2.0 (Malthus), and 2.5. The numbers in Table 1 show the values of and with values of n = 0,1,2, ..., 10 which if multiplied by the value K will get the value and . Note that for the total number of children per couple of 2 people or pair factor ( ) is 1, then it can be seen in Table 1 the comparison coefficient of the serial rate does not change, but always remains . This shows that the amount of population and food produced all the time is always available safely, never going up or down. Meanwhile, the greater the pair factor or the more the number of children on average per couple the coefficient of comparison of the series of rates decreases. This shows that the increase in population is far greater than the increase in food, so that the time of food insecurity and the food crisis is accelerating. Strengths, Weaknesses, and Mistakes of Malthus Theory a. Advantages 1. Strong and fundamental natural law that can be expanded and developed 2. Applies specifically such as: Theorem of Pythagoras (trigonometric mathematics), Newton's Law (Physics), Adam Smith's Economic theory (macroeconomics), Cobb-Douglas production theory (microeconomics) and so on. 3. "Early Warning Bells" will occur "Explosion of Population" which will lead to famine and mass death due to the food crisis. 4. Is the "Forerunner" inspiration of the end of the Family Planning program (KB) throughout the world. Based on the original Malthus equilibrium point between the population and food in Indonesia it is predicted that it will occur around 100 years from now. 2. Based on the theory, the results of the development of the theory of Malthus equilibrium point between the population and food in Indonesia are predicted to occur around 200 years from now. 3. The population at equilibrium point in Indonesia is predicted to be between 1 billion and 2 billion with food needs of around 120 billion kg to 240 billion kg of rice per year. III. RESEACH METHOD AND RESULT This study took the case of Indonesia, using secondary data in 2010 sourced from the 2016 BPS as follows: Food production Table 1 column (Malthus) obtained the ratio of and whose value is around 0.368 is located between and , while for located between with . This comparative number is located in lines n = 2, n = 3, and n = 4. The average or middle number is n = 3, which means the third generation of the initial generation (n = 0). Because according to Malthus' theory the distance between generations is 25 years, then the food crisis that causes mass starvation and death in Indonesia will occur at the fastest nT = Table 1 column the ratio of and is obtained, whose value is around 0.420 is located between and , while for located between and . This comparative number is located in lines n = 6, n = 7, and n = 8. Average or middle number n = 7 which means generation seventh from the initial generation (n = 0). Because based on the development of Malthus's theory of intergenerational distance (Indonesian data) of 23 years, the food crisis that causes mass starvation and death in Indonesia will occur at the fastest nT = (7) (23) = 161 years from 2010 to come or in 2171. Note that equilibrium points for pair factor = 1.5 are definitely faster than for pair factor = 1.3 and slower than for pair factor = 2 (Malthus) or around generation n = 4, 5, and 6 While for pair factor = 2.5 which is greater than Malthus' theory, it will be even faster, which is around generation n = 1, and 2. This explanation shows that the more the number of children on average per couple and the smaller the interval between generations, the faster the equilibrium point will occur. Based on the results of calculations in Table 2 and Table 3, the population graph and food of the original Malthus theory and its development as a function of time, calculated from 2010 as the initial calculation until the occurrence of equilibrium points, can be described as in Figure 1. This is caused by differences in 3 things, namely the number of children on average per couple, distance between generations, and life expectancy. As a criticism of the original Malthus theory and the results of its development, there are two weaknesses in the food series as follows: 1. The initial food data used is data in 2010 obtained equilibrium point EP1 and EP2 as shown in Figure 1. If the food data was initially advanced for example in 2000, the equilibrium points EP1 and EP2 also progressed to the year before 2085 and year 2171. 2. The amount of food needed in equilibrium points EP1 and EP2 is not necessarily the maximum capacity of all agricultural areas in Indonesia. The numbers mentioned in Figure 1 can be below or above the maximum capacity. However, assuming these two weaknesses do not occur, the analysis and discussion of the results of the study are as follows: Analysis and Discussion of Research Results It can be seen in Table 2 and Figure Table 3. Look again at Table 2, the greater the pair factor a or the more the number of children per wife the fewer the number of generations needed or the faster it reaches the equilibrium point, even for pair factor a equals 1 or the number of children on average per couple equilibrium points never occur or occur in infinite places (~). This is very easy to understand because for the average number of children per couple 2 people, the population from time to time never increases, the last number always equals the number of dead. IV. CONCLUTION AND DISCUSSION A. Conclussion 1. The population and food model for calculating equilibrium points between population and food has been reduced so that equilibrium points can be known. The equilibrium point of the original Malthus theory in Indonesia calculated from 2010 (the latest population census data) will occur in the third generation (generation of great-grandchildren from the population of 238 million from the 2010 census).	2021-06-22T17:56:07.834Z	2021-03-30T00:00:00.000	{ "year": 2021, "sha1": "d826e91363876dbb9a94b9d061f9bc3ba9f490bc", "oa_license": null, "oa_url": "https://doi.org/10.35940/ijrte.e5209.039621", "oa_status": "GOLD", "pdf_src": "MergedPDFExtraction", "pdf_hash": "dbb279f7c841736791055291b0281bd38de99452", "s2fieldsofstudy": [ "Environmental Science", "Economics" ], "extfieldsofstudy": [ "Economics" ] }
146588	pes2o/s2orc	v3-fos-license	The Clinical Pharmacokinetics and Pharmacodynamics of Warfarin When Combined with Compound Danshen: A Case Study for Combined Treatment of Coronary Heart Diseases with Atrial Fibrillation Warfarin is used as anticoagulant and Compound Danshen prescription (CDP) is able to promote blood circulation. The combination might produce a synergic effect for patients of coronary heart diseases (CHDs) with atrial fibrillation (AF). Whether the combination increases the bleeding risk of warfarin is unclear, so the effects of Compound Danshen dripping pill (CDDP) on the pharmacokinetics (PK) and pharmacodynamics (PD) profiles of warfarin was investigated in patients. The dose and blood concentrations of warfarin, the four indicators of blood coagulation, prothrombin time, activated partial thromboplatin time, thrombin time, fibrinogen, and international normalized ratio value were compared when with and without CDDP treatment. The population PK (PPK) and PPK-PD models were established to assess patient demographics, genetic polymorphisms and CDDP as covariates. And the Seattle Angina Questionnaire was used to evaluate clinical efficacy, and the bleeding risk of combination was analyzed. The results indicated that CDDP had little influence on PK and PD profiles of warfarin in most patients and the combination of CCDP and warfarin would be a promising alternative regime for CHD with AF patients. The study was registered on China Clinical Trial Registry with number ChiCTR-ONRC-13003523. INTRODUCTION It was reported that about 10%∼15% patients in coronary heart diseases (CHDs) would be associated with atrial fibrillation (AF). Moreover, about 30% patients in AF would be associated with CHD (Jason et al., 2014). The main hazard of AF is thromboembolic complications. Anticoagulants may reduce the risk of death rate in AF patients by 38% (Alpesh, 2013), so about 70-80% patients with AF are suitable for long-term use of warfarin, which was initially used in humans in the early 1950s as vitamin K antagonist (Liu et al., 2014;Darcy et al., 2017). It is more than 100 years history in China that Compound Danshen prescription (CDP) has being applied to treat CHD, which consists of Radix salvia miltiorrhizae, Radix notoginseng, and Borneolum (Xin et al., 2013). In order to get market approval in the United States, the Compound Danshen dripping pill (CDDP), one Chinese patent drug of CDP, had recently completed a multinational phase III clinical trial. Pre-clinical and clinical studies have suggested that CDDP may increase coronary flow-rate and superoxide dismutase activity, expand blood vessel, promote blood circulation, relieve blood stasis, improve microcirculation, and improve hemorheological property, as well as decrease myocardial oxygen consumption (Pei et al., 2004;Zheng et al., 2007). Concurrent use of CDDP with warfarin may be a desirable combination that may produce a synergistic effect, to relieve the symptoms of CHD with AF by CDDP part and meanwhile to decrease the incidence of thromboembolic complication by warfarin. Warfarin treatment is difficult to handle due to its narrow therapeutic window with a large inter-individual variability in the dose-response relationship (Zeng et al., 2016). Both pharmacodynamic (PD) and pharmacokinetic (PK) factors may contribute to more than 10-20 fold inter-individual variability in dose requirement (Hamberg et al., 2007;Steven et al., 2011;Zeng et al., 2016). Whether the CDDP could have impact on the PK and/or PD characteristics of warfarin increasing the bleeding risk as a result, is not clear. It is necessary, therefore, to get the information about the interactions between CDDP and warfarin. Only one literature report has been retrieved to address the interactions between CDDP and warfarin in rats (Chu et al., 2011), but any information on such interactions in humans has not been reported. We conducted the study to explore the potential effects of CDDP on the PK and PD of warfarin in patients. During two periods on and off CDDP, we collected the dose and blood concentration of warfarin, the four indicators of blood coagulation, international normalized ratio (INR) value, and to establish appropriate population pharmacokinetics (PPK) and population pharmacodynamics (PPD) models to assess patient demographics, genetic polymorphisms and CDDP as covariates to evaluate the interaction effects of CDDP on warfarin. The seattle angina questionnaire (SAQ) was used to evaluate the effect of warfarin combined with CDDP on CHD with AF patients. In addition, 2 years follow-up was done after the two periods to learn about the drug compliance, the incidence of bleeding and other important outcomes, such as myocardial infarction, severe arrhythmia, revascularization, death and so on. We hope the study could provide useful clinical information for patients of CHD with AF. Patients The study was conducted in four hospitals in Tianjin from November 2013 to January 2016. Participants, suffered from CHD with AF, had been administrated warfarin with a long time. The study included two periods, in the first period patients took warfarin (Orion Corporation, Finland) at dose titration manner guided by INR values on the daily determined basis to guarantee the INR value maintained in the range of 2-3. When the INR value reached stable and maintained for successive 2 weeks on the fixed dose of warfarin, then the participants switched to the second period, in which, 10 dripping pills of CDDP (Tianjin Tasly Group Co., Ltd, China) were added orally to patients three times per day for at least 4 weeks. The dose of warfarin was re-adjusted according to the changed INR value, until the INR value was stable again and the dose of warfarin was retained for 2 weeks. Four blood samples (4 * 3 ml) were collected at the end of each period for warfarin concentration assay. The sampling time points were arranged at trough (before the administration of warfarin), at peak, two random times on elimination phase (before the next dose of warfarin). All the time points of blood sampling and warfarin dosing were recorded accurately. This study was carried out in accordance with the recommendations of Ethics committees of the Second Affiliated Hospital of Tianjin University of TCM with written informed consent from (ICF) all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Ethics committees of the Second Affiliated Hospital of Tianjin University of TCM in August 2013. Genotyping The information of genes which affect the metabolism of warfarin was obtained by literatures (Lin G.G. et al., 2015;Zeng et al., 2016). In this study, genotyping of VKORC1, CYP2C9 * 3, CYP4F2, EPHX1, and PROC were detected. Genomic DNA was isolated from peripheral blood leukocytes using a Genomic DNA Purification kit. Individual single-nucleotide polymorphism (SNP) loci were amplified using the polymerase chain reaction, which provided a template for allele-specific primer extension. All genotyping were performed using the gene sequencing methods (Kumar et al., 2008;Jorgensen et al., 2009;Steven et al., 2011;Lin G.G. et al., 2015). The VKORC1 was classified by detection of 1173 C > T variant (rs9923231), the CYP2C9 * 3 was classified by detection of 1075 A > C variant (rs1057910), the CYP4F2 was classified by detection of C > T variant (rs2108622), the EPHX1 was classified by detection of G > A variant (rs2292566), the PROC was classified by detection of G > T variant (rs5936). Bioanalysis Plasma warfarin concentrations were determined using a highperformance liquid chromatography tandem mass spectrometry method. Good chromatographic separation was achieved on an Astec CHIROBIOTIC V column (250 mm × 4.6 mm i.d., particle size 5 µm) with acetonitrile-5 mm ammonium acetate with 0.1% acetic acid in water (30:70, v/v) as the mobile phase at a flow rate of 0.50 mL/min (Jin et al., 2012). The column effluent was analyzed using a mass spectrometer in multiple reactions monitoring (MRM) mode by AB Triple Quad 5500 system in positive mode. S-warfarin, R-warfarin, and Tolglybutamide (Internal standard, IS) were extracted from plasma samples by protein precipitation with acetonitrile. Calibration curves were linear with 50.00-2000 ng/ml for S-warfarin and R-warfarin. Both intra-day and inter-day precision and accuracy of S-warfarin and R-warfarin were well within acceptance criteria (15%). The mean absolute extraction recoveries of S-warfarin, R-warfarin, and IS from human plasma were all more than 60.00%. The validated method has been successfully applied to determine of S-warfarin and R-warfarin in human plasma. Then total warfarin was the sum of S-warfarin and R-warfarin. Software The population PK and PK-PD models were developed using a nonlinear mixed-effect modeling approached with the NONMEM TM (nonlinear mixed-effect modeling, version VII, level 2.0, ICON Development Solutions, Ellicott City, MD, United States). Goodness-of-fit diagnostic plots were prepared with R software (3.2.1, R-project. org). All models were run using the first-order conditional estimation method with interaction (FOCEI). Structural Model After inspection of the PK profiles, a one-compartment model with first-order absorption was adopted as the optimal base model for warfarin, R-warfarin, and S-warfarin. Structural PK model was fit to plasma concentrations, and typical values of absorption rate constant (Ka), apparent volume of distribution (V/F), and oral clearance (CL/F) were calculated (where F denotes bioavailability). In this study, each individual parameter was expressed approximately as a coefficient of variation to be a log-normal distribution with the mean of population parameters according to results of previous researches (Maria et al., 2003;Eunice et al., 2010;Mark et al., 2010;Juno et al., 2013). Where P ij was the PK Parameters j for ith individual, P TVj was mean of predicted population of PK Parameters j, η ij was a between-subject random variable distributed normally. PK Parameters j was just about Ka, V/F, or CL/F. The residual error model was assumed to be a mixed error model as following: Where C Obs was the observed plasma concentration, C Pred was the model prediction concentration. Both of multiplicative residual error (ε 1 ) and additive residual error (ε 2 ) were assumed as a normal distribution. Covariate Models These covariates were first explored graphically and each potential covariate individually added to the base model if graphical trends were shown. For the covariate models, stepwise of a forward inclusion step and a backward elimination step method was used. When a variable was considered for entering in the final model, it must reduce the objective function value (OFV) by more than 3.84 if p < 0.05 (5% significance level assuming a one degree of freedom, OFV > 3.84, df = 1; OFV > 5.99, df = 2; OFV > 7.81, df = 3). The variable that had the biggest impact on the OFV could enter first and subsequent variables added according to their impact on the OFV. The forward process described above was repeated again until no further covariates were incorporated into the model. Then, the backward elimination step was implementing. The variables were retained in the model if its removal caused an increase in OFV at least 6.63 if p < 0.05( OFV > 6.63, df = 1). The relative contribution of each covariate to the goodness of fit was evaluated by deleting it from the full model. With these restrictive criteria, only covariates showing statistically significant and clinically relevant contributions were kept in the population PK (PPK) model. In our research, the covariate of weight was described by allometric scaling equation: Where P i,j was the PPK parameters, P TV was a reference value of PPK parameters, WT was weight, WT was median of WT,θ 2 was effect value of WT to PPK parameter. For the categorical covariates, the following equation was adopted: Where COV was a categorical covariate which had m levels, θ l was adjusted value for P TV to P i,j . For categorical covariates, the linear models were employed: Where COV was continuous covariates, COV was median of continuous covariate, θ 3 was a coefficient for the effect of covariate to parameter. PK-PD Model Development The PK-PD model was developed only for S-warfarin, because S-warfarin is the main active ingredient, which is 3-5 times more potent than R-warfarin. According to the plot of relationship between INR and concentration of S-warfarin, the Emax model was selected as the PD model. For the covariate models, stepwise of forward and backward method was used as mentioned in PK model. Model Diagnostics The PK models for warfarin, R-warfarin, S-warfarin, and PK-PD model for S-warfarin was evaluated by the goodness of fit of these models using visual inspection of diagnostic scatter plots of the observed plasma concentrations (DV) versus mean population predicted plasma concentrations (PRED), DV versus individual predicted plasma concentrations (IPRED), conditional weighted residuals (CWRES) versus time, individual weighted residuals (IWRES) versus population predictions (PRED). Model Validation Visual predictive check (VPC) A visual predictive check (VPC) was performed to evaluate the prediction of PK models for warfarin, R-warfarin, S-warfarin, and PK-PD model for S-warfarin. The VPC were conducted by comparing 1000 datasets simulated from the final parameters with the observed plasma concentrations. The 95% predicted intervals (PIs) obtained from the simulation were superimposed and compared with the observations. Bootstrap A nonparametric bootstrap analysis was used to assess the stability of the parameter estimates and to confirm the robustness of the models. The 1000 bootstrap sample datasets were re-sampled from random sampling with replacement from the original data using individual as sampling unit. Next, population parameters of final PK and PK-PD models for each dataset were estimated. Then, the median and 95% confidence intervals (CI) were constructed by obtaining the 2.5th and 97.5th smallest values out of 1000 parameters estimated from bootstrap sample datasets. Comparing with the mean and 95% CI, each estimated parameter derived from the mean and its standard error of the final parameters. SAQ and Follow-Up The SAQ was applied to evaluate clinical symptoms of patients. The score of SAQ was collected after each period of clinical trial, and compared by paired t-test. In order to learn about more information about warfarin and CDDP administration, the follow-ups were done every 6 months and lasted for 2 years. The contents of follow-up included the length of taking the combination of warfarin and CDDP and some important and Patients Sixty-four patients were enrolled from four hospitals in Tianjin from November 2013 to January 2016. Fifty-nine patients completed the trial, among which 21 patients were female, and the mean age was 63 years (49-79 years). The demographic information of the patients was presented in Fixed Dose and Concentration of Warfarin In the two periods, the fixed doses of warfarin were 3.39 ± 1.04 mg and 3.36 ± 0.92 mg (P = 0.691), respectively. The steady-state concentration in the two periods was 1.2225 ± 0.5329 mg kg −1 and 1.1849 ± 0.4949 mg kg −1 (P = 0.587) for warfarin, 0.8337 ± 0.3597 mg kg −1 and 0.8128 ± 0.3132 mg kg −1 (P = 0.650) for R-warfarin, 0.3887 ± 0.2253 mg kg −1 and 0.3721 ± 0.2301 mg kg −1 (P = 0.535) for S-warfarin. So, the fixed dose and the steady-state concentration of warfarin were no statistically different between warfarin alone and warfarin plus CDDP. The results were shown in Table 2. Four Indicators of Blood Coagulation and INR Value The results were shown in Table 3. The four indicators of blood coagulation, prothrombin time (PT), activated partial thromboplatin time (APTT), thrombin time (TT), fibrinogen (FIB), and INR value between the two periods at the fixed warfarin dose had no statistical differences which was compared by paired t-test. Add.RE (sd) 0(FIX) Add.RE (sd) 0(FIX) Add.RE (sd) 0(FIX) * When CDDP was 0 represented the first period that patients was not administrated CDDP, and CDDP was 1 represented the second period that patients was administrated CDDP. Structural Model for PK Models The PK profile of warfarin was in accordance with onecompartment model (Eunice et al., 2010;Steven et al., 2011) or two-compartment model (Jiang et al., 2006;Hamberg et al., 2007) based on literatures, including warfarin, S-warfarin, and R-warfarin. One-compartment and two-compartment models were both investigated for warfarin, S-warfarin and R-warfarin in this study. By comparing OFVs, goodness of fit to the models, as well as rational of parameters, one-compartment was chosen as the optimal ones for initial structure models for warfarin, S-warfarin, and R-warfarin. Covariate Models for PK Models Once the base structural models were established, potentially significant covariates were evaluated as described. The OFVs of structure models for warfarin, R-warfarin, and S-warfarin were -622.414, -897.240, and -1448.94. When the covariate of weight was on CL, OFV were -623.958, -898.783, and -1449.40 for warfarin, R-warfarin, and S-warfarin, respectively. When the covariate of weight was on V, OFV were -622.423, -897.335, and -1448.94 for warfarin, R-warfarin and S-warfarin, respectively. Comparing with structural models, these OFVs were not more than 3.84, so weight was not the significantly covariates for warfarin, R-warfarin, and S-warfarin. For warfarin models, in the forward models, three covariates were brought in, PROC on CL, LU on CL, PROC on Ka, but in the backward models, LU on CL was eliminated. At last, the covariates of PROC on Cl and PROC on Ka were in the model. For R-warfarin models, PROC on Ka and PROC on CL were in the models, then PROC on CL was removed, so PROC on Ka was the significantly covariate. For S-warfarin model, the covariates PROC on Ka, CDDP on KA, CYP2C9 * 3 on CL, EPHX1 on V and VKORC1 on CL were in the forward models, then two covariates were rejected, PROC on Ka, CDDP on KA and CYP2C9 * 3 on CL were in the models as the significantly covariates. The detail information about the covariates for warfarin, R-warfarin, and S-warfarin were shown in Table 4. PK-PD Model On the basis of the final PK model, individual estimates of S-warfarin concentrations were predicted and used in the development of PD model. Graphical analyses of the INR observations versus time for S-warfarin demonstrated the Emax model may be more suitable for PD model. So direct Emax models, as well as BIOPH Emax models were investigated in our research. The parameters of the two models were all within a reasonable range, but OFV of BIOPH Emax model decreased 16.35 compared with direct Emax model. Moreover, there was a considerable time delay between INR response and drug concentration. Therefore, BIOPH Emax model was at last considered as the PD model of S-warfarin. Once the basic structural model was established, potentially significant covariates were evaluated. From the consequence of forward and backward method and the reasonable parameters, AST was finally brought on KE0 of PD model for S-warfarin. The detail information about covariates of PK-PD model for S-warfarin was shown in Table 5. Model Diagnostics The model diagnostic plots were shown in Figure 2. Figures 2A,B demonstrated that all the data points distributed uniformly in both sides of line y = x. Figures 2C,D demonstrated that CWRES and IWRES distributed uniformly in both sides of line y = 0, and the absolute value were less than 4. So, these models adequately described the plasma concentrations, suggesting good fitness of the PK models for warfarin, R-warfarin, S-warfarin, and PK-PD model for S-warfarin. Visual Predictive Check The VPC plots were shown in Figure 3. Most of the observations were in the 95% PIs, so the fit of the models were acceptable in terms of visual or statistical biases for the prediction. Bootstrap The estimated parameters and 95% values from all bootstrap runs for the PK models of warfarin, R-warfarin, and PK-PD model of S-warfarin were given in Table 6. The data indicated that the parameter estimated in PK models and PK-PD model had little bias and the models were fairly robust. The Results of SAQ and Follow-Up The results of SAQ were shown in Table 3. The score of SAQ during trail were 19.71 ± 5.05 for the first period and PD Add.RE (SD) 0 (FIX) * When CDDP was 0 represented the first period that patients was not administrated CDDP, and CDDP was 1 represented the second period that patients was administrated CDDP. 21.02 ± 5.07 for the second period. There was significant difference between two periods (P = 0.002). During 2 years of follow-up, the mean length of taking CDDP is 0.96 ± 0.80 years and 1.70 ± 0.83 years for warfarin. Severe arrhythmia occurred in one patient, revascularization in two patients, death in one patient, and 16 patients were hospitalized due to other conditions. The incidence of severe arrhythmia, revascularization, death, and hospitalization were 2.44% (1/41), 4.88% (2/41), 2.44% (1/41), and 39.02% (16/41). DISCUSSION In this study a total of 404 blood samples, instead of 472 as required by protocol (eight for each patient), were collected for warfarin assay. There were 600 INR values obtained in the study. Being serious in nature of the heart disease, the compliance of the enrolled patients in the study was relatively low. With the limited number of patients enrolled in this study, we use the principle of PPK algorithm with more detail information collected from the participants, as described in the Guidance for Industry PPK published by FDA which states "Since patients are studied in more detail in this design, the design requires fewer subjects, and the relationship of trough levels to patient characteristics can be evaluated with more precision". Similarly, in some other PPK-PPD studies (Hamberg et al., 2007;Parker et al., 2015;Agarwal et al., 2016;Pier et al., 2017), these numbers of subjects enrolled were comparable to our study. These CHD patients with AF had been taking warfarin for a long period of time to decrease the risk of thromboembolic complications. The underlying conditions included hypertension (28.8%), diabetes (18.7%), and cerebral ischemic stroke (16.9%) in the study. Various medications were concomitantly used in the patients due to its complicated property of the disease. More frequently used drugs included β-blockers (39.0%), nitric esters (33.9%), statins (22.0%), diuretics (22.0%), cardiac glycosides (18.7%), calcium channel blockers (15.3%), and antidiabetic drugs (15.3%). Identification of the effect of CDDP on profiles of PK and PD of warfarin were established on a self-control design, maintaining all concomitant medications except warfarin from beginning to end of trial. The INR value, fixed warfarin doses and trough concentration of plasma warfarin had not significantly difference (P > 0.05) with or without CDDP. That suggested there is no drug-drug interaction between CDDP and warfarin in human, and CDDP did not affect anticoagulant mechanism of warfarin because CDDP did not interfere metabolic process of warfarin in human. The PPK models for warfarin, S-warfarin and R-warfarin and PPD model for S-warfarin were developed with assessing patient demographics, genetic polymorphisms and CDDP as covariates. The PK behavior of warfarin, S-warfarin and R-warfarin was in accordance with one-compartment model or two-compartment model based on literatures (Jiang et al., 2006;Hamberg et al., 2007;Eunice et al., 2010;Steven et al., 2011). The Onecompartment models were more optimal and reasonable for warfarin, either S-warfarin or R-warfarin in our study. The consequence of CYP2C9 * 3 on CL of S-warfarin was identical to other researches (Chanan et al., 2016;Darcy et al., 2017), but the age, gender, weight had no significant effects on S-warfarin or R-warfarin which were inconsistent with some research (Hamberg et al., 2007;John et al., 2012;Lin R.F. et al., 2015). The gene of CYP4F2, PROC, VKOR, CYP2C9 * 3, and EPHX1 were investigated in this research. There was no precise conclusion about the effects of gene subtypes on PK and PD characteristics of warfarin due to limited distribution rate of each individual subtype, which was consistent with other literatures (Özer et al., 2011;Radka et al., 2015). The change of the fixed dose of warfarin in EPHXI gene subtype A/A is higher than the other gene types. Because only two patients with EPHX1 gene subtype A/A were enrolled, it is difficult to make the conclusion that CDDP affect the PK and PD characteristics of warfarin on this kind of patient. The study result also suggested by PPK model and PPD model that there may be no influence of CDDP on PK and PD of warfarin in patients, although CDDP was as a covariate on Ka of S-warfarin. There had a great variation of Ka with a higher RSE (43.7%, 129.4%, 74.6% for PK models of warfarin, R-warfarin, and S-warfarin) meaning an incredible consequence about absorption. The bioavailability of warfarin is more than 95% (Mark et al., 2010;Juno et al., 2013), some reports had applied 100% bioavailability when developing models (Maria et al., 2003;Hamberg et al., 2007;Steven et al., 2011). Moreover, there were few reports to evaluate the absorption of warfarin due to its high bioavailability. The SAQ was applied to evaluate efficacy effect of the co-treatment of warfarin and CDDP in the patients of CHD with AF. It's showed that there was significant difference in the SAQ score with or without CDDP. It may indicate that CDDP can improve the life quality of CHD with AF patients when both INR and dose of warfarin are stable. During 2 years followup, many patients still took the combination for a long time, and there were no report about bleeding. So, the combination of CDDP with warfarin might relieve clinical symptoms and provide benefits for patients with CHD and AF. However, this was not a randomized clinical trial, some researches would be needed to further demonstrate the clinical efficacy. CONCLUSION In summary, robust and stable PK-PD models have been successfully developed for evaluating the effect of CDDP on the PK and PD of warfarin. The results indicated that CDDP did not influence the INR stability and PK characteristic of warfarin when warfarin was administrated simultaneously with CDDP in most CHDs patients. Moreover, The SAQ and follow-up results showed the CDDP combined with warfarin might provide benefit in clinical practice for patients. This study would provide some useful information of the combined regimen of CDDP and warfarin for the treatment of CHDs with AF, but the result in Chinese genetic subtypes of EPHX1 and the clinical efficacy study need to be confirmed further. AUTHOR CONTRIBUTIONS CLv and YH wrote the manuscript. CLiu, YH, XG, and HS designed the research. ZY, LS, XD, JS, YL, and KY performed research. CLv, JL, ZL, and RW analyzed data.	2017-11-21T18:08:23.098Z	2017-11-21T00:00:00.000	{ "year": 2017, "sha1": "01873f8495d69001ea862bc5e995ae4e9d05fedb", "oa_license": "CCBY", "oa_url": "https://www.frontiersin.org/articles/10.3389/fphar.2017.00826/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "01873f8495d69001ea862bc5e995ae4e9d05fedb", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
71934051	pes2o/s2orc	v3-fos-license	Costodorsalis – an Additional Slip of Pectoralis Major Muscle-a Case Report Occurrence of variant muscular slips from pectoralis major muscle is rare. In this report, we present a rare case of aberrant muscular slip associated with the pectoralis major muscle which we call costodorsalis. This muscular slip originated f rom the 6th rib near the costochondral junction and ran along the lower border of pectoralis major muscle. It crossed the axilla from media l to lateral side and merged with the latissimus dorsi muscle. This type of origin and insertion is unique and has not been reported earlier . The knowledge of this muscle variation may be of special importance to the anesthesiologists, physiotherapists and plastic surgeons . INTRODUCTION The pectoralis major muscle takes its origin from the clavicle, sternum, 2nd to 6th ribs and the respective costal cartilages and the aponeurosis of the external oblique muscle of the abdomen.It is inserted to the lateral lip of the intertubercular sulcus of the humerus.It is supplied by the medial and lateral pectoral nerves.The muscle forms the anterior wall of the axilla and helps in adduction, flexion and medial rotation of the arm at the shoulder joint.The variants of pectoralis major muscle usually do not cause any symptoms but are of academic interest.They may present as a surgical problem, if they cause any symptoms.The chondro-epitrochlearis, costo-epitrochlearis or costohumeralis are some of the rare muscular variations that may arise from pectoralis major muscle crossing the axilla and inserting to the medial inter muscular septum or medial epichondyles of the humerus.In the present case, we found a variable muscular slip which is a very rare muscular anomaly of the pectoral region. CASE REPORT During the routine dissection classes for the medical undergraduate students at the Melaka Manipal Medical College (Manipal Campus), we observed a variation in the right upper limb of a 40-45 year old male cadaver.The skin, superficial and deep fascia were removed to expose the pectoral region, axilla and arm.The pectoralis muscle and its variant muscle slip were cleaned and photographed.The variant slip arose from the outer surface of the 6 th costal cartilage (Fig. 1).This slip ran parallel to the lower border o pectoralis major muscle and crossed the axilla and merged Department of Anatomy, Melaka Manipal Medical College (Manipal Campus).International Centre for Health Sciences.Manipal University, India.with the aponeurosis of the latissimus dorsi to have same insertion as the latter (Fig. 2).It had the same innervation as the pectoralis major muscle.The variation noted was unilateral and there was no evidence of any other anomaly or pathology in the body.associated with axillary arch muscle in 7 to 13% of the population.Jaijesh (2005) reported the presence of unilateral chondro-epitrochlearis muscle with absence of normal twisted insertion of the pectoralis major muscle.Samuel & Vollala (2008) also reported such a muscle and the blood and nerve supply to this variant muscle.The arterial supply to the chondro-epitrochlearis muscle was from the lateral thoracic artery and the nerve supply was through a branch of the medial pectoral nerve.Loukas et al. (2005) presented a rare case of an accessory muscular slip originating from the pectoralis major and inserting to the medial epicondyle of the humerus and medial brachial intermuscular septum.They proposed a new nomenclature of this variant slip thoraco-epicondylaris.Lama et al. (2010) reported the presence of chondro humeralis and axillary arch of Langer; a rare combination of variant muscles with unique insertion.Bilateral epitrochlearis muscle associated with the bilateral absence of the axillary arch and absence of the normal twisted insertion of pectoralis major muscle has been reported by Flaherty et al. (1999).Presence of a pectoralis tertius muscle was reported by del Sol & Vásquez (2009). In the present case, we found a variable muscular slip which can be called 'costodorsalis' since it was arising from the rib and merging with the latissimus dorsi.Its unique insertion pattern has not been reported earlier.This anomalous slip crossed the base of axilla from medial to lateral side.It might surprise the surgeons doing any surgery of the axilla.The anomaly is of specific importance because of its potential to cause cosmetic defects and to restrict abduction of the arm.Thus, it may be of particular interest to surgeons, physiotherapists and plastic surgeons.RESUMEN: Es poco frecuente la aparición de variaciones de un fascículo muscular desde el músculo pectoral mayor .En este trabajo, presentamos el caso de un fascículo muscular aberrante asociado con el músculo pectoral mayor que denominamos costodorsal.Este fascículo muscular se originó en la 6ª costilla cerca de la unión costocondral y corrió a lo largo del margen inferior del músculo pectoral mayor.Cruzó la axila de medial a lateral y se fusionó con el músculo latísimo del dorso.Este tipo de origen y la inserción es único y no se ha informado anteriormente.El conocimiento de esta variación muscular puede ser de especial importancia para los anestesistas, fisioterapeutas y cirujanos plásticos. DISCUSSION Many muscular variations in relation to the pectoralis major have been reported before, but it remains yet important to continue describing the rare variations due to their clinical implications.These variations are important in defining the anatomical features in relation to the clinical diagnosis and surgical procedures.The pectoralis major muscle may be looked upon as composed of four portions -a clavicular, a sternal, a costal and an abdominal, the last being the portion that arises from the aponeurosis of the external oblique.These parts vary in the extent of their attachments and in the degree of separation that they present.The abdominal portion may extend to the umbilicus.On the sternum the muscles of the two sides may decussate across the middle line.The sternocostal portions of the muscle may be deficient or missing the clavicular, but in rare cases the entire muscle may be absent.The clavicular portion of the muscle may be fused with the deltoid while the sternocostal portion may extend laterally to the latissimus dorsi.Rarely a slip may run from the pectoralis major to the biceps, the pectoralis minor, coracoid process, joint capsule or brachial fascia (Anson, 1966).Rao et al. (2009) have reported a rare costohumeralis muscle.It is an accessory slip of pectoralis major originating from the sixth rib near costochondral junction and running along the lower border of the pectoralis major muscle and getting inserted onto the medial epicondyle of the humerus.Chiba et al. (1983) have reported the presence of a muscle called chondro-epitrochlearis Fig. 1 . Fig. 1.Dissection of the pectoral region and axilla showing the additional slip associated with the pectoralis major muscle.	2017-10-20T06:35:40.952Z	2011-01-01T00:00:00.000	{ "year": 2011, "sha1": "6f57fc84f6f453999a555c56b8182a87e418bf0b", "oa_license": "CCBYNC", "oa_url": "http://www.scielo.cl/pdf/ijmorphol/v29n2/art17.pdf", "oa_status": "GOLD", "pdf_src": "Anansi", "pdf_hash": "6f57fc84f6f453999a555c56b8182a87e418bf0b", "s2fieldsofstudy": [], "extfieldsofstudy": [ "Philosophy" ] }
221008404	pes2o/s2orc	v3-fos-license	Is hepatocellular carcinoma surveillance in high-risk populations effective? Several professional societies recommend hepatocellular carcinoma (HCC) surveillance in high-risk patients including patients with cirrhosis from any etiology and subsets of noncirrhotic chronic hepatitis B virus infection. The efficacy of HCC surveillance to increase early detection and improve survival has been demonstrated in a large randomized controlled trial among hepatitis B virus patients and several cohort studies among those with cirrhosis. However, the effectiveness on HCC surveillance, when applied in clinical practice, is lower due to low utilization of HCC surveillance among at-risk patients, poorer test performance given operator dependency and differences in patient characteristics, and downstream process failures such as treatment delays. Interventions to increase surveillance utilization and improve surveillance test performance should improve surveillance effectiveness in the future. alpha-fetoprotein (AFP). This strategy has been shown to increase early detection and improve survival in a large randomized controlled trial (RCT) in HBV patients and several cohort studies in patients with cirrhosis [12,13]. In this article, we discuss the efficacy and effectiveness of HCC surveillance in patients with cirrhosis, including factors that may contribute to a gap between efficacy and effectiveness. Efficacy versus effectiveness Efficacy studies investigate the benefits and harms of an intervention under highly controlled conditions, evaluating how well the intervention can work [14]. Efficacy studies have clear methodologic advantages and have high internal validity; however, they often highly select patients (e.g., minimal comorbidity and adherent), providers (e.g., highly knowledgeable and skilled) and tools (e.g., most recent ultrasound technology), and use more intensive protocols than routine practice (e.g., patient navigation to complete screening and follow-up). The ideal study design to evaluate efficacy is a placebo-controlled RCT because it minimizes confounding and other sources of bias, although this can also be done in prospective cohort studies as well. In contrast, effectiveness studies examine interventions under circumstances that more closely approach realworld practice and evaluate how well an intervention does work in clinical practice [14]. Effectiveness studies include more heterogeneous patient and provider populations and less-standardized treatment protocols as part of routine clinical practice. Although effectiveness studies sacrifice internal validity, they have higher external validity than efficacy studies. A gap between efficacy and effectiveness can be explained by ineffective intervention, poor implementation, lack of provider recommendation or lack of patient adherence [15]. Efficacy of HCC surveillance The efficacy of HCC surveillance has been evaluated in a large RCT among patients with chronic HBV and prospective cohort studies in patients with cirrhosis [12,13]. The RCT from China used block randomization in >18,000 patients with HBV infection who were allocated to surveillance or no surveillance. Patients randomized to surveillance had significantly higher proportion of early-stage HCC detection (60.5 vs 0%) and curative treatment receipt (46.5 vs 7.5%), resulting in significantly reduced HCC-related mortality (83.2 vs 131.5 per 100,000 persons; hazard ratio (HR): 0.63; 95% CI: 0.41-0.98). Although this study provides level I data supporting HCC surveillance, there have been concerns that the analysis failed to adhere to intention-to-treat principles and did not account from block randomization [16]. These data cannot be extrapolated to patients with cirrhosis given increased nodularity and heterogeneity potentially impacting ultrasound sensitivity and increased risk of liver-related mortality. When a large randomized trial was attempted in patients with cirrhosis, it had to be terminated given poor enrollment and authors concluded an RCT may not be feasible in light of patient and provider preferences [17]. However, several cohort studies provide level II data supporting HCC surveillance in patients with cirrhosis [18][19][20]. A systematic review of 47 studies demonstrated that surveillance was associated with increased tumor detection odds ratio (OR): 2.08; 95% CI: 1.80-2.37), increased curative treatment receipt (OR: 2.24; 95% CI: 1.99-2.52) and improved 3-year survival (OR: 1.90; 95% CI: 1.67-2.17) [13]. Surveillance remained associated with improved survival in the subset of studies adjusting for lead time bias and length time bias, although studies were still prone to other biases inherent in cohort studies such as selection bias and residual confounding [18]. Overall, we believe these data provide strong, albeit imperfect, data supporting HCC surveillance as efficacious in patients with chronic HBV or cirrhosis. Effectiveness of HCC surveillance The potential gap between surveillance efficacy and effectiveness was highlighted by a recent case-control study from the Veterans Affairs (VA) Health System, which failed to find an association between surveillance receipt and improved survival [21]. The authors found surveillance receipt did not significantly differ between patients who died of HCC and a matched group of cirrhosis patients who had not died of HCC. As above, it is unclear if these results are related to surveillance being an ineffective strategy when implemented in clinical practice or if this were related to poor implementation with the VA health system. Most notably, a decision analysis found surveillance utilization and test performance are two of the most important determinants of HCC surveillance effectiveness and ability to afford a survival benefit [22]. Surveillance utilization Several large studies have demonstrated HCC surveillance is underused in clinical practice. The first large studies to demonstrate underuse of surveillance were large population-based studies from the Surveillance Epidemiology and End Results-Medicare and national VA databases, with both studies showing less than 20% of patients had received an ultrasound or AFP in 2 of the 3 years prior to HCC diagnosis [23,24]. Since that time, several cohort studies have similarly shown HCC surveillance is underused, with a recent meta-analysis finding a pooled surveillance utilization of 24.0% (95% CI: 18.4-30.1%) [25]. In subgroup analyses, the highest surveillance receipt was reported in studies including patients from subspecialty gastroenterology clinics and the lowest surveillance receipt in studies using population-based cohorts (73.7 vs 8.8%). Lower surveillance receipt was also observed among patients with nonviral etiologies such as alcohol-related and metabolic-associated fatty liver disease. Finally, surveillance utilization was significantly lower in US-based studies compared with those from Europe or Asia (17.8 vs 43.2 and 34.6%). These figures pale in comparison with screening utilization observed in other evidence-based cancer screening programs, such as those for colorectal cancer, breast cancer and cervical cancer, which typically report surveillance receipt exceeding 60% among at-risk patients [26]. HCC surveillance implementation can be broken down into several discrete steps: patients must be engaged in healthcare and have a clinic visit, providers must accurately identify at-risk patients, providers must order appropriate surveillance tests; surveillance tests must be scheduled and the patient must adhere with the surveillance recommendations. There appears to be breakdowns at each of these steps in clinical practice, although the most common issues are lack of engagement in healthcare/clinic visits and lack of provider recommendation for surveillance testing in patients with known cirrhosis [27]. Patients' nonadherence to surveillance testing can occur but appears to be relatively rare. Survey studies have been conducted among both providers and patients to gain further insight into potential barriers for surveillance completion. Primary care providers appear to believe HCC surveillance is efficacious for early tumor detection and reducing mortality; however, they had some important misconceptions about surveillance logistics including some believing physical examination or liver enzymes levels were useful for detecting HCC [28,29]. Providers also reported several barriers to performing surveillance including not being up-to-date with surveillance recommendations, considering surveillance outside of scope of primary care, and limited time in clinic with competing clinical concerns. Patients demonstrate high levels of HCC-related knowledge but similarly believed eating a healthy diet could preclude the need for HCC surveillance and that surveillance was not needed in the setting of a normal physical exam and labs [30,31]. Patients expressed worry about developing HCC and desire to complete HCC surveillance but reported barriers including difficulty with scheduling, costs of the tests and transportation difficulties. Surveillance receipt was significantly lower in patients who reported barriers to HCC surveillance. Several studies have evaluated interventions to increase HCC surveillance utilization. Studies have reported significant increases with inreach efforts including primary care provider education or an electronic medical record reminder. For example, Del and colleagues reported an increase in surveillance-detected HCC after primary care provider education (55.3 vs 34.8%), compared with no significant change among a control group without education (39.2 vs 25.9%) [32]. In the largest study evaluating an electronic medical record reminder, Beste and colleagues reported increased adequate HCC surveillance (≥2 imaging studies within 18 months) from 18.2 to 27.6%, whereas control sites without the intervention had no appreciable change in surveillance (16.1 vs 17.5%) [33]. Recently, a large pragmatic RCT has evaluated a population health outreach strategy to increase HCC surveillance. One-time screening within 6 months was significantly higher in the mailed outreach arm than usual care arm (44.5 vs 24.3%); the addition of patient navigation did not significantly increase one-time screening completion (47.2%) compared with outreach alone [34]. In a follow-up study, the team found continued benefits of outreach and navigation over longer periods of time; semi-annual surveillance over an 18-month period was performed in 23.3% of outreach/navigation patients, 17.8% of outreach-alone patients and 7.3% of usual care patients (p < 0.001 for both vs usual care and p = 0.02 for outreach ± navigation) [35]. Although the trial demonstrated a benefit of these interventions, surveillance receipt among those who received mailed outreach and patient navigation was disappointingly low. Some of this may relate to the difficult-to-reach patient population, although these results may also imply the need for more intensive interventions in the future. Overall, these studies provide a roadmap to start addressing surveillance barriers, increase surveillance utilization and thereby improve surveillance effectiveness. Surveillance test effectiveness Abdominal ultrasound has been the cornerstone of HCC surveillance for many years because it is readily available, noninvasive, inexpensive and safe with no risk of radiation or contrast exposure. Ultrasound has a high sensitivity of 84% (95% CI: 76-92%) for detecting HCC at any stage; however, it only has a sensitivity of 47% (95% CI: 33-61%) for early-stage HCC detection [36]. Further, its effectiveness can be affected by operator expertise and patient-level factors such as obesity and liver disease severity, leading to wide center-to-center and patient-to-patient variation in sensitivity [37,38]. A retrospective cohort study found 20% of ultrasound exams were of suboptimal quality for HCC surveillance, with this being significantly more likely in obese patients and those with NASH cirrhosis [39]. The lower sensitivity of ultrasound in patients with obesity and nonviral cirrhosis is concerning, given an increasing proportion of HCC related to NASH [40]. Surveillance value not only depends on potential benefits, in other words, sensitivity for early detection and improved survival, but also potential harms. Screening-related harms can be categorized as physical, financial or psychological. Data on harms of HCC surveillance have been limited to date, with only two studies quantifying physical harms and no studies examining financial or psychological harms. Both studies examining physical harms found 20-30% of cirrhosis patients experience a false-positive or indeterminate surveillance result that prompts diagnostic evaluation with computed tomography (CT) or MRI [41,42]. Moderate-severe harm, defined as repeated imaging or invasive evaluation such as biopsy, was observed in 10%. While some harms were related to suboptimal test specificity, there were patients who experienced harm related to nonguideline concordant management of indeterminate results, for example, subcentimeter liver lesion [41]. The latter category may increase the gap between surveillance value in efficacy and effectiveness settings. There has been an increasing use of surveillance CT or MRI in clinical practice; however, there are limited data supporting routine use of cross-sectional imaging for this indication. A small single-center RCT comparing CT and ultrasound-based surveillance failed to find a significant difference in early detection (62.5 vs 55.5%; p = 0.93) or HCC-related mortality (8.8 vs 6.0%; p = 0.46) despite higher costs in the CT arm [43]. In a prospective cohort study comparing MRI-and ultrasound-based surveillance among 407 cirrhosis patients, Kim and colleagues found MRI-based surveillance had higher sensitivity for early HCC detection than ultrasound (83.7 vs 25.6%) [44]. However, data about MRI performance in non-HBV patients and its cost-effectiveness are needed prior to routine use of MRI for surveillance. Studies would also likely need to evaluate other potential concerns such as physical harms (radiation and contrast exposure), costs and limited radiologic capacity. A decision analysis suggested MRI may be particularly useful among patients at the highest risk of HCC; however, accurate risk stratification models are not yet available for routine use in practice [45]. Although MRI-based surveillance may increase surveillance effectiveness in the future, further data evaluating these novel imaging techniques in larger cohorts are still needed. In parallel, there has been an increasing interest in serum biomarkers that may improve sensitivity for early HCC detection. The potential for novel biomarkers to improve surveillance effectiveness is supported when comparing outcomes in the Western world to those in countries like Japan, where biomarkers are incorporated into the routine surveillance recommendations and most patients are found at an early stage [46]. AFP remains the best studied biomarker but has poor sensitivity for HCC when used alone [47]. A meta-analysis found use of combination ultrasound and AFP improved early HCC detection compared with ultrasound alone, with sensitivities of 63%, (95% CI: 48-75%) and 45%, (95% CI: 30-62%), respectively [36]. Although this was associated with decreased specificity (84 vs 92%), the diagnostic odds ratio of the two in combination was higher than that of ultrasound alone. Recent data suggest using longitudinal changes in AFP values, rather than a single threshold, may also more accurately identify patients with early-stage HCC [48,49]. Other emerging biomarkers have similarly failed to achieve high sensitivity for early detection when used alone, so there is an increasing interest in combinations of biomarkers. For example, GALAD, which includes gender, age, AFP-L3%, AFP and des-gamma-carboxy prothrombin (DCP), has been evaluated in a large case-control study with 6834 patients (2430 HCC and 4404 chronic liver disease) and achieved sensitivities ranging from 60 to 80% for early HCC detection [50,51]. A methylated DNA marker panel was also recently shown to have promising performance in a large case-control study with sensitivity exceeding 70% for early HCC detection with specificity of approximately 90% [52]. Although these data for emerging biomarkers are promising, they still require validation in large Phase III cohort biomarker studies [53]. Downstream process failures Surveillance and early detection is part of the larger screening continuum and is dependent on timely diagnostic evaluation and appropriate treatment to translate into a survival benefit [15,54]. Some studies have suggested diagnostic and therapeutic delays may be associated with worse outcomes including stage migration, lower receipt of curative treatment and worse survival for other cancers including colorectal and breast cancer [55][56][57][58]. Although few studies have examined these issues among HCC patients, existing literature suggests failures in these downstream processes. In a retrospective single-center study, nearly one in five patients experienced diagnostic delays, defined as time from presentation to diagnosis exceeding 3 months [59]. Diagnostic delays may be related to providers missing or misinterpreting abnormal screening results, inefficient system-level workflows and scheduling processes, patients missing appointments or inaccurate diagnostic tests with suboptimal sensitivity [60]. Given an expected tumor doubling time of approximately 4-6 months, diagnostic delays of >3 months could allow for substantial tumor growth; however, studies have not yet demonstrated stage migration or worse survival with diagnostic delays in HCC [61,62]. Similarly, a single-center retrospective study also demonstrated a median time-to-treatment of 1.7 months, with nearly a third of patients experiencing therapeutic delays, defined as time from diagnosis to treatment exceeding 3 months [63]. In multivariable analysis, treatment delays were associated with significantly worse survival. Beyond therapeutic delays, several studies have highlighted underuse of HCC therapy in treatment-eligible patients, including low receipt of curative treatment among patients diagnosed at an early stage [64]. Treatment patterns are also notable for significant disparities with lower curative treatment receipt among racial/ethnic minorities and those of low socio-economic status. Multidisciplinary care and being seen in high-volume centers have both been shown to improve timely treatment and appropriate treatment receipt, which result in improved stage-by-stage survival [65][66][67][68]. Conclusion There is level I data in HBV patients and level II data in patients with cirrhosis suggesting HCC surveillance is efficacious for both early HCC detection and improving survival. However, effectiveness of HCC surveillance may be substantially lower given low utilization, poor test performance given operator dependency and differences in patient characteristics, and downstream process failures such as treatment delays. Future perspective Ongoing interventions to increase surveillance utilization and evaluate novel imaging-and blood-based surveillance strategies should improve surveillance effectiveness in the future. We believe blood-based biomarkers have the greatest potential to address both surveillance underuse and test effectiveness over the next 5-10 years. By doing so, this would significantly increase the proportion of tumors detected at an early stage and thereby reduce HCC-related mortality. involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript. Open access This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/	2020-07-30T02:05:11.269Z	2020-07-24T00:00:00.000	{ "year": 2020, "sha1": "e4f004fd1ed02a32e5341c453f9978b7f9c49551", "oa_license": "CCBYNCND", "oa_url": "https://doi.org/10.2217/hep-2020-0012", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "184bb8776181435154c7293425877789b87f139b", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
248947843	pes2o/s2orc	v3-fos-license	Club cell protein (CC)16 as potential lung injury marker in a porcine 72 h polytrauma model Background Polytrauma and respiratory tract damage after thoracic trauma cause about 25% of mortality among severely injured patients. Thoracic trauma can lead to the development of severe lung complications such as acute respiratory distress syndrome, and is, therefore, of great interest for monitoring in intensive care units (ICU). In recent years, club cell protein (CC)16 with its antioxidant properties has proven to be a potential outcome-related marker. In this study, we evaluated whether CC16 constitutes as a marker of lung damage in a porcine polytrauma model. Methods In a 72 h ICU polytrauma pig model (thoracic trauma, tibial fracture, hemorrhagic shock, liver laceration), blood plasma samples (0, 3, 9, 24, 48, 72 h), BAL samples (72 h) and lung tissue (72 h) were collected. The trauma group (PT) was compared to a sham group. CC16 as a possible biomarker for lung injury in this model, and IL-8 concentrations as known indicator for ongoing inflammation during trauma were determined by ELISA. Histological analysis of ZO-1 and determination of total protein content were used to show barrier disruption and edema formation in lung tissue from the trauma group. Results Systemic CC16 levels were significantly increased early after polytrauma compared vs. sham. After 72 h, CC16 concentration was significantly increased in lung tissue as well as in BAL in PT vs. sham. Similarly, IL-8 and total protein content in BAL were significantly increased in PT vs. sham. Evaluation of ZO-1 staining showed significantly lower signal intensity for polytrauma. Conclusion The data confirm for the first time in a larger animal polytrauma model that lung damage was indicated by systemic and/or local CC16 response. Thus, early plasma and late BAL CC16 levels might be suitable to be used as markers of lung injury in this polytrauma model. Introduction Despite achievements to reduce injury-related mortality and morbidity in the last decades, the number of polytraumatized patients admitted to hospitals remains high with approximately 4.8 million human deaths worldwide caused by traumatic injuries every year [1]. In polytraumatized patients blunt thoracic trauma is common, and around 20-25% of deaths in severely injured patients treated in emergency departments are caused by chest injuries [2]. Furthermore, thoracic trauma has been shown to be associated with increased posttraumatic inflammation compared to polytrauma patients who did not undergo thoracic trauma, increased rates of pneumonia and Acute Respiratory Distress Syndrome (ARDS) and lower survival rates of patients [3][4][5]. It is of utmost importance for the medical community Johannes Greven and Jan Tilmann Vollrath contributed equally. to detect changes in the lungs as early as possible to prevent acute lung injury (ALI). Club cell protein (CC)16 is a 15.8 kDa homodimeric protein and the major protein secreted in airways by the non-ciliated bronchiolar Club cells [6,7]. It occurs in very high concentrations in the epithelial lining where it appears to play an antioxidant and anti-inflammatory role [7]. Its role as a potential biomarker to detect pulmonary damage in polytraumatized patients or to identify polytraumatized patients at risk for secondary pulmonary complications has been extensively studied during the last decades. Wutzler et al. reported that in patients who have sustained multiple injuries, the early elevation of CC16 serum levels can be used to accurately identify patients with versus those without lung injury [8]. Furthermore, serum CC16-levels have been shown to correlate with the volume of lung contusion in polytraumatized patients [8], and were proposed to serve as a biomarker to monitor the progression of pulmonary contusion [9]. Negrin et al. investigated CC16 levels in polytraumatized patients with severe thoracic injury and reported that CC16 levels exceeding 30.51 ng/ml on day two after trauma may allow a firmer diagnosis for the development of pneumonia [10]. Another study reported a second peak in CC16 levels in patients developing pneumonia after polytrauma with severe thoracic trauma [11]. Xu et al. investigated ex vivo in vitro association of CC16 and polymorphonuclear leukocytes (PMNL) in polytraumatized patients with or without pneumonia and observed significant effects of CC16 on PMNL in patients with pneumonia [12]. The authors hypothesized that CC16 might reduce the migratory capacity of PMNL, and thus, modulate their function in patients with respiratory complications after trauma [12]. Several factors affecting circulating CC16 have been identified as, e.g. on the one hand chronic exposure to tobacco smoke has been reported to cause a decrease in CC16 levels [13], while on the other hand acute exposure of firefighters to fire smoke has been reported to increase CC16 serum levels [14]. Following the hypothesis that CC16 might indicate lung epithelial injury during polytrauma in a large animal model with high translational aspects, we investigated the biomarker character of CC16 in a standardized porcine trauma model consisting of lung contusion, liver laceration, tibial fracture, and hemorrhagic shock followed by fluid resuscitation and fracture fixation with external fixator. Animal care All experiments were performed in accordance with the German legislation governing animal studies, following the "Principles of Laboratory Animal Care" [15]. Official permission was granted by the North Rhine-Westphalia State Office for Nature, the Environment and Consumer Protection (Landesamt für Natur-, Umwelt-and Verbraucherschutz Nordrhein-Westfalen, Recklinghausen, Germany, AZ 81.02.04.2018.A113). Male German landrace pigs were housed with a 12 h day/night rhythm at about 22 °C and allowed to acclimatize to their surroundings for a minimum of 7 days before the surgery. This experiment was part of a larger study of pigs that sustained polytrauma (PT group) and non-injured controls (sham group) (German Landrace, Sus scrofa). The pigs had a 35 ± 5 kg body weight (BW). For the current study, eight pigs from each group (sham and PT group) were included in the project. All animals underwent clinical examination and vaccination (21 days before experiments; circovirus and mycoplasma) by a veterinarian before the experiments started to exclude pre-infections. This study presents partial results obtained from a large animal porcine polytrauma model [16]. General instrumentation and anesthesia The experimental setup was established at the Department of Trauma Surgery, RWTH, Aachen in 2016 and can be found elsewhere [17]. In brief, anesthetized (propofol 2%/2 mg/ kg BW/h i.v.) and analgized (fentanyl 40-90 μg/kg BW/h i.v.; additional midazolam 0.1-0.25 mg/kg BW/h to prevent shivering) male pigs were intubated, ventilated, and continuously monitored (electrocardiographic recording/pulse oximetry). All pigs were further monitored by arterial pulse contour cardiac output (PiCCO, Pulsion Medical Systems, Germany), and catheters were placed into the right femoral artery, the left femoral vein [two-lumen hemodialysis catheter (Arrow International, Germany)], the right jugular vein [central venous catheter (Arrow International, Germany)], and the bladder (suprapubic catheter). Instrumentation was applied under sterile conditions. Induction of trauma For the induction of blunt thoracic trauma, a pair of steel and lead panels (each 5 × 10 cm in the surface, 0.8 and 1.0 cm thickness, respectively) was applied to the left dorsal lower chest. Blunt lung contusion was induced by a bolt gun machine (Blitz-Kerner, turbocut JOBB GmbH, Germany). Cattle-killing cartridges (green, 9 × 17; Dyna-mitNobel AG, Troisdorf, Germany) were used. The bolt shot onto this panel simulated blunt lung contusion as previously described [17,18]. O 2 was defined at 21% during the trauma period, hence simulating the ambient air. An additional laparotomy was done to approach the liver to mirror a penetration injury. The midlobe of the liver was cut crosswise (3 cm) to half the liver thickness in depth, and uncontrolled bleeding was allowed for 30 s. After that, liver packing and suturing of the laparotomy was performed. Furthermore, a tibia fracture was carried out using a cattle gun. Pressure-controlled and volume-limited hemorrhagic shock was induced by withdrawing blood until a mean arterial pressure (MAP) of 40 ± 5 mm Hg was reached. Here, a maximum of 45% of total blood volume was drawn from the left femoral vein. The shed blood was kept in blood bags for reinfusion purposes. Hemorrhagic shock was maintained for 90 min. After this period, the animals were resuscitated in accordance with established trauma guidelines (ATLS ® , AWMF-S3 guideline on Treatment of Patients with Severe and Multiple Injuries ® ), and the tibia fracture was fixed by external fixation [19,20]. Resuscitation phase In addition to pre-warmed crystalloids (SterofundinISO and pediatric electrolyte solution 2-4 ml/kg BW/h), the animals received their previously withdrawn blood to restore hemostasis. The animals were mechanically ventilated and monitored on a special intensive care unit (ICU) for 72 h post-injury according to well-established ICU treatment guidelines (AWMF-S3 guideline on Treatment of Patients with Severe and Multiple Injuries ® ). The hemodynamics for this model have been described before [17]. After 72 h animals were euthanized using potassium chloride followed by harvest of the organs. Sham animals Sham animals were not subjected to injury, laparotomy or hemorrhage but were identically instrumented and received the same anesthetic and intensive care procedures as the trauma animals. Blood processing and analysis Blood samples were obtained immediately after implementation of the central venous catheter (0 h) and 3 h, 9 h, 24 h, 48 h and 72 h after polytrauma in prechilled ethylenediaminetetraacetic acid tubes (EDTA, S-Monovette ® , Sarstedt, Nümbrecht, Germany) and kept on ice. Blood was centrifuged at 2000×g for 15 min at 4 °C and supernatant was stored at −80 °C until further analysis. CC16 and IL-8 in the bronchoalveolar lavage fluid Bronchoalveolar lavage (BAL) was performed after collection of the right lung with 100 mL PBS using a perfusor syringe and a perfusor line. The obtained BAL was given into EDTA containing tubes (Sarstedt, Germany) stored oat 4 °C and subsequently centrifuged at 2200×g to measure the CC16 and IL-8 concentration in the remaining supernatant by ELISA. Samples were stored on −80 °C until measurement. CC16 levels in lung tissue After lung perfusion with PBS via the pulmonary artery three pieces of the right lobe of the lung were immediately snap frozen using liquid nitrogen and then stored at −80 °C until sample analysis. For lung tissue sample analysis, lung tissues were crushed with the help of a mortar and pestle, previously cooled down with liquid nitrogen. The remaining tissue powder was incubated with RIPA lysis and extraction buffer (10 µl buffer/1 mg tissue; Thermo Fisher Scientific, Massachusetts, USA) for 15 min at 37 °C. The suspension was centrifuged at 2200×g for 15 min on 4 °C. To determine CC16 and/or IL-8 concentrations in BAL and plasma samples, those were spun down for 15 min at 2200×g at 4 °C and subsequently the supernatant was used for testing. Lung tissue samples were prepared as described above. Total protein determination in BAL Protein concentrations were determined in BAL using the Pierce ® BCA Protein Assay Kit (Thermo Scientific) following the manufacture's manual. ZO-1 immune histology of lung tissue The left lobe of the lung was flushed with 4% formalin for overnight fixation and kept in 70% ethanol until paraffin embedding and sectioning. Paraffin-embedded lung samples (left lobe) were sectioned (3 µm), deparaffinized, rehydrated and stained with anti-ZO-1 antibody (Affinity Biosciences, USA). Following deparaffinization, antigen retrieval was done under steam atmosphere using R-Universal epitope recovery buffer (Aptum, Kassel, Germany) for 1 h (Retriever 2010, Prestige Medical). The endogenous tissue peroxidase activity was blocked with hydrogen peroxide according to the manufacturer's instructions (Peroxidase UltraVision Block, Dako). After washing with water and PBS, Anti-ZO-1 antibody (1:100, Cat.#: AF5145) was applied as a primary antibody. After the incubation for one hour at room temperature, and a subsequent washing procedure, a secondary antirabbit horseradish peroxidase-linked antibody (Nichirei Biosciences Inc.) was applied to detect specific binding. As substrate, 3-amino-9-ethylcarbazole (AEC, DCS Innovative Diagnostik-Systeme, Hamburg) was applied. Then, the sections were counterstained with hematoxylin. The relative staining intensity of the AEC substrate per slide was evaluated using the ImageJ software in a blinded manner by an independent examiner. Statistical analysis GraphPad Prism 6.0 software (GraphPad Software Inc. San Diego, CA, USA) and BM SPSS Statistics 27 (IBM, New York, USA) were used to perform the statistical analyses. The data distribution was tested by the D'Agostino-Pearson test to test for normal distribution. For repeated measures, the non-parametric Friedman`s test was applied. Dunn-Bonferroni post hoc test for multiple comparisons was used. The comparison between polytrauma and sham groups was performed using the unpaired T test. Data are given as mean ± standard error of the mean (SEM). A p value below 0.05 was considered statistically significant. Systemic CC16 course after polytrauma The time course of CC16 concentrations in the comparison of PT and sham animals has shown a clear difference between the two groups by significantly increased values in the polytrauma group at 3 h and 9 h after trauma compared to sham (3 h: 1601.1 ± 116.0 pg/g vs. 1065.71 ± 99.93 pg/g; 9 h: 1452.13 ± 98.67 pg/g vs. 1018.68 ± 56.36 pg/g; p < 0.05, Fig. 1A). CC16 expression in lung tissue and BAL The comparison of the CC16 concentration in lung homogenates has shown significantly higher CC16 levels in the lung tissue of sham animals compared to PT (2393.32 ± 219.14 µg/g vs. 1929.55 ± 253.36 µg/g; p < 0.05, Fig. 1B). CC16 levels in the BAL from polytraumatized 0.05, Fig. 1C). Local inflammatory changes in BAL The measurement of IL-8 and total protein concentrations in BAL of sham and PT animals revealed that IL-8 and total protein content were significantly higher in the PT group compared to sham at 72 h after polytrauma (p < 0.05, Fig. 2A, B). Examination of lung injury and epithelial barrier breakdown after polytrauma The evaluation of lung injury and loss of tight junctions by immune histological staining with a ZO-1 specific antibody has shown that the sham group had a significantly higher ZO-1 protein expression intensity compared to the PT group (p < 0.05, Fig. 3D). Discussion Elevated systemic CC16 levels in conditions of acute as well as chronic lung damage have been shown before [8,14,21]. Nevertheless, there are still scientific debates about which markers are reliable, especially in multiple injuries, to assessing lung injury. Many of these markers were not specific enough or were subject to major changes in their expression levels regarding the overall damage, a disease state/pre-existing conditions, or the etiology and have been influenced by the time of measurement [22][23][24][25][26]. Our results clearly indicated that CC16 has been significantly changed both locally in the lung tissue as well as systemically after polytrauma with thoracic injury. Interestingly, the systemic levels of CC16 increased very early after polytrauma. These findings are in direct agreement with data from Wutzler et al. and Lin et al. [11,27]. In these studies, an early increase within the first hours after trauma was detected reaching baseline levels within 24 h again. In addition, Wutzler et al. have shown a second CC16 increase in sera obtained from traumatized patients who developed pulmonary complications [11]. The time course of the CC16 increase and the renewed decrease to baseline level within 24 h was reproduced in the underlying polytrauma model. Due to multiple sampling time points of plasma before and directly after trauma induction, resulting in a higher temporal resolution, it has been shown that the concentrations' peak was around 3 h after trauma with baseline values being reached again after 24 h. Whether this was dependent on the trauma severity, or the intensity of lung damage remains to be investigated in further studies. The secondary increase of CC16 serum levels in clinically later occurring respiratory problems has been described by Wutzler et al., and was not reproduced in our 72 h model, but is also might not be expected within this short ventilation period in our model [11]. Prolonged observational time points are necessary to validate the clinically observed relevance of CC16 as a potential indicator for the secondary lung complications after trauma. Another interesting effect regarding the different concentrations of CC16 over time in the different compartments might have occurred due to a short-term high, but then decreasing and persisting release of CC16. While CC16 increased very quickly systemically and decreased after a short peak, CC16 was only significantly increased in the The results are represented as the mean ± SEM, *p < 0.05 vs. sham BAL at 72 h after polytrauma. Furthermore, CC16 was present in lower concentrations in the lung of polytraumatized animals after 72 h compared to sham animals. These differences may have occurred due to an initial storage dependent release, which was possibly followed by a synthesis and expression dependent release. Lung damage or respiratory complications are often driven by inflammatory changes involving the proinflammatory cytokine IL-8. The above described significantly increased values of IL-8 in the BAL obtained from polytraumatized animals have been associated with the loss of pulmonary barrier function, and a CC16 increase found in polytraumatized animals. In parts this has been confirmed by Reynolds et al. who demonstrated in their mouse model over-expressing transgenic human IL-8 that the prolonged presence of high IL-8 concentrations, as observed in trauma, exerted a damaging effect to the lung [28]. The authors demonstrated modifications in the context of long-term structural damage, with damaged and leaky epithelial tight junctions. Yang et al. showed that this might have been based on surfactant A and B expression depending on IL-8 occurrence [29]. In addition, Lin et al. have demonstrated that CC16 levels at ICU admission directly correlated with the severity of ARDS, but they only provided limited information for prognostic purposes [27]. In the present study, lung injury has been confirmed by increased total protein content in the BAL, which indicated functional disruption of the lung barrier. This is in line with the findings of Herrero et al. who found a direct correlation between ZO-1, loss of barrier function and edema formation in their mouse model [30]. In our study, sham animals have shown significantly higher values of membrane-bound ZO-1, which represented a more intact and functional lung barrier here. Taken together, our findings indicate that CC16 may constitute as an early biomarker for the pulmonary barrier loss that is associated with increased lung injury in this large animal polytrauma model, thus suggesting that this model reflected the pathophysiological modulations that have been observed in the human situation of polytrauma. Furthermore, this study provides insights into changes of CC16 levels in lung tissue Limitation Since the measurements were made exclusively in comparison of a polytrauma to a sham group and no isolated thoracic trauma group was included, the influence of other organ injuries, cannot be completely excluded. The characteristics as a lung epithelial injury marker can only be drawn by comparison to other animal models. Therefore, to be able to make a definitive statement, further investigations would have to be carried out taking into account an isolated thoracic trauma group including different degrees of trauma and outcomeoriented longer time courses. Furthermore, further studies need to be undertaken to evaluate the informative value of CC16 and its expression patterns under pre-disease conditions. Nevertheless, our study indicates that CC16 might also be used as a marker for lung injury in the porcine model of polytrauma. The potentially lung-protective inhalation of CC16 would be a worthy further study to consider if this is a therapeutic target. The temporal limitations to 72 h of the monitoring phase do not allow conclusions to be drawn about unaffected outcome at this time, or the development of later occurring organ and/or multiple organ failure in this model. Further studies would need to pay more attention to important cell populations such as e.g. neutrophils and other immune-related cells. Conclusions Our data demonstrated that CC16 may be used as a potential marker of lung injury in this porcine polytrauma model. Early increased systemic CC16 levels indicated lung damage after polytrauma, while the local CC16 increase in BAL may be used as a later marker even 3 days after polytrauma to indicate the lung injury which is associated with the barrier breakdown and inflammatory changes.	2022-05-22T06:22:55.788Z	2022-05-21T00:00:00.000	{ "year": 2022, "sha1": "49132bc99a1cdcbb51a0f3ec30821c17d2fafcbd", "oa_license": "CCBY", "oa_url": "https://link.springer.com/content/pdf/10.1007/s00068-022-01997-w.pdf", "oa_status": "HYBRID", "pdf_src": "Springer", "pdf_hash": "66afe17e89f8477f12269089d3bd732e87f029a2", "s2fieldsofstudy": [ "Medicine", "Environmental Science" ], "extfieldsofstudy": [ "Medicine" ] }
6365043	pes2o/s2orc	v3-fos-license	Optimum organ volume ranges for organs at risk dose in cervical cancer intracavitary brachytherapy Purpose To analyze the optimum organ filling point for organs at risk (OARs) dose in cervical cancer high-dose-rate (HDR) brachytherapy. Material and methods In a retrospective study, 32 locally advanced cervical cancer patients (97 insertions) who were treated with 3D conformal external beam radiation therapy (EBRT) and concurrent chemotherapy during 2010-2013 were included. Rotterdam HDR tandem-ovoid applicators were used and computed tomography (CT) scanning was performed after each insertion. The OARs delineation and GEC-ESTRO-based clinical target volumes (CTVs) contouring was followed by 3D forward planning. Then, dose volume histogram (DVH) parameters of organs were recorded and patients were classified based on their OARs volumes, as well as their inserted tandem length. Results The absorbed dose to point A ranged between 6.5-7.5 Gy. D0.1cm3 and D2cm3 of the bladder significantly increased with the bladder volume enlargement (p value < 0.05). By increasing the bladder volume up to about 140 cm3, the rectum dose was also increased. For the cases with bladder volumes higher than 140 cm3, the rectum dose decreased. For bladder volumes lower than 75 cm3, the sigmoid dose decreased; however, for bladder volumes higher than 75 cm3, the sigmoid dose increased. The D2cm3 of the bladder and rectum were higher for longer tandems than for shorter ones, respectively. The divergence of the obtained results for different tandem lengths became wider by the extension of the bladder volume. The rectum and sigmoid volume had a direct impact on increasing their D0.1cm3 and D2cm3, as well as decreasing their D10, D30, and D50. Conclusions There is a relationship between the volumes of OARs and their received doses. Selecting a bladder with a volume of about 70 cm3 or less proved to be better with regards to the dose to the bladder, rectum, and sigmoid. Purpose Twenty percent of deaths from cancer are due to gynecological (GYN) malignancies [1]. 3D image guided brachytherapy (IGBT) in combination with external beam radiotherapy (EBRT) is a favorite treatment option for all stages of cervical cancer, especially for inoperable patients [2,3,4,5,6]. To achieve sufficiently high-dose levels to the tumor, organs at risk (OARs) should be spared to reduce radiation related morbidity. Due to the dose-response effect, it is important to limit dose levels below defined dose constraints. The dose to the target and OARs can be optimized by using brachytherapy applicator with appropriate geometries and dwell time optimization based on the anatomical situation at the time of brachytherapy. The filling status of organs, like the bladder or rectum, has an effect on organ dosimetry [7]. In 2005 and 2006, GEC-ESTRO (Groupe Européen de Curiethérapie -European Society for Radiotherapy and Oncology) published recommendations on target definition and dose-volume histogram (DVH) parameters [8,9]. In 2012, the American Brachytherapy Society (ABS) endorsed the GEC-ESTRO recommendations to have "a common language" between distinctive brachytherapy centers [10,11]. The upcoming ICRU/GEC-ESTRO report 89 summarizes the state-of-the-art recommendation for reporting and recording dose and volume parameters [12]. Dose-volume histogram of OARs and clinical target volumes (CTVs), recommended by GEC-ESTRO, can be affected by bladder volume [13,14,15]. In the female pelvis, OARs are located in the close proximity to the intracavitary applicators, and there is a high-dose gradient from the dwell positions inside these applicators. Therefore, the influence of the bladder volume on the received dose by the OARs located in the pelvis can be evaluated to find the optimized bladder volume. The effect of the bladder distention on received doses by the bladder, rectum, sigmoid, and small bowel during the HDR brachytherapy procedure has been discussed by previous scholars. Some of their research results were based on the doses to ICRU-38 points. Therefore in their studies despite the use of sectional images, some standard planes were chosen to deliver an acceptable dose to point A [7,15]. Several studies have been performed on a fixed protocol of bladder fillings but they could not find an optimum bladder volume to minimize the dose to OARs [16,17]. The aim of this study is to find an optimum bladder volume to minimize the dose to pelvic OARs. Due to the lack of reproducibility in comparing previous research methods, this study was arranged as an extensive survey in four major HDR brachytherapy centers of Tehran, Iran, within more than two years research. In these centers, 3D forward treatment planning, manual optimizations, and dose reporting criteria were conducted based on the GEC-ESTRO recommendations for GYN brachytherapy, especially for the delineation of the intermediate and high risk-CTV (CTV IR and CTV HR , respectively) [5,6]. Then, effects of the bladder volume on the DVH parameters of the bladder, rectum, and sigmoid were assessed. The effect of tandem length on these parameters was also evaluated. Meanwhile, an optimum bladder distention was proposed for the bladder volume in GYN intracavitary brachytherapy, so that delivers the minimum dose to aforesaid normal tissues. Case selection A total of 32 patients (97 insertions) were selected from locally advanced cervical cancer patients treated in the four HDR brachytherapy centers from March 2010 to June 2013. Their median age was 56 years old (ranged between 47 and 64) and their median weight was 72 kg (ranged between 64 and 78). No exclusion was made either based on the patients' age or weight. The patients were treated with 3D conformal EBRT of 45-50 Gy for 5-6 weeks, with concurrent chemotherapy (50 mg cisplatin weekly). For all cases, the CTV HR was the cervix with no parametrical extension at time of brachytherapy. Therefore, all of these patients were candidates for intracavitary HDR tandem-ovoid brachytherapy without interstitial needles [18]. These patients were in different stages (IB2, IB, IIB) according to the FIGO staging [19]. The dose to point A of the studied cases was obtained between 6.5-7.5 Gy. Applicator insertion In this study, Rotterdam HDR tandem-ovoid applicators (Nucletron, an Elekta company, Elekta AB, Stockholm, Sweden) with different tandem lengths (i.e., 3, 4, 5, and 7 cm) and ovoid caps of 2-3 cm diameter were used for different uterine and vaginal dimensions [20]. The cases were chosen from the patients treated using a 30-degree tilted tandem applicator, which is suitable for a wide range of anatomies. The main reason for this inclusion criterion was that in two of the study centers, the only available tandem angle was 30° and in the other two, 93% of the cervical cancer patients were treated with this angle. By selecting these cases, one degree of freedom for statistical analysis was omitted. The insertions were performed under general anesthesia or partial sedation in the operation room. Patient preparation and imaging In all of the considered centers, a Foley catheter was inserted into the patients' bladder before implantations. Approximately 1 cm 3 of Meglumin compound (Amp VISIPAQUE™ 320 [50 cm³], GE Healthcare, Ireland), as a contrast agent, was mixed with 6 cm 3 normal saline and injected into the catheter balloon. In two centers, the bladder was filled with about 120 cm 3 of normal saline solution and clamped right before the CT scanning. Therefore, scans were performed with a specified bladder volume. Then, the clip was opened and the bladder emptied passively (without help from a syringe) until the beginning of the treatment when the procedure was repeated and, then, the patient was treated by assuming the identical bladder filling. In the third center, the Foley catheter was kept open during the whole procedure (i.e., imaging and treatment). In the fourth center, the Foley catheter was completely clamped from the beginning to the end of the procedure. The time between the clamping and CT scanning was about 30 minutes, and the time between clamping and the dose delivery was approximately one hour. Positioning devices were used to guarantee the reproducibility of the patients positioning. In all of the patient preparation, scanning, and transferring steps, a flat backboard was used and the patients were fixed in the supine position, as they were during the time of the treatment. Computed tomography scans of the whole pelvic were acquired using slices with a thickness of 3 mm (Somatom ® DR, Siemens healthcare, Forchheim, Germany). Contouring, treatment planning, and treatment Patient's CT images were all imported to a 3D brachytherapy treatment planning system (Flexiplan ® , version 2.6, Isodose control, The Netherlands). The CTV HR and CTV IR of each patient were contoured by a radiation oncologist on the basis of GEC-ESTRO GYN working group recommendations and using the available MR images as guides for delineations (at the time of diagnosis and also before the brachytherapy) [8,9]. The bladder, rectum, and sigmoid were also contoured. The rectum was considered from the levator ani muscles until it was horizontally oriented in the axial slices. From there onwards, the sigmoid was considered to have started and was countered until the first part of the descending colon. As the study was a multicenter survey, the small bowel dose was not evaluated because of the inconsistency in its delineation. The brachytherapy prescribed dose per fraction was between 7-9 Gy to D 90 of the CTV HR . This dose prescription was based on the linear quadratic model and EQD2 concept by considering the α/β of 10 Gy for the tumor and 3 Gy for late reacting normal tissues [21,22]. The treatment aims were to deliver 80-90 Gy as D 90 of CTV HR (EQD2) and not more than 70-75 Gy to the rectum and sigmoid, and 80-90 Gy to the bladder (D 2cm³ ). These were the dose constrains permitted for the whole radiotherapy procedure (EBRT plus typically 3-4 fractions of brachytherapy). The planning aim for one individual fraction was, therefore, to reach an absorbed dose of around 7 Gy. Inserted applicators were chosen from the software predefined library (reconstructions were done using three points). Treatment planning was initiated by rough dwell position/time suggestion of Flexiplan, followed by 3D adaption and manual optimization to ensure the desired planning aims were achieved (i.e., D 90 of CTV HR was not less than 80-90 Gy and OARs doses were no greater than the dose constraints). Classification The bladder, rectum, and sigmoid DVH parameters such as D 0.1cm³ and D 2cm³ were recorded in terms of percentages of the planning aim dose after each treatment plan. Patients were classified into four distinctive groups based on their bladder volumes (i.e., group A with less than 70 cm 3 , group B with 70 to 110 cm 3 , group C with 110 to 170 cm 3 , and Group D with greater than 170 cm 3 ). The rational for this type of patient classification was to have subgroups, in which the bladder volumes were the same as ones in the routine filling protocol of the clinics. Furthermore, another classification was made based on the physical tandem length and the patients were categorized into two groups (1) the patients who were treated with tandem lengths equal to or less than 4 cm, and (2) those who were treated with tandem lengths longer than 4 cm (Table 1). To consider the effect of patient rectum and sigmoid volumes on the DVH parameters of these organs, the volumes of these organs were also divided into different groups. The patients' rectal volumes were classified into R 1 (≤ 40 cm 3 ), R 2 (between 40 and 80 cm 3 ), and R 3 (> 80 cm 3 ). Also, the patients' sigmoid volume classification was as S 1 (≤ 35 cm 3 ) and S 2 (> 35 cm 3 ) ( Table 1). As mentioned earlier, two analysis steps were performed in order to find the approximate amount of the bladder volume, which caused the sigmoid and rectum to receive the minimum and maximum dose. In these analyses, we encountered some results that encouraged us to focus more on two out of the classified groups (i.e., B and D). Therefore, these groups were further divided into the following subgroups; B was divided into B 1 : 60 to 70 cm 3 , B 2 : 70 to 75 cm 3 , B 3 : 75 to 90 cm 3 , and B 4 : 90 to 110 cm 3 . Group B 1 intentionally has an overlap with group A, for finding the minimum bladder volume to deliver the minimum dose to the bladder more accurately. Group D was divided into D 1 : 170 to 180 cm 3 , D 2 : 180 to 200 cm 3 , D 3 : 200 to 220 cm 3 , and D 4 > 220 cm 3 , even though their statistical population was low. Note that the subgroups of B start from 60 cm 3 in order to find the sigmoid minimum dose point more clearly. To summarize the final results, we averaged the extracted DVH parameters of each classified groups (in terms of percentage of planning aim dose) and presented them below. Data analysis The DVHs of studied organs were compared descriptively among the cases in different classified groups to examine the effect of the bladder volume and tandem lengths on the OARs dose. In addition, we tried to find the optimum normal tissue volumes, which would lead to the minimum possible dose to the OARs. Numerical data analysis was performed through one-way ANOVA and general linear model (GLM)-univariate tests (confidence interval [CI] of 95%) using SPSS (SPSS Statistics for Windows, Version 17.0. Chicago: SPSS Inc. USA). To ensure the validity of the results and to prove the suitability of our data classification, normality of data distributions and homogeneity of their variance were tested by Kalmogorov-Smirnov (KS) and Levene tests, respectively. Considered continuous patients data included their FIGO stage, age, weight, and OARs' D 2cm³ (as the maximum dose to the organs). Also the normality of the classified group was checked by choosing the classification criteria as the "Factor List" parameter in the SPSS. Results For the normality test, we checked that all KS tests passed significantly, i.e. sig. > 0.05. Also, the normality distribution was checked graphically by the normal Q-Q and box plots, indicating that all normality distribution criteria passed (plots are not shown). The significance values of Levene tests were greater than 0.05 and tests for investigating the homogeneity of data variance passed. The effect of bladder volume The results of this study indicate that the minimum dose to the most exposed small volumes of the bladder (i.e., D 0.1cm³ and D 2cm³ ), increased by the bladder volume enlargement. An example of this claim is Figure 1, which illustrates how the bladder wall closes off to the high dose region by increasing its volume. On the other hand, large volume doses (i.e., D 10 , D 30 , and D 50 ) decreased, although with a lower slope ( Figure 2). For volume parameters (e.g., V 100 ; the V 100 is the volume, which receives the D 90 ) there were no significant differences (p value > 0.05) between different groups. However, the overall results indicated a reduction in irradiated bladder volumes with the increase in total bladder volume ( Figure 2). In all the classified groups, except V 100 , the differences in bladder DVH parameters were statistically significant (e.g. p value was obtained as 0.007 and 0.02 for D 2cm³ and D 0.1cm³ , respectively). The effect of tandem length The statistical results of the effect of tandem lengths on the bladder dose tests were significant (p values were 0.024 and 0.004 for D 2cm³ and D 10 , respectively), except for D 0.1cm³ . The results show that the dose to bladder and the irradiated volume of bladder increased when longer tandems were used (Table 2). This increase in the dose was equal to 4.8% and 4.9% (% of the planning aim dose) for D 2cm³ and D 10 of this organ, respectively. The effect of bladder volume and tandem length No statistically significant outcome on the bladder dose was obtained when considering applicator lengths and bladder volumes (GLM test), simultaneously. The effect of bladder volume on rectum dose The rectum dose increased as the bladder volume increased up to about 140 cm 3 . For higher bladder volumes, the rectum dose showed a decreasing trend ( Figure 3). However, the ANOVA test results were not significant. As it is illustrated in Figure 3, the dose to the rectum was maximum when the volume of the bladder was about 120-140 cm 3 . In this range, D 0.1cm³ and D 2cm³ of the rectum became about 90% and 69% of the planning aim dose, respectively. On the other hand, these rectum DVH parameters (i.e., D 0.1cm³ and D 2cm³ ) became 82% and 60% when the bladder volume was below 70 cm 3 , and they were 84% and 62% when the bladder volume was more than 170 cm 3 , respectively ( Figure 3). The effect of tandem length on rectum dose Like the bladder, the length of the tandem had a significant and direct influence on the rectum DVH parameters (p value of 0.004 and 0.026 for D 2cm³ and D 10 , respectively). For treatments with longer tandem lengths, the rectum dose was higher ( Table 2). For patients who were treated with longer tandem lengths (> 4 cm), D 2cm³ and D 10 of the rectum were 5.5% and 4.2% (percentage of the planning aim dose) higher, respectively. Multivariate statistical test results showed that there were no statistically significant effects from tandem lengths on the rectal dose. In different subgroups of bladder volumes, the rectal dose was only about 4% higher for those treated with tandem lengths of > 4 cm. The effect of rectum volume and tandem length on rectum dose As expected, by increasing the rectum volume, the D 0.1cm³ and D 2cm³ parameters increased; however, D 10 , D 30 , and D 50 decreased (Figure 4). While the rectum volume increased, the effect of tandem lengths on the rectum's DVH parameters became more dominant; this can be seen in Figure 4 where the discrepancies of the two related diagrams intensify for more filled rectums. Rectum D 0.1cm³ and D 2cm³ discrepancies were about 3% (of the planning aim dose) for cases with rectum volumes of less than 40 cm 3 and 10-13% for rectum volumes greater than 80 cm 3 . The effect of bladder volume on sigmoid dose Obtained results for the sigmoid are presented in Figures 1 and 5. The minimum dose to the sigmoid was obtained in the case of a bladder volume of 70 to 75 cm 3 . In this range, for example, D 0.1cm³ and D 2cm³ became 44% and 31% of the planning aim dose, respectively. The effect of tandem length on sigmoid dose As expected, tandem lengths had a direct influence on the sigmoid dose. The sigmoid received a significantly higher dose when longer tandems were used. For example, when the bladder volume was more than 170 cm 3 , the mean values of sigmoid D 0.1cm³ and D 2cm³ (in terms of the percentage of the planning aim dose) were 34% and 22% for patients with T > 4 cm and T ≤ 4 cm, respectively. For those whose bladder volume was 70-110 cm 3 , these values were 17% and 11%, respectively. However, this difference in sigmoid dose was decreased when the volume of the bladder was 110-170 cm 3 . For example, the difference in sigmoid D 0.1cm³ and D 2cm³ was only ~3% in this particular range. Figure 6 illustrates that the effect of the bladder volume on received doses by the sigmoid are less pronounced for patients treated with shorter tandem lengths (≤ 4 cm) than the longer ones (> 4 cm). The effect of sigmoid volume and tandem length on sigmoid dose By increasing the sigmoid volume, its minimum dose to the most exposed volumes (i.e., D 0.1cm³ and D 2cm³ ) increased to about 10% and 8% of the planning aim dose, respectively. However, the D 10 , D 30 , and D 50 decreased as expected (Table 3). Multivariate statistical test results for simultaneous effects of sigmoid volume and tandem length on sigmoid dose were significant (Table 3). For example, p values of this GLM test were 0.042 and 0.036 for D 2cm³ and D 10 , respectively. Discussion The effects of bladder, rectum, and sigmoid volumes as well as the tandem length on DVH parameters of the said organs in HDR intracavitary brachytherapy of cervical cancer patients were assessed. Also, the optimum volume range of the bladder was searched so that within this range, the mentioned OARs received their minimum dose values. In a study previously conducted by the same authors, the effects of two bladder filling situations (full and empty) on the OARs dose of individual patients were compared. In this paper the analysis was focused to find the optimum organ volume range, which were achieved via analyzing the data of 97 brachytherapy cases [23]. The first reason for the mentioned patient classification was that different brachytherapy centers apply different bladder filling protocols; they typically inject 1 to 3 full 60 cm 3 syringes of normal saline to the bladder Foley catheter before scanning it [15,24,25]. Another protocol for the centers, which believe in empty bladders is to leave the Foley clamp open [10,20]. However, according to our findings, even in opened Foley conditions, there would still be about 60-70 cm 3 urine. Another reason for this type of classification was to reach an acceptable statistical population for each subgroup. Our results for the evaluation of bladder dose variations by increasing the bladder volumes are in close consistency with previous studies (Figure 2) [14,16,26]. For example, D 2cm³ of this organ increased from about 75% to 95% (of the planning aim dose) by increasing the bladder volume from less than 70 to more than 170 cm 3 . It is difficult to extrapolate our values for empty bladder filling status. In principle, the dose to the bladder should decrease further. On the other hand, for a completely empty bladder, the anterior bladder wall would be in contact with the posterior bladder wall and, thus, the total volume of irradiated bladder wall is increased. This effect is not directly visible with DVH analysis. Therefore, there is no evidence to treat patients with a completely empty bladder at this point. According to the results attained from the rectum, it can be concluded that the rectum dose can be affected by the bladder volume. Increasing bladder volume to about 140 cm 3 would cause pushing the applicator posterior towards the rectum and, thus, increasing the rectum dose. However, for patients whose bladder volumes were more than 140 cm 3 , this effect on rectal dose was negligible. This statement is in agreement with previous research results [14,16]. By continuing the bladder distension (more than 140 cm 3 ), most of its movement would be upwards (towards the head). It yields a tandem tip orientation, which is more or less horizontal (in the patient supine position), to be directed to the sigmoid and intestine. Comparing the results of Figure 3 and Figure 5, it can confirm this claim. However, these findings are in contrast with some previous studies [7,24], which may be due to the difference in statistical population, dose optimization protocol, and the utilized applicator types. The dose to the sigmoid can be decreased by filling the bladder up to 75 cm³. For higher bladder volumes, the effect turns opposite again ( Figure 5). This can be explained by taking into account the shifting of the bladder due to its various volumes. The discrepancy of the results in this study from those obtained by other authors [14,24] can be explained by the separate delineation of the rectum and sigmoid in the current study. For example, Yamashita et al. defined the rectum and sigmoid colon together as the recto-sigmoid in their study. They found significant reduction of hot spot dose to the small bowel, parallel to the growth of the bladder D 2cm³ dose, without any significant outcome in the rectum or sigmoid [24]. The most important factors in choosing the tandem lengths by the physician is the patient's uterine canal length. However, our results showed that the dose to the OARs will be increased by increasing this length. Knowing the relationship of this increase in the dose and the OARs volumes can help the clinicians with choosing the appropriate protocol for the patient preparation before the treatment. An increase in rectum and sigmoid volume has a direct impact on increasing their hot dose points and decreasing their D 10 , D 30 , and D 50 . Increasing the D 2cm³ and D 0.1cm³ of these organs will result in telangiectasia, bleeding, and fistula [21,27]. Therefore, controlling these parameters by finding the optimum combination of these organ volumes, bladder volumes, and also tandem lengths can prevent the occurrence of such complications [28]. When the bladder volume was more than 110 cm 3 , D 0.1cm³ and D 2cm³ of the sigmoid became constant to some extent and even were decreased for tandem lengths of ≤ 4 cm ( Figure 6). This means that the shifting of the uterus and its consequent effects on the surrounding organ dose is related to the tandem length. Considering the simultaneous effects of tandem lengths and sigmoid volume, it can be concluded that loading short tandems, a fuller sigmoid will get more dose than an emptier one. On the other hand by inserting a longer tandem, a sigmoid with a lower volume will be exposed to higher doses (Table 3). Therefore, even though the tandem length is not a changeable factor in treatments, the patient's anatomical conditions (i.e., filling status of OARs) can be adjusted to the appropriate length to expose the OARs by a lower dose. For example, the prescription of a two-day bowel preparation before brachytherapy can have a significant impact on the patient's treatment quality. As mentioned in the above, we used the ANOVA test (GLM) in this study. With knowledge of the assumption of the sample dependence in this test, we were encouraged to perform another statistical analysis by taking into account just one fraction of the data for each patient (i.e., considering only 32 cases) and repeating this test on them. Resulted data for evaluating the bladder dose dependency on its volume, showed a similar trend to the previous results in D 10 , D 30 , and D 50 for the bladder. But, for D 0.1cm³ and D 2cm³ of this organ, we were unable to detect any statistically significant differences among different bladder volumes in contrast to our previous results. This might be due to the small statistical population of the new analysis. Furthermore, for the rectum and sigmoid, the trend of this new investigation was the same as the previous one, proving our assumption about the independency of the 97 cases. In the current research, the patients were not forced to obey any precautions before their therapy, and the effects of bladder, rectum and sigmoid volumes on their received dose were taken into account, in routine treatment conditions. Therefore, we were unable to consider the effects of these organ volumes separately, and consequently some of our results were affected by this fact. For more accurate results, it is recommended that more research is conducted to consider these factors separately by applying specific protocols for patient preparation before the study. The other limitation of this study was that only the physical tandem length could be analyzed. It remains unclear if the main impact is due to the inserted length of the tandem or the length of source loading inside the intrauterine channel. Conclusions There is an optimum volume range for the bladder, rectum, and sigmoid, which can have a direct relationship with the received dose during intracavitary brachytherapy. The effect of a tandem length on the DVH parameters of OARs can be quantified. The most important findings of this study for different organs can be summarized as follows: Bladder: High-dose parameters of bladder are increase by increasing the bladder volume. Tandems longer than 4 cm increase the dose to the bladder up to about 4% (of the planning aim dose) more, compared to the shorter tandems; Rectum: Rectum dose reaches its minimum and maximum levels for empty and 120-140 cm 3 bladders, respectively. Choosing a longer tandem leads to about 5% more dose to the rectum. For the cases with tandems > 4 cm, D 2cm³ of the rectum with bladder volumes more than 80 cm 3 is about 10% (of the planning aim dose) higher compared to the cases with bladder volumes less than 40 cm 3 . Rectum volume does not have any drastic influence on the dose to rectum for patients whose tandems are ≤ 4 cm. Sigmoid: When the bladder volume is 70-75 cm 3 , the sigmoid dose will be minimum. Tandem lengths have a direct relationship with sigmoid dose. By increasing the sigmoid volume (i.e., more than 35 cm 3 ) the sigmoid dose is increased (e.g. 13% for D 2cm³ ) for the cases who were treated with tandem lengths of ≤ 4 cm. However, the sigmoid dose is decreased when increasing its volume for tandems of > 4 cm. In conclusion, choosing a bladder with a volume of about 70 cm 3 or less is recommended when taking into account the high dose volume parameters for bladder, rectum, and sigmoid.	2018-04-03T04:28:27.337Z	2016-04-01T00:00:00.000	{ "year": 2016, "sha1": "6369b5964f8b21fa6af875aafe6d40b948ce3d85", "oa_license": "CCBYNCSA", "oa_url": "https://www.termedia.pl/Journal/-54/pdf-27483-10?filename=optimum%20organ.pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "6369b5964f8b21fa6af875aafe6d40b948ce3d85", "s2fieldsofstudy": [ "Medicine", "Physics" ], "extfieldsofstudy": [ "Medicine" ] }
263277241	pes2o/s2orc	v3-fos-license	Dissecting the efficacy of the use of acupuncture and Chinese herbal medicine for the treatment of premature ovarian insufficiency (POI): A systematic review and metaanalysis Premature ovarian insufficiency is a multi-factor gynecological disease that has become a major global health problem. In recent years, several trials have explored the treatment of premature ovarian insufficiency using Chinese herbal medicine and acupuncture, but the efficacy and safety of this combination remains controversial. This systematic review and meta-analysis aimed to comprehensively evaluate the efficacy and safety of combining Chinese herbal medicine with acupuncture to treat premature ovarian insufficiency. From eight different databases, we retrieved randomized controlled trials wherein Chinese herbal medicine and acupuncture had been compared with western medicine in the treatment of premature ovarian insufficiency. The bias risk assessment stipulated by the Cochrane Collaboration's tool was utilized to evaluate the quality of the chosen randomized controlled trials. This meta-analysis was executed with the help of Review Manager 5.3 and Stata 10.0. The quality of evidence was assessed using the Grading of Recommendations, Assessment, Development, and Evaluation framework. A total of 10 randomized controlled trials involving 594 premature ovarian insufficiency patients were included in the analysis. Compared with western medicine, co-treatment with acupuncture and Chinese herbal medicine exhibited a significantly higher total effective rate (relative risk: 1.21; 95% confidence interval: 1.12–1.31; P < 0.01, I2 = 0%), but lower levels of luteinizing hormone (standardized mean difference: −0.57; 95% confidence interval: −1.06, −0.08; P < 0.05, I2 = 80%), follicle-stimulating hormone, and Kupperman index score. Moreover, the combined intervention increased estradiol level in the serum. Overall, the data demonstrate that acupuncture plus Chinese herbal medicine is an efficacious and safe treatment option for POI patients. These findings must be verified by conducting large-scale, multicenter, high-quality, and long-term randomized controlled trials. Introduction Premature ovarian insufficiency (POI) is characterized by compromised hormonal and reproductive functions in females under the age of forty.It currently affects approximately 1% of women in the general population, and its incidence is persistently increasing [1]. POI is a gynecological endocrine illness caused by multiple factors, which may be related to socioeconomic status [2], autoimmune disorders [3], or prenatal ethanol exposure [4].POI makes females more susceptible to multiple health risks, such as infertility [5], lipid disorders, early progression of cardiovascular illnesses [6], osteoporosis [7], psychiatric disorders, and other adverse events [8,9].Besides, it increases the economical and psychological burden on their families and society [10]. Despite the huge global efforts to manage this disease, no established treatment exists.Hormone replacement therapy (HRT) is recommended for women with POI, especially to treat the symptoms of low estrogen.It alleviates vasomotor and genitourinary symptoms and prevents bone loss and cardiovascular disease.However, the medical history of patients must be strictly assessed before HRT is administered; for example, in breast cancer survivors, HRT is generally contraindicated by the international recommendations.The potential risks associated with HRT have been previously reviewed [1]. Several POI patients have opted for alternative medicine, particularly traditional Chinese medicine (TCM), which has an exceptional concept of the etiology of POI [11].Chinese herbal medicine (CHM) in combination with acupuncture has been widely used to treatment multiple diseases, including POI [12][13][14][15].Previous systematic meta-analyses have shown that both CHM and acupuncture have positive therapeutic effects on POI [16,17].A recent meta-analysis indicated that a combination of acupuncture and other treatment options (CHM/western drugs) produces a superior therapeutic effect against POI compared with that produced by CHM or western drugs alone [18].However, this study had a relatively small sample size and did not distinguish between CHM and western medicine in the control group.Here, we have designed a more rigorous study that interrogates the therapeutic efficacy of combining acupuncture and CHM versus the use of Western drugs in patients with POI.This study identifies complementary therapeutic approaches for POI and can guide its clinical treatment. Material and methods This review has adopted the guidelines described in Preferred Reporting Items for Systematic Reviews and Meta-Analyses [19].The PROSPERO registration number is CRD42020190573. Data sources and searches Systematic data searches, without language restrictions for randomized controlled trials (RCTs), were conducted across the Wanfang database, Chinese Scientific Journals Database, PubMed, Chinese BioMedical Database, Web of Science, Embase, China National Knowledge Infrastructure, and Cochrane Library.We examined the efficacy of acupuncture plus CHM versus western drugs for treating patients with POI from the time each database was established to 31 July 2022.We devised a formula for retrieval following the PICOS strategy.The specific retrieval formula for China National Knowledge Infrastructure was "subject = (acupuncture + body acupuncture + electroacupuncture + needling therapy + needle acupuncture) AND (premature ovarian failure + primary ovarian insufficiency + premature ovarian insufficiency + POI + POF) AND (randomization + randomized controlled + random grouping + RCT + clinical research)".The search strategies applied to PubMed have been reported in Supplement Digital.In addition, we manually searched the reference lists of all selected articles to find probably related research. Inclusion criteria We included RCTs that involved patients diagnosed with POI according to a clear diagnostic criterion, such as the guidelines of the European Society of Human Reproduction and Embryology, regardless of their ethnicity or nationality, and wherein they received acupuncture (including needle acupuncture and electroacupuncture) plus CHM, irrespective of the course of treatment, prescription name, dosage form, and dosage.Studies that compared acupuncture plus CHM with western drugs, and those with at least one clear outcome, such as the follicle-stimulating hormone (FSH) level, total effective rate, estradiol (E 2 ) level, luteinizing hormone (LH) level, Kupperman index (KI) score, and the occurrence of complications, were also included. Exclusion criteria Studies involving animal experiments, commentaries, editorials, experience introduction, conference articles, reviews, graduation theses, and case reports were excluded from our analysis.We also excluded duplicate publications, studies whose original data were difficult to obtain, and those with intervention measures such as moxibustion, warm acupuncture, auriculotherapy, acupoint catgut embedding, or auricular acupoint pressing.Besides, articles without a clear outcome index were excluded. Data extraction Using a pre-designed, standardized data retrieval form, H. F. Li and W. J. Chen independently extracted data as follows: first author and year of publication, TCM syndrome differentiation, the number of persons, age, disease duration, interventions and comparisons, treatment duration, and outcomes.The third author (J.X. Zhang) resolved the disagreements between the two reviewers. Outcome measures The primary outcome was the total effective rate, which was assessed according to the Guidance Principle of Clinical Research on New Drug of Traditional Chinese Medicine [20].For trials that assessed the treatment effect as different classification without total effective rate, we integrated the effective classification into "totally effective" for analysis.The secondary outcomes were serum E 2 , FSH, and LH levels, KI score, and the incidence of adverse events. Quality assessment Two authors (W.J. Chen and H. F. Li) independently evaluated the bias risk for the selected trials in accordance with the Cochrane Handbook.Briefly, the criteria included: selective reporting, blinding of participants and personnel, random sequence generation, incomplete outcome data, blinding of outcome assessments, and allocation concealment.Each RCT was categorized as high, low, or unclear.Any disagreement was resolved by a third researcher (J.X. Zhang). Grading of Recommendations, assessment, development, and evaluation (GRADE) The quality of evidence was assessed by GRADE, and categorized as very low, low, moderate, or high.The criteria consisted of study design, indirectness of the evidence, inconsistency in the outcomes, and risk of bias, among others. Statistical analysis We executed this meta-analysis with the help of Review Manager (RevMan) (Version 5.3, Copenhagen; The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) and Stata (Version 10.0; Stata Corporation, College Station, TX, USA).For the study results, relative risk (RR) with 95% confidence interval (CI) were utilized as binary variables.Mean differences (MDs) and 95% CIs for the continuous data were computed using some measure evaluation units and methods.Then, the standardized MD (SMD) was calculated.Cochrane's P values and I 2 were used to assess heterogeneity among the RCTs.In cases where high heterogeneity existed due to methodological and clinical factors, the random effect model was employed even if I 2 was small.Subgroup analysis was carried out, premised on different types of acupuncture.Egger regression tests and funnel plots were executed to assess possible publication bias.Additionally, sensitivity analysis was undertaken by chronologically removing RCTs to test the stability of the primary outcome. Ethics and dissemination No ethical approval was required because this systematic review is based on published studies.Fig. 2. The risk of bias for each included study. Risk of bias Although all the studies were randomized, only four employed the random method (random number table) [21,24,26,27].None of the included RCTs used any blinding or concealed allocation of investigators and patients.Only one reported the number of dropouts as well as the reasons for them [28].None employed selective reporting.None of the studies computed the sample size in advance.A summary of the risk of bias is depicted in Fig. 2. Evaluation of the therapeutic effect 3.4.1. The total effective rate Nine studies assessed the total effective rate.Our data illustrated that patients who received acupuncture in conjunction with CHM exhibited a higher total effective rate than those treated with western drugs (RR: 1.21; 95% CI: 1.12-1.31;P < 0.01, I 2 = 0%, Fig. 3).Similarly, subgroup analysis showed that the acupuncture-CHM combination was associated with a superior total effective rate (Electroacupuncture: RR: 1.22; 95% CI: 1.07-1.40;P < 0.01, I 2 = 0%; Needle acupuncture: RR: 1.20; 95% CI: 1.09-1.33;P < 0.01, I 2 = Fig. 3. Forest plot for total effective rate between the acupuncture plus CHM and the western drugs group.(CHM, Chinese herbal medicine). Adverse events One study recorded episodes of nausea and vomiting and elevated alanine transaminase (ALT) [21], the occurrence of which was significantly lower occurrence in the acupuncture plus CHM group compared with that in the western medicine group. However, a definitive conclusion could not be drawn from this study. Publication bias and sensitivity analysis While the funnel plot for the total effective rate illustrated an asymmetrical distribution (Supplementary Fig. 1), Egger's test illustrated no probable publication bias (P = 0.06, Supplementary Fig. 2).Sensitivity analysis showed unchanged effect estimates, which depicted the robustness of the pooled results (Fig. 8). GRADE assessment The quality of evidence was evaluated by using GRADEpro.The quality of chosen RCTs was low, thus the "Risk of bias" was classified as "serious".Data regarding the serum FSH, LH, and E 2 levels, and KI score had high I 2 values, so the "Inconsistency" was classified as "serious".The results of the GRADE assessment for total effective rate were moderate, while the quality for other outcomes was low (Table 2). Summary of evidence This meta-analysis, which included outcomes from 10 RCTs with an aggregate number of 594 POI patients, provides a quantitative estimate of the clinical efficacy and safety of combining acupuncture and CHM to treat POI.The pooled data demonstrated that acupuncture plus CHM had a higher total effective rate than western medicine.The combination therapy yielded better outcomes with regards to serum LH, FSH, and E 2 levels, and KI score, without obvious adverse events.Subgroup analysis premised on the different types of acupuncture, such as electroacupuncture and needle acupuncture, was further performed to detect potential sources of heterogeneity.For some outcome measures, the heterogeneity did not decrease significantly, implying that it may originate due to other reasons, including different dosage forms and compositions of the CHM, different diagnostic criteria, or various needling points.However, due to the poor methodology, further studies confirming the beneficial effects and safety of the combination therapy are required. Significance of the study POI is a frustrating gynecological endocrine disease, whose etiology remains unclear.Some studies have associated it with iatrogenic or endogenous factors.Besides limited literature on its optimal management, no existing drug that can halt its progression, a situation that has severely affected the physical and mental health of women [31].Appropriate HRT plays an imperative part in the management of POI patients with strict indications and contraindications [32].However, it introduces minor risks of diseases such as breast cancer, cardiovascular disease, stroke, and venous thromboembolism, depending on the type of HRT, duration, and individual health risks [33].Therefore, many patients have resorted to complementary and alternative medicine for managing POI [34].Among all the options, a combination of two TCM techniques-CHM and acupuncture-has emerged as a common therapeutic approach for POI in China as well as other Asian countries, even for young survivors of ovarian and breast cancer [20,35,36]. In the CHM, the pathogenesis and etiology of POI are often associated with blood stasis and kidney deficiency, which influence each other [37].Hence, treatment via CHM aims to regulate hormone levels and improve ovarian function by improving blood circulation and tonifying the kidney.Both clinical and animal studies have illustrated that Chinese activating blood herbs and nourishing kidney herbs that are used to relieve POI symptoms, such as Chinese Yam, Dodder, and prepared Radix rehmanniae, exert phytoestrogen-like effects [38,39].They can regulate or enhance immune functions and improve ovarian blood flow, thus regulating reproduction [40][41][42][43].These herbs can also promote follicle growth through the phosphoinositide 3-kinase/Akt signaling pathway [44]. On the other hand, two different but widely used types of acupunctureelectroacupuncture and needle acupuncture-have been shown to enhance ovarian blood flow, regulate the reproductive endocrine system, and improve ovarian function in women with POI [45]; [46].The three most frequently utilized acupoints are Guanyuan (RN4), Sanyinjiao (SP6), and Zigong (EX-CA1), while the most employed meridians are the Ren Meridian, the Spleen Meridian of Foot-Taiyinand, and the Bladder Meridian of Foot-Taiyang.Most of the acupoints are distributed in the lower limbs, lumbar region, abdomen, and chest.A study found that acupuncture achieves the same therapeutic effect as estrogen in the treatment of POI, and might be acting by upregulating gene and protein expression in the phosphoinositide 3-kinase/Akt/mammalian target of rapamycin signaling pathway [47].Moreover, electroacupuncture pretreatment exerted a certain protective effect on ovarian health in rats with POI by upregulating ovarian granular cells and inhibiting the expression of Bcl-2 and Bax [48].Although acupuncture can produce noticeable and rapid effects without causing an adverse reaction, it is restricted by its action span and area [49]. Dating back to an ancient Chinese practice mentioned in Huangdi's Internal Classic (Huang Di Nei Jing), the combination of acupuncture and CHM is a comprehensive therapeutic option for preventing and treating diseases [14], [50]."when disease forms, acupuncture can be used to treat its outside and CHM to treat its inside" is one of the most important classical quotes in TCM.Acupuncture and CHM act via different mechanisms.While acupuncture can relieve pain, regulate qi and blood, and treat meridian diseases, CHM can tonify qi, nourish yin, and effectively treat Zangfu diseases [51].By affecting signaling pathways, acupuncture and CHM can play a regulatory role and produce a synergistic curative effect, lengthening the duration of efficacy, minimizing the side effects, and shortening the course of treatment, especially in POI patients who respond poorly to acupuncture or CHM alone.The use of acupuncture and herbs can treat hormonal disorders, creating a conducive environment for conceiving and maintaining pregnancy [52].Besides, the combined treatment option could improve ovarian function in rats with a decreased ovarian reserve.The treatments might be preventing or delaying apoptosis of follicular granulosa cells, slowing follicular atresia, and positively regulating follicular development in rats [53]. In summary, the acupuncture-CHM combination may innovate and optimize the treatment pattern, offering more options and broadening the scope of treatment for POI. Limitation While we have expansively analyzed and evaluated all the selected RCTs, this study was inevitably associated with some shortcomings.First, the selected RCTs had low quality since they employed unclear blinding methods, attrition bias, selective bias, and allocation concealment.Secondly, despite undertaking an unbiased literature search with no language restriction, all the RCTs chosen for this review were performed in China and published in Chinese with no relevant foreign experiments, which potentially contributes to a bias and negatively impacts their representativeness.Thirdly, only a few RCTs employed long-term follow-up, implying that the long-term safety of the interventions is unclear.Fourth, though the intervention and control cohorts were stringently integrated into this meta-analysis, high or moderate heterogeneity still existed, likely due to the different ingredients of CHM and the diverse acupuncture operations performed in the studies.Thus, the lack of high-quality RCTs inevitably impeded the robust assessment of the efficacy of the intervention, which impacted the stability and accuracy of the evaluation test. Conclusion To summarize, the results of our meta-analysis illustrated that the use of acupuncture in conjunction with CHM is a safe and efficacious therapeutic option for patients with POI.Nevertheless, since the quality of the selected RCTs is relatively low, further studies investigating follow-up and adverse events are needed to confirm these findings.Well-designed, large-sample, and multicenter clinical studies are required in the future to investigate the efficacy of acupuncture combined with CHM for treating POI, and to derive a scientific, objective, and reliable conclusion. Fig. 1 . Fig. 1.The inclusion process for the literature. Fig. 4 . Fig. 4. Forest plot for the LH levels between acupuncture plus CHM and western drugs group.(LH, luteinizing hormone; CHM, Chinese herbal medicine). Fig. 6 . Fig. 6.Forest plot for the E 2 levels between acupuncture plus CHM and western drugs group.(E 2, estradiol; CHM, Chinese herbal medicine). Fig. 7 . Fig. 7. Forest plot for the KI score between acupuncture plus CHM and western drugs group.(KI, Kupperman index; CHM, Chinese herbal medicine). Fig. 8 . Fig. 8. Sensitivity analysis for the total effective rate. Table 1 The basic characteristics of the included trials. Table 2 Results of GRADE evaluation.	2023-10-01T15:06:50.567Z	2023-09-01T00:00:00.000	{ "year": 2023, "sha1": "badd788a44922008594672c716d874f1d617620f", "oa_license": "CCBYNCND", "oa_url": "http://www.cell.com/article/S240584402307706X/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "a07813c3f301ac7d2ec81c1279ce20f5b26fa8b6", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
251696105	pes2o/s2orc	v3-fos-license	Exploring Effective Practices of an Elementary STEM Block Program : Creating a STEM-driven culture incorporating engineering habits of mind and 21st century skills at an early age could impact students’ STEM interests and knowledge. Therefore, early exposure to effective engineering design practices could create a foundation for a STEM program. This exploratory case study examined the integration of a STEM program in an elementary-level school. Survey, interview, focus group, and observational data were analyzed and coded to determine effective practices of the STEM program. This paper focuses on the emergent themes of the (a) critical role of the specialist, (b) instructional design, and (c) integration of the engineering laboratory. The STEM specialist at Gemini Elementary School provided the teachers with the foundation for the in-depth acquisition of STEM content and pedagogical skills. Teachers were provided with team planning time that focused on the instructional design of the STEM Block lessons. Through collaborative settings, teachers and the specialist were able to design modern real-world problems for students that allowed students to apply engineering design practices to find solutions. The results of this study point to the need to increase the number of STEM programs embrace engineering design in elementary schools. Introduction The United States (U.S.) is the land of opportunity that developed some of the world's greatest scientific and technological innovations (National Science Foundation, 2016).However, its future world standing may depend on its populace being prepared to think from a scientific and technical perspective; and, most importantly, people's economic opportunities may be limited to those with science, technology, engineering, and mathematics (science, technology, engineering, mathematics; STEM) backgrounds (Langdon et al., 2011;Sargent, 2017).A STEM-literate workforce and citizenry enhance the U.S. capability to compete globally and sustain efficient, diverse industries and production necessary to meet 21st century demands (NSTA, 2021).Exposing all students to STEM education at an early age may enhance their interest in STEM careers and provide an equal opportunity to all students (DeJarnette, 2012).The importance of providing students with early exposure was brought to light by the early learning framework presented by the Partnership for 21st Century Skills (P21, 2017) and other researchers (Bybee & Fuchs, 2006). Incorporating engineering opportunities into STEM schools could be one tool educators use to capture students' interest and engagement while introducing them to real-world applications and STEM careers.According to Tanenbaum (2016), early interest in STEM concepts is the foundation for creating a competitive STEM workforce prepared to meet the challenges of the rapid development of scientific and technological innovations.However, there is little research regarding implementing integrated STEM initiatives in elementary schools (Peters-Burton et al., 2019).In addition, there is a dearth of research regarding STEM schools; thus, a need for STEM school case studies needs to be conducted (Lesseig et al., 2019).This case study sought to identify features of an integrated STEM program and how the use of an engineering laboratory supports integrating the STEM curriculum in a STEM elementary school Abstract: Creating a STEM-driven culture incorporating engineering habits of mind and 21st century skills at an early age could impact students' STEM interests and knowledge.Therefore, early exposure to effective engineering design practices could create a foundation for a STEM program.This exploratory case study examined the integration of a STEM program in an elementary-level school.Survey, interview, focus group, and observational data were analyzed and coded to determine effective practices of the STEM program.This paper focuses on the emergent themes of the (a) critical role of the specialist, (b) instructional design, and (c) integration of the engineering laboratory.The STEM specialist at Gemini Elementary School provided the teachers with the foundation for the in-depth acquisition of STEM content and pedagogical skills.Teachers were provided with team planning time that focused on the instructional design of the STEM Block lessons.Through collaborative settings, teachers and the specialist were able to design modern real-world problems for students that allowed students to apply engineering design practices to find solutions.The results of this study point to the need to increase the number of STEM programs embrace engineering design in elementary schools. in Texas.The following research question was explored in this study: What features of an elementary-level integrated STEM program have the potential to contribute to its effectiveness? STEM's Impact on a Global Economy A strong STEM foundation will help secure the United States' position as a world leader in innovation and contribute to the economic opportunities available to all citizens.More recent BLS projections include all STEM jobs to reach over 10 million occupations by 2030, which is a 10.0% increase from 2019 (BLS, 2020).Quality jobs could be critical to keeping the U.S. globally competitive because of the purchasing power they provide its citizens to fuel the national and global economies.A dynamic global economy and workforce have prompted an increase in the discussion concerning STEM education and concern over a shortage of prepared STEM workers and educators worldwide (Kennedy & Odell, 2014;Sargent, 2017).To remain competitive in a global economy, key stakeholders have encouraged students to develop an interest in STEM (Cunningham et al., 2015;National Science Board, 2007; President's Council of Advisors on Science and Technology, 2012).People's standard of living may also be connected to STEM innovations and their impact on the U.S. economy (Shernoff et al., 2017). Additionally, maintaining the U.S.'s ability to compete globally comes from having STEM-literate citizens and a workforce that also includes technical workers that "help secure our health and safety, revitalize our utility infrastructures, monitor our food production, and improve our manufacturing efficiencies and capabilities" (NSTA, 2021, para. 5).However, the lack of students' STEM preparedness threatens the U.S.' economic growth (Rozek et al., 2017) and may threaten students' future opportunities for work.Building a solid understanding of the importance of an engaging and practical STEM application may involve capturing students' interest earlier to help develop them for a competitive and innovative global market. Our rapidly changing world has demanded that students be flexible and respond quickly to the innovations, advancements STEM careers require, and new jobs that have not been created (Morgan et al., 2013).Businesses have been compelled to adapt quickly and require a workforce trained in critical thinking and problem-solving skills to meet complex 21st century challenges (Miller, 2017).Early interest in STEM concepts is foundational for creating a competitive workforce (Tanenbaum, 2016).Therefore, efforts to incorporate real-world problem solving and critical thinking in U.S. educational systems, especially with elementary students, may equip students with 21st century skills, thinking, and competencies. Obstacles to STEM Education However, obstacles exist to implementing STEM education.Inconsistencies in integrating STEM into K-12 educational systems create problems in comparing what pedagogical approaches best support learning and how results compare among disciplines (NRC, 2014).These inconsistencies add to the difficulty of making connections among STEM subjects and courses, so students may fail to understand how the integration of these subjects applies to realworld applications (Breiner et al., 2012).The difficulty in defining STEM and STEM integration could limit the J. of Res. in Sci. Math. and Tech. Edu.\| 197 implementation of STEM education in more schools and limit the students entering the STEM workforce in the future (NRC, 2014;Schneider et al., 2016).Educators, policymakers, and other stakeholders must agree on common terminology to increase the effectiveness of STEM education. Since there are various definitions of STEM and multiple interpretations of STEM integration (Roehrig et al., 2021), it is difficult for stakeholders to agree on exactly how to increase the opportunities for STEM education in the U.S. (Schneider et al., 2016).Some researchers define integrated STEM as "the approach to teaching the STEM content of two or more STEM domains, bound by STEM practices within an authentic context for the purpose of connecting these subjects to enhance student learning" (Kelley & Knowles, 2016, p. 3).In comparison, others state that it is "an effort to combine some or all of the four disciplines of science, technology, engineering, and mathematics into one class, unit, or lesson that is based on connections between the subjects and real-world problems" (Moore et al., 2014, p. 38).The lack of agreement on the definition makes it difficult for educators to implement comprehensive and uniform STEM curricula. In addition, elementary school educators may be reluctant to implement STEM curricula due to their lack of STEM content knowledge (Hammack & Ivey, 2017;Watson et al., 2020).In a study by Hammack and Ivey (2019), participants appeared adamant that they would not implement engineering concepts from training due to a lack of background knowledge.Moreover, many elementary teachers teach multiple content areas and may not have adequate planning time to implement STEM content that is not assessed in standardized testing elementary (Hammack & Ivey, 2019).Therefore, some elementary teachers may perceive incorporating engineering, a necessary STEM concept, as not feasible due to administrators' focus on "state-mandated assessments and reading" (Hammack & Ivey, 2019, p. 516).Alleviating issues and concerns facing gatekeepers, such as elementary teachers, may encourage openness to new pedagogical approaches. Benefits of Integrating STEM Multiple interpretations have led to the evolution of STEM education and its role in society, expanding interpretations and benefits of integrating STEM concepts throughout schools.The National Science Teachers Association (NSTA) is taking a more expansive understanding of STEM education.It includes fields such as computer science, "the designed world," and robotics that integrate science, technology, engineering, and mathematics concepts to solve realworld problems by creating modern solutions (NSTA, 2021).STEM integration helps determine how things work and how technologies are created and provide students with authentic learning experiences, including problem-solving, innovation, and design, which are three themes with high priorities on every nation's agenda (Hernandez et al., 2014). Advocates of integrated STEM education argue that it is essential to teach STEM across disciplines in an integrated approach so that students and teachers develop an awareness of how to make connections between real-world problems and improve learning (NRC., 2014;Subramanian & Clark, 2016).An interdisciplinary integration requires focusing on a real-world problem and using 21st century skills (i.e., critical thinking, problem-solving, and content knowledge) to solve a problem (Wang et al., 2011).This approach is crucial because it helps students build the skills needed to understand our ever-changing world and obtain skills needed for the STEM workforce (Smith et al., 2015). Students' interest in STEM increases as their curiosity, exploration, and understanding of how integrated STEM concepts are connected and impact the world around them (English, 2016).DeJarnette (2012) suggested the need to expose students to STEM education at an early age.An integrated STEM approach in schools may be one way for students to learn about global real-world issues and create interdisciplinary solutions that address complex problems (English, 2016;NRC, 2014;Sanders, 2008;Tsupros et al., 2009).Additionally, Sanders and Wells (2005) posit that intentionally integrated concepts and practices of SteM and applications of technical/engineering design-based learning can result in a trick-down effect where other content areas integrate STEM into instruction. One of the most incredible benefits of providing an integrated STEM curriculum to students is acquiring 21st century skills (P21, 2015) and engineering habits of mind (Loveland & Dunn, 2014).The National Science Teaching Association (NSTA), in conjunction with the Partnership for 21st Century Learning (P21) and National Research Council (NRC), includes "life and career skills; adaptability; complex communication/social skills; nonroutine problem solving; self-management/self-development; and systems thinking" as part of 21st century skills (NSTA, 2011, p. 2).Complex communication and social skills include effective collaboration with peers that builds STEM knowledge, promotes discourse with different viewpoints, and comes to a consensus when working together (Loveland & Dunn, 2014).Systems thinking, an underrepresented skill within classrooms (Salado et al., 2019), allows students to make connections across STEM disciplines and is a natural way to apply engineering habits of mind (Lippard et al., 2019).Six engineering habits of mind (i.e., systems thinking, creativity, optimism, collaboration, communication, and attention to ethical considerations) encompass many 21st century skills (Loveland & Dunn, 2014;National Academy of Engineering & National Research Council, 2020).Thus, including these skills and habits of mind fosters students' STEM knowledge and understanding across disciplines (English, 2021). Integrating Engineering in STEM With increasing STEM education awareness, educators understand the importance of hands-on activities and their role in providing the necessary foundation for STEM learning and achievement (Margot & Kettler, 2019).Many schools are beginning to stress the integration of all areas of STEM using the engineering design process (EDP).The EDP is the practice of "testing the most promising solutions and modifying what is proposed based on the test results leads to greater refinement and ultimately an optimal solution" (NRC, 2012, p. 210).In some instances, critical concepts of the EDP are applied in math and science courses.According to English (2019), this will provide students with foundational problem-solving skills in engineering design that are transferable and applicable across STEM disciplines and fields.According to Capraro et al. (2013), there are many engineering design models schools can adopt and follow. The K-12 engineering framework expects educators to focus on design and problem solving while incorporating STEM concepts (Strimel, 2012).Applying STEM processes and practices such as the EDP by educators and students J. of Res. in Sci. Math. and Tech. Edu.\| 199 helps create a STEM mindset for students (Peters-Burton et al., 2019).In addition, educators should "promote engineering habits of mind" (Berland, 2013, p. 22) because they incorporate several 21st century skills (P21, 2015). Similarly, the K-12 science education standards incorporate STEM (mainly science) into engineering design challenges.Engineering allows students to visualize and apply complex concepts relevant to them and our society (Morgan et al., 2013).An interdisciplinary approach and brainstorming of possible solutions allow students to increase their creativity and ingenuity as engineers do in the real world (Gormley & Boland, 2017;Marcos-Jorquera et al., 2016).Furthermore, modeling a "think tank" environment to "learn and adapt to innovate solutions to new problems" provides students with the opportunity to demonstrate their ability to work together like engineering teams (Larson et al., 2017, p. 2).Incorporating a STEM-driven mindset can provide a shared vision and expectation for learning that promotes a STEM-driven school culture (Waters & Orange, 2022).This culture includes engineering habits and applications that could be essential to instill in students at an early age. According to Berland (2013), incorporating engineering into science is a great idea, but teachers' pedagogical methods and classroom philosophy may not be conducive to integrating the EDP.Classrooms working on the EDP should apply their multidisciplinary approach to learning through solving problems with a focus on engineering objectives rather than simple math and science knowledge and skills (Margot & Kettler, 2019).Additionally, engineering education can be used in science classrooms to accentuate engineering habits of the mind that are important in the engineering process and applicable across STEM fields (Lippard et al., 2019;Schnittka, 2012), and most science classrooms only focus on science curricula.A significant difference to note is that scientific inquiry focuses on gathering empirical data which supports a hypothesis and explains why something is occurring.In contrast, the EDP has students create a project or solve a problem using constraints and specifications (Nadelson et al., 2011).Teachers who incorporated the EDP into their professional practice and classroom experiences often included their students in their successes and failures as they worked through classroom experiences with them."This visible failure, she [the teacher] made clear, served as evidence-for themselves and the students-of their ongoing pursuit of new ideas and continuous improvement" (Peters-Burton et al., 2019, p. 454).Wendell et al. (2017) support the importance of learning from failure in the classroom but believe it needs to be intentionally scaffolded into students' EDP experiences so they understand the critical role reflections play in the EDP. The International Technology and Engineering Educators Association (ITEEA) created the Engineering by Design (EBD) program to integrate science, technology, and mathematics into the engineering design process to better prepare today's students for future advancements in technology and engineering (Strimel, 2012).During the EDP, students can collaborate while increasing their critical thinking, creativity, and communication skills while solving real-world problems.The EDP emphasizes that all activities (i.e., researching, calculations, budgeting, creating, and testing) should be contextualized around the design challenge (Strimel, 2012).For example, many engineers must work under specific constraints, such as budget constraints, limiting the design by eliminating different possibilities (Dym et al., 2009).Constraints, however, are usually limited to understanding budgets, time, and materials, rather than real-world problems associated with social and political challenges (Roehrig et al., 2021). Additionally, EDP challenges should include real-world connections.Maiorca and Mohr-Schroeder (2020) discovered that design challenges such as the Marshmallow Challenge may engage students in EDP but lack an authentic learning experience.According to Kapur and Bielaczyc (2012), teachers who promote this culture also challenge students to be in a state of productive struggle where they "are challenged yet not frustrated and remain sufficiently engaged in problem-solving" (p.50).Productive failure is a critical element of the EDP.English (2021) echoes this need but emphasizes the importance "to increase awareness of, and capitalise [sic] on, children's skills as independent problemsolvers, who relish challenges, preserver in the face of failure, and learn from both what "works" and what does not" (p.117).The goal is to engross students in such a significant way that they learn the engineering design process while creating authentic learning experiences that increase their critical thinking, STEM knowledge, and ability to connect to real-world innovations (Maiorca & Mohr-Schroeder, 2020).Therefore, engineering is an excellent way to integrate STEM concepts (NASEM, 2020). Methods This exploratory case study (Creswell, 2007;Lichtman, 2010) examined the integration of a STEM program in an elementary-level school.Gemini Elementary STEM School (GES) is a pseudonym for a suburban elementary school in Texas.The researcher selected it as the site for this study because it implemented a STEM program known as the STEM Block that included an engineering laboratory.A case study was an appropriate qualitative framework to employ because selecting one program within the school to study allowed the researcher an in-depth exploration of how GES integrated STEM with attention to the site context and educators' perceptions that framed the program (Lichtman, 2010).The study addressed the following research question: What features of an elementary-level integrated STEM program have the potential to contribute to its effectiveness? Participants Forty-four GES K-5 STEM educators were asked to participate in this study based on their role as a school leader or teacher and their knowledge of integrating the STEM curriculum into students' STEM Block experiences.Participants volunteered in the study's K-5 STEM Educators' Perception Survey (n=17), focus groups (n=16), and interviews (n=4) and did not object to the use of a pseudonym to protect participants' anonymity.All GES educators were invited to participate in the focus group sessions.Due to time constraints and school commitments, 16 chose to participate.Table 1 presents the demographics of the focus group and interview participants.The teachers, STEM content specialist, and librarian are female, and the principal is male.This data allowed the researcher to cross-reference the survey data with the focus group and interview data to identify participants who attended the focus groups and interviews and completed the survey. Survey The researcher created the K-5 Educators' Perception Survey to gather preliminary data regarding educators' perceptions of STEM education and the components of a successful STEM school.The anonymous online survey consisted of a cover letter containing implied consent by participants taking the survey, 15 open-ended items, four dichotomous items (three yes/no items and one gender), and ten nominal items, which included demographic questions and took an average of 14 minutes to complete.At the principal's request, the survey was emailed to the STEM content specialist at the site, where she forwarded the email containing the cover letter and link to the survey to 44 staff members.For this paper, the survey items the researcher focused on were how teachers perceive how planning with team members impacts STEM Block lessons, using the EDP in the engineering laboratory, and effective practices of a STEM school. Focus Groups After reviewing the survey data, the researcher wrote the semi-structured focus group questions to provide a more thorough description of STEM integration and components of a successful STEM school.The researcher emailed the STEM specialist a copy of the focus group questions to share, and participants received a paper copy during the session.There were seven focus groups with 16 participants.Focus groups were offered throughout the day at 30minute intervals.This allowed the teachers to attend during their conference or lunch periods.The principal and specialist encouraged teachers to attend during their conference period. Interviews The researcher conducted semi-structured interviews lasting 20-40 minutes each.The researcher interviewed the principal on the phone and a STEM specialist and librarian on site.A copy of the interview questions was emailed before the interview and focused on educators' perceptions of the elements of a successful STEM school.After the initial focus group and observational data were collected, a follow-up interview was conducted with a second-and third-grade teacher to provide a deeper insight into the engineering laboratory. Observations The STEM specialist offered the researcher to observe the students' EDP experience.The researcher thought this would be a great way to observe and compare the program to the previous data collected.Therefore, the researcher agreed to observe the fourth-grade EDP STEM Block on Cargo Boats for four consecutive days.The observations lasted approximately 60-90 minutes in duration.The researcher took notes in her field journal and pictures in the engineering laboratory to document how students work through the Cargo Boats EDP during their STEM Block week. In addition, the researcher set up two flip video cameras on the first two days of the observations so she could check her notes in her field journal. Data Analysis Before importing data into QSR International's NVivo10 qualitative data analysis software, the researcher examined the data to explore possible trends between the transcribed and observational data.Queries were run to detect highfrequency words (e.g., STEM, engineering, science, students, technology, curriculum, teachers, technology, and math), which provided a synopsis of the overall data.Since a more in-depth analysis was necessary to understand participants' perceptions fully, text search queries were run to look for potential relationships between topics involving the integration of STEM content.The researcher ran variations of the text queries (i.e., integrate, integrating, and integration) to examine different aspects of the same concepts.Figure 1 Common trends in the data were looked for when coding was complete.The trends were explored to see where the codes overlapped and could be combined into themes.The analyses resulted in the themes presented in the following section. Findings Based on the survey, interview, focus group, and observational data, seven common themes emerged from the more extensive study of K-5 STEM educators' perceptions of STEM education: instructional leadership team, professional development, teacher collaboration, making connections, vision and culture, 21st century skills, and integration of the engineering laboratory.Based on the intertwined themes, we collapsed these initial themes into three major themes: (a) critical role of the specialist, (b) instructional design, and (b) integration of the engineering laboratory for this paper. Critical Role of the Specialist According to participants, the primary integrative curriculum at GES occurs during the "STEM Block," an engineering laboratory for all grade levels.Qualitative data underscored that the support provided by the GES STEM specialist fostered a robust instructional design foundation for teachers.During year one of implementation, the specialist created and designed the STEM Block engineering design lessons used across all grade levels.Additionally, she facilitated the EDP experiences for students and modeled best teaching practices for teachers in the engineering laboratory.In subsequent years, the STEM specialist encouraged teachers to co-teach the EDP classes.As a result, teachers were more comfortable and willing to learn EDP lessons, integrate STEM into daily instruction, and begin leading the EDP classes in the third and fourth years of implementation. The vision of GES focuses on preparing, inspiring, and creating lifelong learners who can use critical thinking skills to address future global challenges.The "STEM Block" at GES is the primary integrative curriculum, according to the survey, focus group, and interview data, which is an engineering laboratory for all grade levels… indicated support provided by the GES STEM specialist provided the instructional design foundation for teachers to feel more comfortable integrating STEM and learning the EDP lessons.Exploration of K-5 STEM educators' perceptions of critical components that made GES a successful STEM school revealed that team planning is essential to creating an effective EDP instructional design for the STEM Block lessons.One first-grade teacher felt the STEM specialist's role was vital to the success of their school.She believed the specialist supported her and was helpful when she stated: our teachers, making sure they understand how to go through the lessons.And so that was a key component, I think, is really helping out those teachers that are more uncomfortable with it or not used to it and taking them through it.So that was probably the biggest parts. The principal perceived that the STEM specialist was a key component of successful instructional design and implementation of STEM curricula in their school.In addition, it was a constant reminder of the goals that encouraged teachers to achieve confidence and build their STEM content knowledge and skills by integrating the EDP. Most of the teachers attribute their initial successes with STEM integration to the implementation of the EDP provided by the STEM specialist.Based on teachers' comments, the specialist trained the teachers on how to facilitate the engineering laboratory lessons and incorporate the EDP into the school's identity and culture."She's pushing the kids to go beyond the boundaries, especially during that design part," the second-grade teacher shared.She also noted that the teachers in the engineering laboratory are watching and learning as the specialist models how to question the students to get them to develop their ideas at a higher level. During the initial phases of STEM education implementation, the specialist trained teachers on the engineering lessons and instructed students in the engineering laboratory during the summer and start of school prior to implementation. During the implementation year, one second-grade teacher, now a third-grade teacher, recalled how the specialist took the burden of designing EDP lessons off the teachers.She recalled what many teachers believed: [The STEM specialist] was the one kind of in charge.There was discussion about it [designing plant packages EDP] with the second Grade [teachers] but for the most part since we were at [year] zero, we had input but it was more about we need to follow our curriculum because that's how we're gonna get our grades, that's how we're gonna get our learning, our [standards] met. The specialist taught the lessons to the students and modeled the facilitation of the lessons for the teachers.It appeared to the researcher that the teachers appreciated the help from the specialist to help them understand the engineering process and integrate STEM into their classrooms.Furthermore, the teachers initially depended on the specialist to learn how to implement the integration of the STEM Block EDP lessons. The specialist's role has transitioned from being the primary facilitator of the EDP for the students to taking a step back and allowing the teachers to facilitate the engineering laboratory.For example, a second-grade teacher said the specialist is in the laboratory two to three times a week.She also stated how teachers are encouraged to begin developing and modifying lessons for the EDP.She recalled: Now that we've been through the process so many years, she kind of puts it on us and just kind of is more of a guide for us, and we'll run things past her, like when I started doing the PowerPoints, I'd run it past her. Since this is the fourth year of implementation, teachers are taking a more significant role in the engineering laboratory.According to the second-grade teacher, the specialist will now guide teachers and ask them how they can modify their ideas to "make it even more of a challenge.""There's still modeling going on, but a lot of us have been through the process so many times, it just kind of becomes natural after you go in there so much," the second-grade teacher concluded.The principal has set aside professional development opportunities at the beginning and throughout the school year.In addition to a teachers' planning period, he provided a half-day of planning each nine-week grading period to allow teachers to collaborate on the EDP during the engineering week. Instructional Design STEM Block lessons were initially developed by the STEM specialist and teachers with numerous years of experience. Teachers' collaboration and methodical planning during the instructional design of the engineering lessons were identified as an essential component of the STEM Block and engineering laboratory.As teachers acquire STEM content, pedagogical knowledge, and skills, they can collaborate effectively, improve the instructional designs of current EDP lessons, and create new ones.The STEM Block lessons (Table 2) for K-5 grade levels were developed to cover a 3-5-day engineering unit using the engineering design process.According to the original program proposal, the purpose of these lessons is to incorporate the EPD into a STEM block every nine weeks for K-5 students.The proposal also stated that language arts, mathematics, and social studies would be integrated into each grade level's different units of study when applicable. As part of the implementation of the engineering laboratory, teachers attended schoolwide professional development Teachers began contributing more to the lessons' instructional design as they gained STEM content knowledge. According to most of the teachers and the specialist interviewed, the principal set aside half-day planning every nine weeks for teachers to plan and review the engineering lesson.According to the interviewees, planning focused on the engineering lessons and was in addition to team planning and PLCs.A second-grade teacher shared that these meetings involved teachers in the same grade level meeting to identify the state standards for that nine-week timeframe.During this time, teachers collaborated to find a critical concept that could be applied to solve a real-world problem and draw arrows connecting the cross-curricular concepts, not just STEM.According to the specialist and second-grade teacher, teachers select a topic within the area with the most connections, usually within science.Then, they begin the daily planning the STEM Block unit (Figure 2). Previous themes are supported when teachers obtain new knowledge and ideas from conferences-they share the information with colleagues and begin collaborating on ways to adapt it for their curriculum.For example, a secondgrade teacher recalled how one of her colleagues attended a conference and won a lab about Fulton's ferry.She remembered her colleague saying, "Well, this would fit with what we're doing right now," and recalled that her team decided, "Well, let's develop this into a STEM lesson."Therefore, they modified the lesson to meet the needs of their STEM students and made connections to the TEKS and standards during their cross-curricular team planning.As a result, increasing teachers' confidence and STEM pedagogical knowledge led to improving the EDP instructional designs. Figure 2 Example of Daily Planning for the Third-grade STEM Block Integration of the Engineering Laboratory The collective analysis of the data supported that the GES engineering laboratory is a primary component of the walls, referred to by educators, and used as anchors for student thinking and action.Throughout the weeklong laboratory, students were prompted to refer to these posters to help them through the EDP.The engineering laboratory facilitators prompted students to identify a problem and its possible solutions during the "Ask" phase.During the "Imagine" phase, students identified the possibilities, explored options, and agreed on the best solutions for their problems.Students created a "Plan" and determined the necessary materials to build their cargo boat.Next, students created a model and evaluated it to decide if they followed their plan and met their goal.Students tested their model at the end of the "Create" phase.During the "Improve" phase, students analyzed and appraised their model to assess whether or not it worked, explored what they could have done differently, and considered what would make it better. In the final phase, "Communicate," students discussed if changes were needed, listened to feedback from others, and determined whether the problem was solved.The laboratory experience appears to culminate in the necessary knowledge and skills students need to be successful engineers and provides a safe atmosphere where the EDP encourages students to make mistakes, learn, and redesign. The specialist proudly shared, "The engineering lab has been key to the success" of Gemini Elementary, citing the STEM Block as a critical component of their success.Several teachers shared their beliefs that students demonstrated a shift in how they think and act due to participating in the engineering lab.One third-grade teacher noted, " [The] engineering lab like I said, it creates that mindset" for students to become an engineer.Students often work in groups of two to three.Students are expected to follow the EDP and communicate their results at the end of each STEM block.The following section describes how fourth-grade students worked through an EDP on buoyancy, offering an example of how the engineering lab is implemented at Gemini Elementary. Cargo Boats EDP According to the focus group data, the Science State of Texas Assessments of Academic Readiness (STAAR) Grade 5 data indicated that students struggled with the concept of buoyancy.Therefore, fourth-grade teachers chose the cargo boats EDP, which integrated social studies and science concepts.Science was the dominant content area used to develop this concept.Still, the inclusion of social studies helped deepen students' learning of the history of shipbuilding and how buoyancy was an important scientific concept that aids in thriving shipbuilding.The researcher observed the introduction of the cargo boat EDP for a fourth-grade class and all phases of the engineering of the cargo boat, except the "Final Testing" and "Communicate" phases.The STEM specialist described the final testing and communication phase experience was described to the researcher by the STEM specialist after completing the fourthgrade engineering laboratory.The teacher, the STEM specialist, and the engineering aide facilitated the project.The fourth-grade science teacher was absent during the Tuesday observation, so a fifth-grade teacher taught the engineering laboratory to the students.The fifth-grade teacher had prior knowledge about the cargo boat EDP but was asked to be a substitute in the engineering laboratory that morning.Therefore, the STEM specialist interjected and often took the lead in facilitating this project during the Tuesday observation.Additionally, there was an engineering laboratory aide who walked around to help students if they had questions.This section will explain how the teacher facilitates the EDP as students work through the EDP during the STEM Block. Before beginning the EDP, the researcher observed the teacher begin the lesson with a PowerPoint titled "Ships for Exploration."During the presentation, students discussed how explorers would cross the Atlantic Ocean during the 1500s (Figure 3).Toward the end of the Ships for Exploration lesson, the researcher noticed the lesson shift from teacher-driven to student-driven EDP.The fifth-grade teacher and STEM specialist, who substituted for the fourth-grade science teacher, explained the project together.Both appeared familiar with the students and the project.Therefore, they could fully describe the expectations and facilitate students' ability to think and act like engineers.The teacher asked, "Can you design a ship that can carry a crew and cargo and withstand wind, rain, and waves in the Atlantic Ocean?"During this moment, the researcher noticed that students appeared to think about the problem and were shifting their mindsets to become cargo boat engineers.Students were then prompted to write the "Ask" question in their science journal (Figure 5).Next, detailed information aided the "Imagine" and "Plan" phases.The teacher showed students the materials list (Figure 6) for their cargo boat build and explained that the king allotted them a $10.00 budget.The teacher provided time for students to ask questions about the type and number of materials.According to the STEM specialist, question time reflects the brainstorming engineers do at the start of a project as they consider realistic options.The STEM specialist explained that the budget was for the prototype and the final design.She continued explaining that the boat should not sink, and they would "take a fan and blow on it" to simulate wind and test their design.Rainstorms would be simulated using a watering can, and typhoons/hurricanes would be simulated by shaking the poolside extremely hard."[There will be] a massive rainstorm that will see if your boat can make it across the sea. This is what our explorers did," the specialist enthusiastically explained as she connected social studies and science. Cargo Boats Material List Before an engineer designs a project, the specialist explains that the questions during the brainstorming session are used to imagine the design and weigh alternatives.During the next phase of the EDP, students were provided additional details and asked to "Imagine" what their boat needed to look like when traveling across the Atlantic Ocean (Figure 7).The researcher observed students sketching their ideas in their science journals as they were encouraged by the specialist not to worry about their budget constraints.The sketch had to be large and labeled with materials. Students were encouraged to look at the materials list.The specialist reminded students that the ship must carry a crew and cargo.The crew consisted of six miniature teddy bears and a cargo of 40 grams of clay.Students had five minutes to complete the "Imagine" portion individually (Figure 8)."Individual ideas only.No talking; by-yourself time," the specialist reminded students.Students continued transitioning from the teacher-centered instruction to a more studentcentered "Plan" phase where the design was collaboratively constructed.After sharing, students combined both designs and ideas and created a "Plan" with a new sketch on a clean sheet of paper in their science journals.The researcher observed groups receiving an approval stamp from either the teacher, specialist, or engineering aide after finishing the new design and completing the budget sheet individually (Figure 9). If needed, students were directed to the mathematics anchor charts in the back of the laboratory (Figure 10).The facilitators reminded them, "The king has told us you have $10.00 for this boat [so] stay within your budget."GES students acted as engineers and used specific constraints such as budget constraints that could impact their design. Completing their Cargo Boat EDP within the proper constraints provided students with a real-world application of correct budgeting and managing materials for their project. Figure 10 Adding and Subtracting Decimals Review Anchor Chart The researcher observed that the "Create" phase began when students received approval for their "Plan" with the new design and budget (Figure 11).The researcher saw students beginning to review and collect materials as they began the "Create" phase (Figure 12).Using duct tape was permitted to build the cargo boat; however, students independently budgeted and measured the appropriate amount.Using their design plan, the researcher observed students work collaboratively to build their cargo boats (Figure 12). They regularly communicated about the build's plan and reality as a check-and-balance and to ensure accuracy. According to the teacher and specialist, students' final products should match their approved designs (Figure 13).This phase marked an apparent move to a student-driven environment with students thinking and acting more like engineers. Figure 13 EDP: Students Create Their Cargo Boats Figure 14 EDP: Approved Plan and Built Cargo Boat Comparison Students tested their cargo boats in a small plastic pool (Figure 15).As stated, the teacher provided "extreme weather conditions," utilizing a fan for wind, a watering can for rainstorms, and shaking the side of the pool for typhoons and hurricane waves.During testing, the teacher and STEM specialist prompted students to think about how they could improve their designs by analyzing what worked and did not.After testing, students were provided time to "Improve" and redesign their cargo boats and retest them.During the trials, students collected and placed data into an Excel spreadsheet. Finally, students shared the results of their designs.The researcher did not observe students during the "Communicate" phase; the STEM specialist provided information on what students completed during this phase and the grading rubric for students' presentations.She said the students wrote a persuasive letter to the king asking him to fund their expedition, which included the ship's design and how well it held up to severe weather conditions, thus allowing it to cross the Atlantic Ocean.The final presentation included a skit in which they presented their cargo ship to the king (Figure 16).Students used the Oral Presentation Rubric: Presentation to the King (Figure 17) to guide them. Discussions STEM Specialist One of the most important findings of this study uncovered the importance of the role of the STEM specialist when developing and implementing the instructional design of the engineering laboratory at GES school.Educator comments revealed that the STEM specialist assumed the role of the EDP's lead facilitator in the engineering laboratory during the initial implementation.Unlike the study by Hammack and Ivey (2019), which concluded that teachers would not implement engineering concepts into their classroom due to a lack of knowledge, this study indicates that teachers acquired a deeper understanding of the STEM content and pedagogical knowledge and skills needed to be a successful facilitator of EDP by participating in team planning and watching the STEM specialist model lessons.Since it is challenging for students to connect STEM content areas and real-world applications (Breiner et al., 2012), teachers should learn how to facilitate EDP to foster these explicit connections.AT GES, this was done by the specialist modeling best practices while she taught the EDP lessons to the students and inviting the teachers to begin co-teaching EDP lessons.This gives teachers an understanding of how students can acquire 21st century skills (P21, 2015).The content specialist can exemplify how to provide students with an in-depth learning experience that could spark students' sense of curiosity and interest.A hands-on approach helps educators understand the content and could give students a more incredible foundation in STEM concepts (Margot & Kettler, 2019) and possibly increase their interest in STEM careers (Tanenbaum, 2016).Providing teachers with modeling opportunities could guide and increase their STEM content and pedagogical knowledge and skills when facilitating an EDP.As teachers' confidence increases during the facilitation of the engineering week, then they may feel more comfortable implementing EDP strategies in their classroom. Team Planning Another significant layer of instructional design included the importance of teacher collaboration during team planning of the EDP lessons.Teachers could share new ideas and help colleagues connect across cross-curricular content areas with increased practice.This essential part could help teachers expand their thinking of STEM fields and ways to include modern real-world problems (NSTA, 2021).Therefore, team planning could be an essential time where teachers share their ideas and collaborate on ways to improve their lessons, integrate modern STEM concepts, make explicit connections to STEM in classroom instruction, and create a budget for the different EDP.Dym et al. (2009) support the integration of different STEM concepts, including budget constraints.Since students fail to understand STEM interdisciplinary connections (Breiner et al., 2012), educators need to discuss them during instructional design and planning.Experienced teachers and content specialists could explain and model new teachers' classroom content and EDP lessons.In addition, STEM integration was not limited to science, mathematics, engineering, and technology but included English and language arts, social studies, and fine arts.This supports the notion by Sanders and Wells (2005) that the intentional integration of STEM areas can lead to the connections of STEM into the instruction of non-STEM classes.The purposeful collaboration helped teachers to be consistent across grade levels and cross-curricular content teams.These collaboration opportunities are often overlooked due to the stress elementary teachers may feel from teaching multiple disciplines and high-stakes testing (Hammack & Ivy, 2019).Therefore, the additional time allotted to teachers during planning is essential for the success of the STEM program. STEM Block In addition to instructional design, this study indicated that the engineering laboratory is the root of the successful implementation of STEM at GES.The primary tool for integrating STEM education was the STEM Block, which included a weeklong EDP where students attended and studied in the engineering laboratory.At GES, the STEM Block provided students with a safe place to act like engineers to brainstorm different engineering designs and solutions.Gormley and Boland (2017) concluded that using an interdisciplinary approach that allows students to brainstorm solutions promotes creativity and ingenuity.Using an integrated approach to teaching STEM supports students' ability to make connections, improving their learning using the EDP (NRC, 2014). Furthermore, Lippard et al. (2019) agree that EDP is a natural way to apply engineering habits of mind.These support this study's findings that all the educators who participated agreed that the engineering laboratory was the most J. of Res. in Sci.Math.and Tech.Edu.\| 219 effective way to integrate all aspects of STEM into their elementary school.The engineering laboratory could create an environment that allowed for the teaching of the engineering processes and was an encouraging space for students to possibly shift their mindsets and become engineers.This environment may foster students' productive struggle when faced with challenges and supports an engineering culture and framework (Kapur & Bielaczyc, 2012). Creating an Engineering Culture The findings indicated that implementing integrated STEM curricula into an elementary school benefited the students and educators.GES's vision of creating a STEM elementary school where students were encouraged to make mistakes, learn, and redesign, created a school culture that embraced failure.Peters-Burton et al. ( 2019) supported the importance of failure as part of the learning practice for teachers and students.Wendell et al. (2017) noted that, when properly scaffolded, reflective decision-making is crucial when participating in engineering design.Waters and Orange (2022) expands on this notion to include adopting a STEM-driven school culture encompassing engineering aspects and habits of mind.A STEM-driven mindset could be an essential aspect of including not only during an engineering STEM Block but also throughout daily activities in classrooms.Encouraging students to reflect on why their design was ineffective and prompt them to improve, redesign, and retest their design could be foundational to supporting this STEM-driven school culture.This instills a culture that promotes engineering habits of mind (Loveland & Dunn, 2014).This culture and approach could provide a safe environment that accepts failure as a natural part of learning and where students' interest and engagement could increase to the integrative nature (English, 2016). Maiorca and Mohr-Schrodder (2020) posited that creating authentic learning experiences that provide real-world connections and applications of 21st century skills enables students to be immersed in an engineering environment. English (2021) also believed that these 21st century engineering skills solidify students' STEM knowledge and application.Furthermore, collaboration and communication among students with adults and peers can be enhanced during stages of the EDP.Students could improve their communication skills as they work through different steps in the research process.In addition, students should be encouraged to find ways to solve problems and collaborate with others in the classroom and engineering laboratory.These are essential 21st century skills that are foundational and transferable to multiple areas across the STEM fields (English, 2019;P21, 2015). Implications This study also indicated that it is essential for principals to invest in personnel, a STEM content specialist, and resources.Furthermore, the STEM specialist must develop teachers' STEM content and pedagogical knowledge and shift their thinking from co-facilitating an EDP to facilitating the EDP during the engineering week.This shift could help build teachers' confidence and skill set.Providing teachers with adequate time throughout the school year to network and collaborate with other STEM educators during planning days and other professional opportunities should be considered.This could support the STEM school culture where students and teachers believe it is acceptable to make mistakes, learn, and redesign (Waters & Orange, 2022).This mindset creates the idea that learning in the classroom and the engineering laboratory after mistakes are made is then enhanced by redesigning and reworking ideas and providing students and teachers a safe environment.Mindset-related implications often influence how students and teachers are willing to take risks and try new designs and ideas.Thus, it is suggested that school leaders, especially principals, create an environment that embraces a STEM-driven school culture (Waters & Orange, 2022) that empowers teachers, promotes innovation and sharing of ideas, and instills engineering habits of mind into students and faculty. Conclusion With a deficit of literature regarding STEM initiatives and case studies (Lesseig et al., 2019;Peters-Burton et al., 2019), this study provides insight into effective practices for integrating STEM into an elementary school.The role of the STEM special was critical to the development of the initial STEM Block lessons and the teachers' STEM content and pedagogical knowledge.Based on educator comments and researcher observations, the engineering laboratory appeared to be a natural integrator for STEM concepts and a key component of Gemini Elementary's perceived success.It created an environment for teachers to develop and hone their STEM content and pedagogical knowledge by watching and co-teaching with the STEM specialist.Additionally, it provided an engaging learning environment for students to utilize and improve their 21st century skills as they enthusiastically took the role of a ship designer. Furthermore, the engineering laboratory was a dedicated setting for the purposeful STEM integration for every student from all grade levels to attend and practice vital engineering habits of mind. Limitations This school was purposely selected due to the implementation of integrated STEM curricula and tried to identify critical components of the school's success.With the study conducted at a STEM school, the participants might have displayed possible bias that would showcase their STEM elementary school as being successful.This might have been due to the initial phrasing of the survey question asking what key components of a successful STEM school were. Figure 1 Figure 1 that taught them how to integrate STEM.The principal also allocated specific planning time before the beginning of the 2014-2015 school year, where teachers focused on learning and reviewing the EDP lessons.The specialist attended these planning sessions to discuss and model the lessons.The teachers attend STEM-specific professional development opportunities and training each subsequent year to help them integrate STEM education. school's implementation of STEM curricula.The students in the laboratory applied science, technology, engineering, and mathematics concepts to solve real-world problems that also reflect parts of the curriculum for one or more content areas.GES students attend four one-week-long engineering laboratory sessions per year.Each occurred during the nine-week grading period during the regularly scheduled science class.Typically, about 50 minutes are devoted daily to science.However, during the STEM Block, the time may fluctuate to accommodate the laboratory's needs.The STEM specialist carefully designed the instruction to build on what is learned each week during an academic year.Each year provided a layer of authentic knowledge and skills necessary for subsequent yearly projects.By the time students complete fifth Grade, students appear to acquire an engineering mindset.The principal concluded, "If they've gone through our building from K [kindergarten] to fifth grade, they'll have gone through the 24 sessions, and you can see them emerge as with that engineering thing, understanding the engineering process."The STEM specialist trained teachers on the STEM lessons, including modeling how to facilitate student learning, behavior, and thinking like an engineer.Additionally, she worked collaboratively with teachers when facilitating and co-facilitating students in the laboratory.Engineering laboratory lessons were taught and facilitated by the grade-level science teacher, STEM specialist, and the engineering laboratory aide, who has worked part-time in this role for three years, helping students as needed.Students worked through different parts of the EDP (e.g., ask, imagine, plan, create, improve, and communicate) throughout the STEM Block week.Posters explaining the process for each engineering phase were posted on the J. of Res.inSci.Math.and Tech.Edu.\| 209 Figure 3 Engage Figure 3Engage Lesson on Exploration in the 1500s Figure 4 Figure 4Diagram Explaining Parts of a Ship Figure 5 EDP: Figure 5 EDP: Student Example of the Ask Phase Figure 7 EDP: Figure 7 EDP: Additional Details Provided in the Imagine Phase Figure 8 EDP: Figure 8 EDP: Student Example of the Imagine Phase Figure 11 EDP: Figure 11 EDP: Approved Plan and Budget Sheet Figure 12 EDP: Figure 12 EDP: Students Review and Measure Materials to Create Figure 15 EDP: Testing Extreme Weather Conditions Table 1 Focus Group and Interview Participant Demographics shows results from text queries referring to participants' different responses.J. of Res. in Sci.Math.andTech.Edu.\|203Additionally, the researcher coded the data according to the most common words (i.e., professional development, planning, time, engineering lab, key components, culture) and then used NVivo to auto-code that data.Common codes comprised professional development, exploratory phase, team planning, makerspace, communication, time, collaboration, STEM education approach, critical thinking, creativity, problem-solving, engineering, future improvement, real-world experiences, connections, and integration.They were a part of the comprehensive study.The coded data regarding the engineering laboratory was compared to the researcher's field notes and observational data. Table 2 STEM Block Examples for K-5	2022-08-21T15:07:59.206Z	2022-09-15T00:00:00.000	{ "year": 2022, "sha1": "ac701595cbee68b10ad2d491e44466b9807bd344", "oa_license": "CCBY", "oa_url": "https://jrsmte.com/download/exploring-effective-practices-of-an-elementary-stem-block-program-12383.pdf", "oa_status": "GOLD", "pdf_src": "ScienceParsePlus", "pdf_hash": "c6337089d70ce5cff98553d3369d7a7c7fd6ee93", "s2fieldsofstudy": [ "Education" ], "extfieldsofstudy": [] }
245041182	pes2o/s2orc	v3-fos-license	Clinical significance of signet ring cells in surgical esophageal and esophagogastric junction adenocarcinoma: A systematic review and meta-analysis BACKGROUND The clinical significance of signet ring cells (SRCs) in surgical esophageal and esophagogastric junction adenocarcinoma (EEGJA) remains unclear now. AIM To explore the association between the presence of SRCs and the clinicopathological and prognostic characteristics in surgical EEGJA patients by combining and analyzing relevant studies. METHODS The PubMed, Web of Science, and EMBASE electronic databases were searched for the relevant literature up to March 28, 2021. The relative risk (RR) with 95% confidence interval (CI) was calculated to assess the relationship between SRCs and clinicopathological parameters of surgical EEGJA patients, and the hazard ratio (HR) with 95%CI was calculated to explore the impact of SRC on the prognosis. All statistical analyses were conducted with STATA 12.0 software. RESULTS A total of ten articles were included, involving 30322 EEGJA patients. The pooled results indicated that the presence of SRCs was significantly associated with tumor location (RR: 0.76, 95%CI: 0.61-0.96, P = 0.022; I2 = 49.4%, P = 0.160) and tumor-node-metastasis stage (RR: 1.30, 95%CI: 1.02-1.65, P = 0.031; I2 = 73.1%, P = 0.002). Meanwhile, the presence of SRCs in surgical EEGJA patients predicted a poor overall survival (HR: 1.36, 95%CI: 1.12-1.65, P = 0.002; I2 = 85.7%, P < 0.001) and disease-specific survival (HR: 1.86, 95%CI: 1.55-2.25, P < 0.001; I2 = 63.1%, P = 0.043). CONCLUSION The presence of SRCs is related with advanced tumor stage and poor prognosis and could serve as a reliable and effective parameter for the prediction of postoperative survival and formulation of therapy strategy in EEGJA patients. However, more high-quality studies are still needed to verify the above findings. INTRODUCTION Esophageal carcinoma is the eighth most common tumor worldwide with an estimated 456000 new cases in 2012 and an increasing incidence has been observed in recent decades (7.9/100000 in males and 1.4/100000 in females) [1][2][3]. Although great progress has been made in the surgical and adjuvant therapy of esophageal cancer in recent years, the survival of esophageal cancer patients remains poor due to the advanced stage at the time of diagnosis [4]. In Western countries, adenocarcinoma is the most common pathological subtype of esophageal cancer, although squamous cell carcinoma accounts for the highest proportion in Asian countries [5]. Signet ring cell (SRC) carcinoma is a rare mucinous subtype of adenocarcinoma that has been reported to be related with aggressive biology in gastrointestinal cancer [6,7]. Actually, the clinical significance of SRCs in gastric and colorectal carcinomas has been widely verified. Nie et al [8] included 19 studies involving 35947 cases and demonstrated that gastric carcinoma patients with SRCs tended to be younger (weighted mean difference = -3.88, P = 0.001) and predominantly female [odds ratio (OR): 1.60, P < 0.001]. Besides, early-stage gastric cancer patients with SRCs were related with a better overall survival (OS) [hazard ratio (HR): 0.57, P = 0.002], but advanced stage patients with SRCs were related with a worse prognosis (HR: 1.17, P < 0.001); however, in the total population, no significant difference in OS between non-SRC and SRC patients was observed (HR: 1.02, P = 0.830). Besides, after analyzing 2454 colorectal cancer patients, Tan et al[9] demonstrated that the presence of SRCs was an independent prognostic risk factor in colorectal cancer patients (SRC ratio < 50%: 2.182, P = 0.005; SRC ratio > 50%: 1.699, P = 0.016). Meanwhile, it has been reported that patients with an SRC ratio > 50% are more likely to experience an advanced stage of disease and worse survival than patients with an SRC ratio < 50%[10]. However, these meta-analyses focused on the gastric and colorectal carcinomas which are more likely to be combined with SRCs and their results were inconsistent with the studies about esophageal and esophagogastric junction adenocarcinoma (EEGJA). Meanwhile, for EEGJA, the clinical significance of SRCs remains unclear because of the inconsistent reports [11][12][13][14][15][16][17][18][19][20]. Therefore, we conducted this systematic review and meta-analysis to explore the clinical significance of the presence of SRCs in EEGJA patients and the impact of SRCs on the clinicopathological and prognostic characteristics, which might contribute to the prediction of prognosis and formulation of treatment strategy for EEGJA patients. MATERIALS AND METHODS This systematic review and meta-analysis were conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [21]. Inclusion criteria The inclusion criteria were as follows: (1) Patients were pathologically diagnosed with EEGJA; (2) All patients received surgical therapy; (3) Patients were divided into two groups according to the presence or absence of SRCs, and prospective randomized controlled trials (RCTs) or retrospective cohort studies were both available; (4) Clinicopathological parameters or prognosis between the two groups were compared and relevant data were provided; (5) Although the HRs with corresponding 95% confidence intervals (CIs) were not directly reported in articles, the Kaplan-Meier survival curves were provided to calculate them; and (6) The articles were published in English. Literature retrieval The PubMed, Web of Science, and EMBASE electronic databases were searched from the establishment date to March 28, 2021. The following terms were used: "esophageal", "esophagus", "esophagogastric", "gastroesophageal", "adenocarcinoma", and "significance of signet ring cell". A combination of subject terms and free words was applied. Besides, the references cited in included studies were also reviewed for availability ( Figure 1). Exclusion criteria The exclusion criteria were as follows: (1) Duplicated studies or studies with severely overlapped data; (2) Case reports, reviews, meeting abstracts, and animal trials; and (3) Adenocarcinomas located in other sites like the stomach were also enrolled without subgroup analysis for EEGJA patients. The titles and abstracts were screened first and irrelevant publications were excluded. Then full texts of potentially related studies were further reviewed for availability. The literature retrieval and selection were conducted by two investigators (Wang YF and Xu SY). Data extraction The data extraction was conducted by two authors (Wang YF and Xu SY) through the Microsoft Excel sheet independently. The following information was extracted from each included studies: The author, publication year, country where the study was conducted, sample size, tumor location (esophageal vs esophagogastric), tumor-nodemetastasis (TNM), SRC ratio, sex, smoking, family history, lymph node metastasis status, relative risk (RR), and HR with 95%CI or corresponding data for their calculation. The association of SRCs with sex, smoking, family history, tumor location, lymph node metastasis, and TNM stage were measured in this study. For the prognostic role of SRCs in EEGJA, the primary outcome was the OS and the second outcomes included the disease-free survival (DFS) and disease-specific survival (DSS). Quality assessment The Newcastle-Ottawa Scale (NOS) was applied to evaluate the quality of included studies [22]. This scale includes object selection, comparability, and exposure assessment and objectively evaluates the risk of bias with the maximum score of 9, and December 16, 2021 Volume 9 Issue 35 studies with a score ≥ 6 were considered to have high quality. The quality assessment was performed by two authors (Wang YF and Xu SY) independently. Any disagreement was solved by team discussion. Statistical analysis All statistical analyses were conducted with STATA 12.0 software. The RR and HR with corresponding 95%CI were calculated to assess the association between the presence of SRCs and clinicopathological characteristics and prognosis of EEGJA patients, respectively. If the HR with 95%CI was not reported in articles directly, they would be calculated from Kaplan-Meier curves [23]. The heterogeneity among the included studies was evaluated by I 2 statistics and Q test. When significant heterogeneity was observed [I 2 > 50% and (or) P < 0.1], the random effects model was applied; otherwise, the fix effects model was used. The sensitivity analysis and metaregression analysis were performed to detect the source of heterogeneity and evaluate the stability of pooled results. Besides, the Begg's funnel plot and Egger's test were conducted to detect publication bias [24,25]. Significant publication bias was defined as P < 0.05. Basic characteristics Among the ten included articles, Chirieac et al [11] and van Hootegem et al [17] enrolled two different subgroups of patients, which were regarded as two studies separately. Thus, a total of 12 retrospective cohort studies from ten publications were included, involving 30322 EEGJA patients. Most cases were from America and the sample size ranged from 163 to 14224. Three studies only enrolled esophageal adenocarcinoma patients. All studies were high-quality studies with an NOS score of 6 or higher. Detailed information is presented in Table 1. Sensitivity analysis and publication bias According to Figure 4, the sensitivity analysis revealed that the pooled results were stable and reliable. Besides, the Begg's funnel plot was symmetrical ( Figure 5) and the P value of Egger's test was 0.572, which indicated nonsignificant publication bias. Meta-regression analysis Due to the significant heterogeneity of the OS, meta-regression analysis was conducted based on some variables including the publication year, country, sample size, location, SRC ratio, TNM stage, and NOS score. Unfortunately, none of these parameters were the causes of heterogeneity. DISCUSSION The current systematic review and meta-analysis demonstrated that the presence of SRCs was significantly associated with tumor location (RR: 0.76, P = 0.022) and TNM stage (RR: 1.30, P = 0.031). Meanwhile, the presence of SRCs in surgical EEGJA patients predicted a poor OS (HR: 1.36, P = 0.002) and DSS (HR: 1.86, P < 0.001). The presence of SRCs might be an independent prognostic factor in EEGJA patients and contribute to the evaluation of prognosis and decision making. Interestingly, Chirieac et al [11] demonstrated that the presence of SRCs predicted a much poor prognosis in patients who received surgery alone (P = 0.05), but for patients who received neoadjuvant chemotherapy and surgery, the presence of SRCs predicted a better prognosis (P = 0.02). This phenomenon indicated that SRCs might play an essential role in the response to chemotherapy. In the study by Corsini et al [18], EEGJA patients with usual type had a much higher pathologic complete response than patients with SRCs (25% vs 10%, P = 0.006). Meanwhile, Solomon et al [20] also revealed that patients with SRCs were less sensitive to neoadjuvant chemoradiotherapy (OR: 6.118, 95%CI: 1.299-28.821, P = 0.022) and less likely to experience downstaging after neoadjuvant chemotherapy (OR: 0.306, 95%CI: 0.099-0.946, P = 0.040). The presence of SRCs may predicted a poor response to chemotherapy, which is opposite to the results reported by Chirieac et al [11]. Thus, more relevant studies are December 16, 2021 Volume 9 Issue 35 still needed to identify the clinical role of SRC in the chemotherapy and radiotherapy of EEGJA patients. Although the pooled results indicated that there was no significant association between the presence of SRCs and DFS of EEGJA patients (HR: 1.21, 95%CI: 0.94-1.57, P = 0.145), Yoon et al [12] and Patel et al [15] both reported that patients with SRCs were more likely to experience a worse DFS (HR: 1.49, 95%CI: 1.15-1.93, P = 0.001; HR: 1.34, 95%CI: 1.02-1.80, P = 0.048). Therefore, we believed that SRCs might also predict a poor DFS in EEGJA patients, which needs more studies to further verify. Besides, Nafteux et al [14] demonstrated that the presence of SRCs was significantly associated with a higher recurrence rate (56% vs 42% for usual type adenocarcinoma, P = 0.003). In their multivariate analysis, SRC ratio > 50% was an independent risk factor for recurrence (OR: 2.070, 95%CI: 1.159-3.696, P = 0.014), which also indicated that the presence of SRCs was related with a poor prognosis in EEGJA patients. Actually, there are still several valuable fields about the clinical significance of SRCs in EEGJA patients that are worth further investigation. First, as mentioned above, the role that SRCs play in the chemotherapy and radiotherapy is unclear, and more studies comparing the clinical outcomes between EEGJA patients with and without neoadjuvant chemoradiotherapy or postoperative chemoradiotherapy are needed. Second, it is necessary to identify whether SRCs could serve as a reliable predictor for the selection of therapeutic strategies. Third, to explore the role of the proportion change of SRC during chemotherapy in predicting the prognosis of EEGJA patients might be significative. Fourth, according to previous reports, the proportion of SRCs may also be related with the prognosis of EEGJA patients and patients with different ratios of SRCs might have different survival rates. December 16, 2021 Volume 9 Issue 35 There were some limitations in this study. First, all included studies were retrospective, which may cause some bias. Second, the association of SRCs with other parameters such as alcohol drinking, differentiation status, Helicobacter pylori infection, and age was not explored due to the lack of relevant data. Third, we failed to conduct subgroup analysis based on the clinicopathological parameters such as age, sex, and TNM stage because the detailed data were not available even if we contacted the authors of included studies. CONCLUSION The presence of SRCs is related with advanced tumor stage and poor prognosis and could serve as a reliable and effective parameter for the prediction of postoperative survival and formulation of therapy strategy in EEGJA patients. However, more prospective high-quality studies are still needed to verify our findings. Research background The clinical role of signet ring cells (SRCs) in surgical esophageal and esophagogastric junction adenocarcinoma (EEGJA) remains unclear now.	2021-12-12T16:42:18.919Z	2021-12-16T00:00:00.000	{ "year": 2021, "sha1": "7e7772cea7faef1fae0b03ae9516ed9a15d7e0bc", "oa_license": "CCBYNC", "oa_url": "https://doi.org/10.12998/wjcc.v9.i35.10969", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "ee6ac05ee8b65bcd274bab78e32a77a160a9e4d1", "s2fieldsofstudy": [ "Medicine" ], "extfieldsofstudy": [ "Medicine" ] }
238670671	pes2o/s2orc	v3-fos-license	State-led bricolage and the extension of collective governance: Hybridity in the Swiss skill formation system This paper explores the extension of collective governance to sectors without collective governance tradition. We introduce the concept of state-led bricolage to analyze the expansion of the Swiss apprenticeship training system – in which employer associations ful ﬁ ll core collective governance tasks – to economic sectors in which training had previously followed a school-based and state-oriented logic. In deindustrializing societies, these sectors are key for the survival of collectively governed training systems. Through a mixed-methods analysis, we examine the reform process that led to the creation of new intermediary organizations that enable collective governance in these sectors. In addition, we compare the organizational features of these organizations with the respective organizations in the traditional crafts and industry sectors. We ﬁ nd that the new organizations result from state-led bricolage. They are hybrid organizations that re ﬂ ect some of the bricoleur ’ s core policy goals and critically build on the combination of associational and state-oriented institutional logics. Introduction Institutional orders emphasizing collective governance have attracted considerable attention as alternatives to more market-based or state-based orders because of their ability to produce desirable socio-economic outcomes (Hall & Soskice 2001;Streeck & Kenworthy 2005;Busemeyer 2015). In terms of institutional logics (Deeg 2005;Carstensen 2017), collective governance relies on a specific logic that has evolved historically (Streeck & Schmitter 1985). This associational logic has an important effect on employers' capacity to provide collective goods and contribute to social goals (Culpepper 2003;Campbell 2007). Yet, due to increasing competitive pressures in a globalizing economy and the decline of manufacturing as stronghold of the associational logic, systems based on employers' provision of collective goods are under pressure to adjust (Martin & Thelen 2007). Prominent examples are dual vocational education and training (VET) systems in coordinated market economies (CMEs), which rely on the associational logic to foster the voluntary provision of training by firms (apprenticeships). In these apprenticeship systems, employer associations are assigned key governance roles as they develop training content that fits labor market needs and foster firms' participation in training (Busemeyer & Apprenticeship training in CMEs is widely acknowledged as a prototypical case of collective governance. However, given that the traditional base of apprenticeship training is in the crafts and manufacturing sectors, which are in relative decline, maintaining both strong collective governance and high participation rates requires deepening its linkage to the expanding service sector (Busemeyer & Trampusch 2012). How can the state extend collective governance to sectors that lack the intermediary organizations needed for collective governance? What does transferring the associational logic of traditional apprenticeships imply for collective organizing in the new sectors? And most importantly in the context of this analysis, which kind of reform process manifests itself in the characteristics of the newly created associations? Answers to these questions will provide a better understanding of new forms of collective organizing outside of the sectors in which collective governance is already well established. In contrast to the emphasis on institutional stability and coherence in the classical conception of varieties of capitalism (Hall & Soskice 2001), we argue that the state may strategically create and enable hybrid organizational forms to solve the problem of transferring collective governance to new sectors. Through what we call state-led bricolage, the state can transfer the associational logic to sectors without collective governance tradition. In this process, the state as bricoleur develops institutional solutions by combining elements of existing institutional principles and practices in new ways, which results in organizations that are different from but potentially resemble the old ones (Campbell 2004). Crucially, the state retreats to its role as enabling state (Schmidt 2009) once the bricolage is achieved. Furthermore, given the state's high capacity as agent of institutional change, the state leaves footprints on the outcome of the change process, most notably by making the new organizations reflect some of its core policy goals. Together, these two features of the reform process lead to hybrid arrangements, as the associational logic gets blended with the logic characterizing the targeted sectors and the state's policy goals are partially imprinted on these logics. Empirically, our study examines a crucial reform of the Swiss apprenticeship system, which is proportionally the world's largest one (OECD 2017a). Collectively governed apprenticeship training represents an unlikely case for state-led bricolage given that in these systems, private actors have a strong sense of ownership (Busemeyer & Trampusch 2012) and thus typically resist state intervention. Yet, by extending the associational logic to new economic sectors through bricolage, the Swiss state intervened to stabilize the apprenticeship system. While in most traditional dual VET countries, the rate of apprentices relative to students has dropped in recent years, Switzerland stands out with a high and stable rate of around 60 percent of all learners in their first year of upper secondary education enrolled in apprenticeships (OECD 2017a). To achieve this stabilization, the state needed to adapt the system to socio-economic change. When in the 1990s, firms' provision of apprenticeship positions dropped dramatically, the system was reformed with the aim of maintaining it. A crucial measure was expanding the VET system to economic sectors that were not yet part of the system: agriculture, forestry, arts, health and social care (see Appendix S1, Table A1). Training in these sectors had never followed an associational logic but a stateoriented one and primarily took place in schools rather than firms. Thus, the reform addressed the challenge to transfer the associational logic of traditional apprenticeships to sectors that followed different organizing principles and lacked traditional employer associations in charge of VET. To transfer the associational logic to the new sectors, the state sponsored the creation of new organizations to fulfill employer associations' tasks. At the formal level, authorities introduced a new umbrella term to include these organizations in the system. Both the new organizations and the traditional employer associations are called "Organizations of the World of Work" (Organisationen der Arbeitswelt, OdA). This deliberately broad term allowed including organizations in VET governance that do not necessarily correspond to the typical organizational form of employer associations in the VET field. Our case study of one New-Sector-OdA (i.e., OdA in the newly integrated sectors) traces the change process and shows that the combination of elements of different institutional logics and the imprinting of some of the state's core policy goals made the resulting New-Sector-OdA a hybrid organization. We find that this organization (OdASanté) dependedat least initiallyon the capacity of the state to act as bricoleur to encourage private actors to collectively organize in new ways. This allowed to integrate an important economic sector into the VET system, which was essential to the survival of collective governance in a changing environment. To explore whether New-Sector-OdA more generally display hybrid characteristics, we subsequently compare the organizational characteristics of all New-Sector-OdA to the Old-Sector-OdA (i.e., OdA in the traditional sectors for apprenticeship training) based on an original survey of all 146 Swiss OdA responsible for initial VET. The statistical analysis supports our argument that the creation of hybrid organizational forms helps to overcome public policy problems related to collective governance. In the next section, we develop our theoretical framework. First, we introduce the institutional logics of the state and the association and we theorize hybridity in the context of state-led bricolage. Then, we operationalize hybridity and describe our methods and data. Next, we present our findings on New-Sector-OdA through an historical account of the relevant VET reform, a case study of an exemplary New-Sector-OdA (OdASanté), and a cross-sectional comparison of New-and Old-Sector-OdA. A final section concludes. Theoretical framework: State-led bricolage and hybridity This section develops our core argument around state-led bricolage and hybrid organizations. We start by introducing the institutional logics of the state and the association as two central building blocks for this argument. 2.1. Institutional logics: The state logic and the associational logic Each of the major institutional orders of contemporary Western societies has a key institutional logic. Deeg (2005, p. 172) refers to logics as "typical strategies, routine approaches to problems, and shared decision rules that produce predictable patterns of behavior by actors" within an institutional system (see also Streeck & Thelen 2005, pp. 12-13;Carstensen 2017, pp. 144-145). We follow Streeck and Schmitter's (1985) famous distinction between four institutional orders and respective logics. The guiding principle of interaction in the logic of the state is hierarchical control, in the logic of the association it is organizational concertation, in the logic of the market the authors refer to dispersed competition, and in the logic of the community to spontaneous solidarity. In this paper, the state and associational logics are of relevance, which is why we introduce them in greater detail (see Appendix S1, Table A2, for additional information). As Streeck and Schmitter (1985, p. 6) put it, "in the ideal-typical state bureaucracy, allocational decisions are made through public policies that are enforced, with the ultimate backing of the state's monopoly on legitimate coercion, by civil servants striving to satisfy their dominant interest in career advancement and bureaucratic stability, on subjects which strive to avoid punishment." In this depiction, the focus lies on the state as bureaucracy rather than partisan aspects. State agencies are the main actors that devise authoritative regulations to which other actors abide because otherwise they face sanctions. The state logic hence resonates with the concept of a regulatory state that devises rules, monitors, and sanctions them (Koop & Lodge 2017). In statist skill formation, VET is typically school based and employers' involvement in training provision limited (Busemeyer & Trampusch 2012), which together with significant involvement of public actors reflects a statist logic. The state logic captures well the pre-reform state-oriented logic of training in economic sectors in Switzerland which were not yet part of the apprenticeship training system. In these sectors, training was carried out under the leadership of regional administrations, while being strongly school-oriented and characterized by weak involvement of employers and their associations (see case study below for details). Consequently, we refer to the pre-reform logic of the more school-based forms of training in the economic sectors that were to be integrated into the collective governance system of apprenticeship training as state logic. In contrast, apprenticeship training in collectivist systems in countries like Germany and Switzerland is typically characterized by an associational logic (Busemeyer & Trampusch 2012). In this logic, "actors are contingently or strategically interdependent in the sense that actions of organized collectivities can have a predictable and determinant effect (positive or negative) on the satisfaction of other collectivities' interests, and this induces them to search for relatively stable pacts" (Streeck & Schmitter 1985, p. 11). In this logic, coordination and networks among employers are strongly developed and provide an alternative, for instance, to coordination by the market. The associational logic allows firms to overcome cooperation dilemmas in various institutional arenas, such as industrial relations or skill formation (Culpepper 2003). The associational logic is characterized by the systematic involvement of employer associations and labor unions in policymaking and implementation (Streeck & Kenworthy 2005). Focusing on employers, 1 ideal-typical associations are membership-based with associations representing employer interests in a broad array of policy fields often along sectoral lines (Farago & Kriesi 1986;Behrens & Helfen 2009). For example, they negotiate collective labor agreements with unions. Typically, such associations of firms are members in higher-level umbrella associations. This allows a hierarchically structured and coordinated aggregation and representation of employer interests (Schmitter & Streeck 1999). Umbrella associations represent employer interests in policy forums such as consultative commissions with public authorities. Moreover, the state conveys public tasks to employer associations, which is thought to improve the efficiency of implementation and lower its costs (Streeck & Schmitter 1985). This cooperation with the state is said to promote the organizational development of employer associations (Traxler & Huemer 2007). However, finding a balance between cooperating with the state to gain influence in policymaking and remaining faithful to members' interests is challenging for such intermediary associations (Schmitter & Streeck 1999). For example, cooperating with the state may imply pursuing more social goals than firms would otherwise consider (Campbell 2007). Although the associational logic increases the organizational capacity of associations, not all associations are equally developed. Many of the traditional associations have existed for a long time (some date back to the mid-19th century), but they can vary, for instance, regarding their degree of professionalism. For example, many associations unite small groups of firms and rely on voluntary member support, while some are highly professionalized and employ paid staff. The role of the state for fostering the associational logic In countries and policy fields that follow the associational logic, employer associations capable of finding a balance between cooperation with the state and representing members' interests are crucial actors (Schmitter & Streeck 1999). Collectively governed skill formation, which combines training at the workplace and in schools, traditionally builds on associations that foster cooperation among private actors as well as with public actors for the administration and reform of training (Culpepper 2003;Busemeyer & Trampusch 2012). However, such capable associations do not exist everywhere. Typically, institutions fostering collective organizing of employers have their origins in the crafts and industry sectors (Thelen 2007). Yet, the importance of the crafts and industry sectors in terms of employment is decreasing. Moreover, economies' skill needs are changing not least due to technological change and demographic developments, which trigger an increased need of, for instance, health and care professionals (OECD 2017b). Consequently, if the involved stakeholders want to maintain the associational logic in skill formation, they need to adapt it to these developments. This adaptation is challenging, however. In countries that rely on the associational logic, the state may be reluctant to or lack the capacity for coercive state intervention. Due to the institutional arrangements between private actors and the state, the state may need to negotiate with the involved actors and gain agreement on reforms, otherwise reforms may be blocked or fail (Schmidt 2009, p. 525). Moreover, as the state has delegated important public tasks to private actors, such as the provision of the collective good of transferable training, it now depends on the private actors' continuous provision of these services (Streeck & Schmitter 1985). Hence, traditionally, under the associational logic, the state acts as enabler. It facilitates private actors' decentralized cooperation (Culpepper 2003). It delegates public tasks to private actors, to business associations, and labor unions, and these actors are involved in defining rules in areas such as wage bargaining or apprenticeships (Hall & Soskice 2001;Streeck & Kenworthy 2005). Thus, the enabling state fosters nonmarket coordinating institutions (Schmidt 2009, pp. 521-522). In the heyday of this associational logic, the state could retreat to a partly passive, reactive, and indirect role, giving substantial autonomy to private actors to manage their own affairs (Streeck & Kenworthy 2005). Yet this institutional arrangement was based on the capacity of intermediary associations to enforce collective decisions among members. In a context of deindustrialization and liberalizing pressures, state inactivity may lead to the exhaustion of institutions such as collective skill formation. Hence, the state may be pushed to adopt a more active role to ensure the sufficient provision of vocational training. Following Schmidt (2009, p. 526), the state could, theoretically, take over the functions it delegated to private actors or coerce employers to offer training slots. However, this option would imply covering important costs and losing the close connection between the labor market and training for which dual VET is valued. In addition, such coercive state intervention may be blocked in the context of a system that is marked by cooperation and negotiation between the state and private actors (Hall & Soskice 2001). However, inactivity is not a desirable option either, as private actors left on their own encounter important obstacles to engage in the provision of collective goods (Culpepper 2003). Consequently, to maintain the existing associational logic in VET, we argue that the state needs to adopt a more active roleat least temporarily. One approach is to expand the system to new economic sectors. 2 Yet, in these sectors the state may lack partners for the delegation of tasks, most importantly intermediary associations of employers that help overcoming cooperation dilemmas among private actors. Moreover, previously involved stakeholders in training in the new sectors might not be easily displaced. In the next section, we argue that through bricolage, the state may attempt combining the associational logic of the apprenticeship system with previous institutional logics in newly integrated sectors. In this way, and in contrast to approaches that emphasize the importance of keeping institutional logics distinct (Hall & Soskice 2001), the state may be able to foster private actors' organizational capacity to fulfill traditional governance functions before resorting to its primarily enabling role and leaving collective good provision to private actors. 2.3. Theorizing state-led bricolage and hybridity 2.3.1. Institutional variety and state-led bricolage While ideal types such as associational and state logics help to understand and explain differences between countries, in reality, a variety of institutional principles and practices can coexist in a single political economy (Campbell 2010). For example, high-quality apprenticeships may exist outside countries characterized by the associational logic in specific industries or regions (Fortwengel et al. 2019). Hence, institutional variety refers to the coexistence of institutional principles and practices that remain distinct. If one sees institutions as resources to accomplish goals (Streeck & Thelen 2005), a next logical step is to understand such institutional variety as enabling institutional change processes as it provides actors with a range of institutional elements to work with (Crouch 2005). In this context, the basic idea behind the concept of bricolage is that actors frequently develop institutional solutions by combining elements of existing institutional principles and practices in new ways, which results in institutions that are different from but potentially resemble the old ones (Campbell 2004). Bricolage thus typically represents change in a rather evolutionary and gradual way (Carstensen 2017). Here, rather than one institutional logic replacing another completely, a new arrangement is created. If such an arrangement blends elements of the two pre-existing institutional logics, these arrangements are defined as hybrids (Haveman & Rao 2006). While bricolage was originally associated with bottom-up, ad hoc, or situationally improvised change (Lévi-Strauss 1962), we follow authors like Campbell (2004) and Carstensen (2017) who suggest that bricolage can also be promoted more strategically by change agents. Carstensen (2017) highlights that bricolage is a key tool of policymakers to respond to policy challenges. Carstensen (2017) shows that in the case of disruptions to a socio-economic systemin his case a financial crisisstates often resort to bricolage to address policy problems. As a result, the pre-reform institutional logic in the relevant policy field may change gradually but significantly. Like Carstensen (2017), we study a case in which bricolage is a response to a policy challengethe intricate task to extend collective governance to new economic sectorswhich led policymakers to seek new policy solutions. In our empirical case, the respective policy processes are linked to policy solutions crystalized in the 2002 VET law and especially the creation of associations that are needed to extend collective governance to sectors previously governed by a state-oriented logic. Importantly, it matters who the bricoleur is. The capacity of a bricoleur to creatively combine institutional elements depends on its social, organizational, and institutional connections to relevant institutions and actorswhich enlarges the institutional repertoire available for bricolage. Furthermore, this capacity is increased by the bricoleur's tangible resources, like financial means or political clout, which are needed, for instance, to make institutional changes stick (Campbell 2004, pp. 74-77). In our empirical case, the bricoleur is not anyone, but the state. The respective Swiss state agencies are in a privileged position not only in terms of their connections to relevant stakeholders in collective skill formation, but also in terms of the resources available to them to facilitate and maintain institutional changes. Hence, our bricoleur has a relatively high capacity to promote bricolage. However, this state-led bricolage is not a new type of bricolage, but rather a bricolage process for which we identify the state as the key change agent. This high capacity also implies that the state can strongly influence the process of gradual change. Therefore, the result of bricolage is likely to reflect, in one way or another, core policy goals of the state, which are likely to deviate somewhat from the goals of associations typically in charge of training occupations in collectivist systems (Bonoli & Wilson 2019). In this way, the state as a bricoleur may leave a footprint on the outcome of the change process. This, in turn, implies that the state's policy goals get partially imprinted on the respective other logic(s)with specific consequences that we are conceptualizing in the following. In our empirical case, the relevant institutional variety available to the bricoleur consists of the state-oriented logic of VET in the sectors to be integrated in apprenticeship training and the associational logic in the traditional industry and crafts sectors. Importantly, the integration of previously state-oriented VET into the dominant dual VET system based on the associational logic potentially reduces institutional variety at the national level. However, in the process of bricolage the state created new collective actors in the affected sectors, which differ from traditional intermediary associations. Hence, we focus our attention on the characteristics of these new organizations, which enabled the expansion of collective governance to new sectors, by building on the rich literature on hybrid organizations presented in the next section. Conceptualizing hybridity and state-led bricolage at the organizational level While the concept of bricolage speaks to both institutions and their organizational carriers, we focus on measuring the outcomes of bricolage for the latter. In our empirical analysis, the main unit of analysis is the "Organizations of the World of Work" (OdA) as essential intermediary organizations within VET governance in Switzerland. The OdA can be seen as the crystallizing points through which the change process related to the 2002 VET reform is institutionalized at the organizational level. More specifically, if our argument around stateled bricolage holds water, we should observe that the newly created OdA, which result from a creative combination of state and associational logics through state-led bricolage, are hybrid organizational forms. To test our argument empirically, we zoom in on the organizational level and the concept of hybrid organizations. Broadly speaking, one can understand hybrid organizations as entities that arise from the blending of two or more institutional logics, that is, "[h]ybrid organizations combine the institutional logics that are materialized in two or more organizational forms" (Haveman & Rao 2006, p. 974). Thereby, hybrid organizations allow overcoming strict boundaries between two institutional logics. Hybrid organizations can play a central role in processes of institutional change and maintenance (Graf 2013). Crucially, while an organizational field may be exposed to considerable tensions between logics, hybrid organizations can contribute to stabilizing the relevant (sub-)field over time as they develop organizational features to accommodate heterogeneity and thus handle tensions internally (Reay & Hinings 2009), as they can "develop structures and processes that allow them to involve a variety of stakeholders, pursue multiple and often conflicting goals and engage in divergent or (seemingly) inconsistent activities" (Mair et al. 2015, p. 733). Thus, we adopt the perspective that hybridsinstead of necessarily being temporary organizational arrangementscan be long-lasting and have the potential to contribute to permanent policy solutions. Institutionalist research has shown that disposing of coercive authority, the state is a potential driver of organizational adaptation processes (Greenwood & Hinings 1993). Although state-promoted hybridization may also be associated with unintended and conflictual outcomes (Gornitzka & Maassen 2000, p. 284), in this paper we argue that the state can strategically develop hybrid organizations to solve a public policy problem, namely establishing collective governance in new sectors. We thus suggest that the state has the capacity to establish hybrid organizations as a means of system development, for instance, to foster exchanges between relevant stakeholders. In sum, we analyze a case of bricolage, whereby the state combines associational and state logics to reach an instrumental policy goal, the extension of collective governance to sectors that previously followed a state logic for offering vocational training. As the two institutional logics are combined, this should foster hybrid arrangements. Furthermore, the state has a high capacity to act as bricoleur. Once more, this is likely to promote hybridity because due to its strong position, the state's policy goals are likely to getat least partlyimprinted on the associational logic in the newly integrated sectors. For instance, the state being the bricoleur may imply that societal goals are promoted beyond what would be typically associated with the traditional associational logic. Consequently, the new intermediary organizations (New-Sector-OdA) created with the 2002 VET reform through what we call state-led bricolage should differ from the organizational forms traditionally prevalent in the VET field (Old-Sector-OdA). We expect New-Sector-OdA to represent hybrid arrangements both due to the footprint of the bricoleur on the reform process and the combination of associational and state logics. Research design: Operationalization, methods, and data This section introduces an operationalization that enables us to measure the hybridity between the associational and the state-oriented logic at the organizational level. We also describe our methods and data. Operationalizing hybridity In the following, we employ our arguments around state-led bricolage and hybrid organizations to develop specific expectations for New-Sector-OdA. As a structuring element, we refer to three well-established dimensions of organizations, which apply to any organization, irrespective of more specific organizational characteristics (Aldrich & Ruef 2006): First, goal direction means that organizations are purposive systems that do not just follow some random path. Second, organizations are maintained through the control of socially constructed boundaries, which implies a distinction between members and non-members. Third, activity system refers to a set of socially constructed routines: "forms, rules, procedures, conventions, strategies, and technologies around which organizations are constructed and through which they operate" (Aldrich & Ruef 2006, p. 6). For each of these organizational dimensions (see Table 1, column 1), we develop indicators regarding associations in the Swiss VET system (column 2). We then derive theoretical expectations of the differences between New-Sector-OdA and Old-Sector-OdA (column 3). The more each of the indicators reflects hybridity of the associational and state logics, the higher is the overall degree of hybridity of the respective organization. Our core expectation is that while both Old-and New-Sector OdA enable collective governance, they differ as Old-Sector-OdA embrace the associational logic, whereas New-Sector-OdA display significant hybrid characteristics in combining associational and state logics. Goal direction Concerning goal direction, we examine four indicators: policy field, membership composition, organization of short-track dual training, and validation of prior learning. ) Membership in Federal Commissions More often member in Federal Commissions (9) Participation in national VET events More frequently participating in national VET events Activity system (10) Degree of professionalism Higher degree of professionalism (11) Responsibility for collective labor agreements Less often negotiating collective labor agreements (12) Responsibility for generally binding collective VET funds Less often responsible for generally binding collective VET funds †If New-Sector-OdA are hybrid organizations combining state and associational logics, then the statements in the last column should hold. See Appendix S1 for coding and data sources. Associations develop and formalize their organizational goals in the form of associational aims, which are codified in their statutes. Such associations represent member interests toward public authorities (Schmitter & Streeck 1999). Yet, the policy field (Indicator-1) in which they pursue aims can either cover various areas or be limited to a narrower area of interest representation such as VET. The more heterogeneous the interests of an organizations' members, the more difficult it becomes to create internal consensus and to represent these interests toward external stakeholders (Traxler & Huemer 2007). We expect that New-Sector-OdA mainly focus on the policy field of VET, as this limits the degree of heterogeneity in at least one organizational domain and thus helps to compensate for the heterogeneity already linked to the combination of associational and state logics. Furthermore, the focus on VET would correspond to what the state, as the bricoleur, was envisaging when setting the stage for these new intermediary organizations. As these associations represent their members' interests, membership composition (Indicator-2) translates into different goal directions. Here, hybridization may be linked to the inclusion of various stakeholders in an organization (Mair et al. 2015). Owing to their role as governance actors before the inclusion in the national VET legislation, we expect that public authorities are included in New-Sector-OdA membership. This, in turn, implies the presence of a state logic next to an associational logic. As indicated in the previous section, it matters which actor promotes hybridization. The goals of this actor may come to the fore in the goals of the newly created organizations. In our case, the state was the key driver of the reform. The state aims to ensure that VET is inclusiveness-enhancing, which is not always the case for employers (Bonoli & Wilson 2019). The Swiss VET reform introduced two inclusion-enhancing measures: The option to organize short-track training (Indicator-3), which lasts only 2 years and aims to integrate weaker students (Di Maio et al. 2019), and validation of prior learning (Indicator-4), which refers to the possibility for adults to gain a nationally recognized VET diploma without going through an apprenticeship, for example by formally recognizing skills they acquired on-the-job (BBV, Art. 31, al. 1). Thus, owing to state involvementand, thus, the state logicin their creation, new OdA should orient their goals more often toward public goods. We expect them to offer more inclusion-enhancing measures such as short-track dual training and validation of prior learning than the employer-dominated Old-Sector-OdA, which build more exclusively on the associational logic and might be reluctant to implement inclusion-enhancing measures. Boundary maintenance We examine five indicators for boundary maintenance: organizational set-up and sponsors, membership in umbrella associations, distance between OdA and firms, membership in Federal Commissions, and participation in national VET events. Some organizational arrangements (Indicator-5) create new boundaries whereas others allow preserving preexisting organizational boundaries. Crucially, setting up a joint venture (rather than creating a traditional association) allows the constituting organizations, that is, organizational sponsors, to safeguard their boundaries. 3 In contrast, being member of a higher-level umbrella association (Indicator-6) implies being part of a system of interest representation with specific boundaries. Membership in associations affects the boundaries and identity of members, as umbrella associations are expected to be able to influence their members' behavior and contribute to the formulation of common goals (Schmitter & Streeck 1999). The organizational set-up as joint venture and membership in higher-level umbrella associations thus indicate specific organizational boundaries. To allow for high levels of heterogeneity and, thus, to enable the bricolage of two institutional logics (i.e., associational and state-oriented) while preserving organizational boundaries, we expect New-Sector-OdA to be more often set up as joint ventures than Old-Sector-OdA. Moreover, to maintain pre-reform boundaries, we expect New-Sector-OdA to refrain from becoming members of umbrella associations. Becoming members of such traditional national-level umbrella associations would hamper their ability to bring together associational and state logics by including employers, employees, and public authorities. Another important aspect of boundaries is the distance between OdA and firms (Indicator-7). Firms are key stakeholders in the collective governance model of dual apprenticeship but were not in the other sectors. The ability to influence firms' behavior is crucial for employer associations in skill formation systems. Firms can either be direct members of OdA or only indirectly through another association that is an organizational sponsor of the OdA. Given their roots in a less employer-dominated and more school-based and state-dominated governance logic, we expect New-Sector-OdA not to include firms as direct members. Thereby, New-Sector-OdA can maintain pre-reform organizational boundaries, and, in this way, preserve elements of the state logic. A further characteristic of associations is that they can negotiate with public authorities. Thus, besides their members, the relationship to public authorities constitutes a decisive organizational boundary. If associations become too closely involved with public authorities, members may lose confidence in the associations' reliability regarding interest representation. Yet, if associations have too little contact with public authorities, they may not enjoy the necessary influence to represent interests toward public authorities (Schmitter & Streeck 1999). Federal Commissions (Indicator-8) are central arenas for steering the VET system. They are composed of a selected group of representatives from all stakeholder groups including state, employer, and employee representatives. Thus, OdA membership in such commissions indicates a close relationship with public authorities and, thus, to the state logic. Given the strong support by the state for New-Sector-OdA, we expect them to be more often represented in Federal Commissions than Old-Sector-OdA in relative terms. Moreover, their participation in national VET events (Indicator-9) indicates proximity to public authorities. National VET events organized by the responsible government agency, the State Secretariat for Education, Research, and Innovation (SBFI), provide crucial platforms for OdA to deliberate on their activities and exchange best practices. Not least due to the need to legitimize the strong support by the state, and to maintain the state logic next to the associational one, we expect New-Sector-OdA to be more regularly present at such events than Old-Sector-OdA. Activity system We examine three indicators for activity system: degree of professionalism, collective labor agreements, and VET funds. The organizational development or degree of professionalism of OdA varies substantially. We assess their degree of professionalism (Indicator-10) by evaluating to which extent they publish and make accessible relevant information on their websites. This information includes visions, missions and aims, statutes and membership regulations, advertising membership, and organizing activities for members or offering them various services (see Appendix S1, Table A3). We argue that the extent to which substantial information is provided on the organizational website reflects professionalism. To create a reasonably stable activity system despite the heterogeneity related to the involved associational and state logics and actors, and given their deliberate design through stateled bricolage rather than traditional evolution, a sustainable hybrid organization is expected to build on a high degree of professionalism. Thus, New-Sector-OdA are likely to display on average a higher degree of professionalism than Old-Sector-OdA. A specific domain, in which traditional employer associations engage, is negotiating collective labor agreements (Indicator-11) with unions. Although, in Switzerland, VET is not tightly linked to wage coordination and unions play a minor role in VET governance , some OdA are in charge of collective labor agreements. Given that collective labor agreements are a core activity for traditional employer associations but not in the newly integrated sectors, and that these collective agreements might overstretch the organizational capacity to combine associational and state logics, we expect them to be less prevalent among New-Sector-OdA. In contrast, although sectoral VET funds (Indicator-12) have a long tradition in certain sectors, only the 2002 VET act introduced the possibility to declare them generally binding. If they create generally binding VET funds, OdA rely on the coercive authority of the state, that is, indirect state support (Traxler & Huemer 2007). Thus, generally binding VET funds commit all firms (not only member firms) employing a specific occupation to financially contribute to training activities. Due to the important role of the state and the state logic in their creation, we expect New-Sector-OdA to develop VET funds more often than Old-Sector-OdA, the latter being exclusively based on the associational logic and more reluctant to develop such coercive measures toward firms. Methods and data This section describes our mixed-methods approach. We begin by providing a historical-institutionalist account of the relevant VET reform and the development and organizational characteristics of one exemplary OdA, OdASanté, which was created in the framework of the VET reform that led to the extension of the associational logic to the new sectors. This first part of the empirical analysis serves to demonstrate how the state acted as bricoleur to creatively combine associational and state logics, a process that is crystalized in the form of new hybrid intermediary associations. We select OdASanté as it represents a key example of a New-Sector-OdA, which is today responsible for one of the most prominent apprenticeship programs in Switzerland. In 2018, the health care apprenticeship was the third most common initial VET occupation offering dual training (BFS 2019). Similar developments also occurred in very different fields such as farming or social pedagogy. While the details for each of the sectors integrated into the collective VET governance system differ, the general transition pattern is well-illustrated by OdASanté. Our case study is based on information from available secondary literature, document analysis, our extensive new OdA database (details below), and four expert interviews. Interview partners for semi-structured interviews were selected based on their expertise and practical knowledge on VET governance in Switzerland. 4 In the second part of the empirical analysis, we conduct a cross-sectional statistical analysis of all New-and Old-Sector-OdA in Switzerland, which allows us to see whether our argument around New-Sector-OdA being hybrid organizations also holds more generally. Thus, the qualitative and quantitative parts of the analysis complement each other. To compare hybrid features of New-Sector-OdA to Old-Sector-OdA, we analyze the complete population of Swiss OdA in charge of initial VET occupations. This population consists of 146 organizations, which are recognized by public authorities as responsible for an initial VET occupation. We collected data on these organizations' characteristics in 2017-2018 (i.e., approximately 10 years after their creation) from various written sources based on a structured questionnaire. To ensure the quality of the data, we developed a coding manual with decision rules. Our data sources are the register of all federally recognized occupations with their responsible OdA, the organizations' websites, their statutes, published registers of collective labor agreements and binding VET funds, member lists of peak associations and federal commissions as well as participant lists of events. Additional phone interviews were conducted with OdA publishing only limited information online. We examine differences between New-and Old-Sector-OdA on our 12 indicators. However, a simple comparison of means does not consider the differences in within-group variance and group sizes. We therefore use bivariate regressions to test if the observed differences in means between New-and Old-Sector-OdA are statistically significant despite the rather small number of New-Sector-OdA and heterogeneity within the two groups. Given the properties of our dataset, this is a conservative test of the differences between the two groups. For most indicators, we use logistical regression except for the indicator "participating in national events," where we use poisson regression, and "degree of professionalism," where we use OLS regression. When the outcome is present in under 15 percent of cases, we use logistic regression adapted for rare events (see Appendix S1, Table A4, for the regression results). Our independent variable is the binary variable "new sector," which is coded as 1 if the OdA is responsible for an occupation in the newly integrated economic sectors. We identified these OdA based on administrative sources. Our classification of New-Sector-OdA was then confirmed by several experts in the field with whom we discussed it independently from each other. Most of our dependent variables are coded as binary outcome variables taking the value 1 if the OdA shows the characteristic. The indicator "participating in national events" is a count variable, which varies from 0 to 5 participations. The indicator "degree of professionalism" is a continuous variable, which varies from 0 to 12 points (see Appendix S1, Table A3, for more information on the variables). Findings: State-led bricolage and the hybridity of New-Sector-OdA In this section, we first offer an historical account of the VET reform that led to the extension of apprenticeship training to new sectors and provide an in-depth case study of OdASanté, a sector in which training previously followed a state-oriented logic. We find that the new OdA resulted from a process of state-led bricolage and features a high level of hybridity. Second, our cross-sectional analysis of all Swiss Old-and New-Sector-OdA shows that this finding of hybridity holds more generally for the latter group. The VET reform process and the creation of OdASanté The Swiss VET reform in the early 2000s expanded the associational logic to new sectors such as health care. The new VET act included all non-university vocationally oriented education and training rather than only training in the craft, industrial, and commercial sectors (BBT 2000). This state-led reform was broadly supported by the relevant stakeholders both inside and outside of parliament (Bundesrat 2000, p. 5,708). In fact, in December 2002 the new VET act was passed with unanimity in both parliamentary chambers. However, the parliament demanding a collective solution to training in new economic sectors does not solve the problem of creating these collective solutions. The parliament has no means to create effective intermediary associations that can fulfill the required governance tasks. The difficult task to implement the reform, and to assist in the creation of these organizations, was therefore entirely left to the federal state agency responsible for VET (BBT, later renamed SBFI). This bureaucratic actor thus faced the challenge of integrating a diverse group of economic sectors (e.g., agriculture, arts, and health care) into a collectively governed system. In these economic sectors, there had been no work-plus school-based apprenticeships governed by employer associationsrather, training had taken various mainly school-oriented forms. Regional authorities (cantonal offices) had been the essential stakeholders for providing training, although in some of these sectors, professional associations, for example of farmers and nurses, had been involved (Wettstein 2020). Additionally, in these often para-public sectors, the state tends to be an important employer (even in agriculture and the arts, as public subsidies play an important role). Thus, the state and especially the cantonal offices had been key governance actors in the pre-reform setting of the concerned sectors. As explained earlier, we classify the pre-reform institutional logic in these sectors as state oriented. Indeed, what unites the newly integrated sectors is that training was not organized in the form of dual apprenticeships but rather based on a state-oriented logic in cooperation with partners. However, otherwise, these new sectors differ significantly. With the major 2002 VET reform, the federal state aimed to align training in these sectors to the industrial-craft apprenticeship model that is based on the associational logic. For implementing apprenticeships in these newly integrated sectors, the responsible state agency promoted the creation of new training occupations. Moreover, given the key role of employer associations in the collective governance of dual apprenticeship training (Culpepper 2003), the state provided the legal framework and assisted in founding associations responsible for these apprenticeships. In this context, one of the state agency's main goals was to prominently involve employers in these new associations. Although the federal state aimed to integrate these sectors into the collectively governed apprenticeship system, it was aware that necessary structures, in particular intermediary organizations in charge of VET governance, were lacking (Bundesrat 2000, p. 5,709). Yet rather than setting up organizations similar to the traditional associations described earlier, the state agency initiated and supported the creation of new organizational forms broad enough to combine institutional features of the state logic and the associational logic. This choice reflected the difficulty for the state to create regular employer associations (Culpepper 2003), but also the need to involve already existing actors into the newly created organizations. Importantly, to describe these new organizational forms, the VET lawadopted by parliament but drafted by the state agencyintroduced the deliberately broad term "Organizations of the World of Work" (OdA) that encompasses the variety of old and new associations in charge of VET. Once the relevant OdA were created in the new sectors, the state bureaucracy stepped back from its pro-active steering role to leave the space to private actorsthus contributing to the strengthening of the associational logic. Overall, we observe that in the creation of these new intermediary associationswhich fulfill collective governance tasks usually delegated to employer associations in the traditional sectorsthe state bureaucracy played a critical steering role in combining state and associational logics. For instance, the state agency was in charge of identifying the respective stakeholders, engaging them in a regular exchange, and moderating their dialog. In the following, our deep dive into the case of OdASanté serves to further substantiate our analysis of this reform process. We begin by tracing the pre-reform governance context for training in the health sector. Historically, nursing training in Switzerland was primarily school-based and the most prominent entrance point for health occupations (Maurer 2013). The respective programs were mainly focused on school diplomas, while the employers' side played a very limited role in their governance (Flury 2004, p. 25). Cantonal health offices were responsible for securing public health, including the availability of health professionals. Together with the cantons, the Swiss Red Cross (SRC) played a relevant role in developing health care diplomas in Switzerland (Wettstein 2020, pp. 118-119). All cantons recognized SRC diplomas and SRC monitoring ensured that training followed nation-wide standards, but training was not clearly positioned within the educational system (Spitzer & Perrenoud 2006;Bender 2011). However, with the growing importance of health care training, cantons were increasingly taking over responsibilities from the SRC (Bender 2011). For instance, the intercantonal conference of health offices established a public-private partnership with the SRC. 5 The SRC's competencies included the definition of training curricula, the monitoring of the schools, and the recognition of foreign diplomas. Yet, cantonal health offices organized and financed the health care training schools (BBT 2000). Overall, the SRC had to carry out the will of the cantons regarding the training (Flury 2004, p. 5). The nurses' association was involved in questions of training by cooperating with the SRC, while employers remained sidelined (Bender 2011). However, the influence of professional associations was rather marginal: training in the health sector prior to the reform mainly involved cantonal health offices, the SRC, and schools (Flury 2004, p. 7). Training in the health sector was thus following a school-and state-oriented logic, while employer associations played virtually no role. In the late 1990s, the Federal Parliament decided to integrate all non-university-based training into the VET system. Hence, the BBT was commissioned to align training in the health sector to the industrial-craft apprenticeship model that is based on the associational logic (Bundesrat 2000). This required a genuine effort on the side of the state agency to nudge the relevant private actors to work together and, at the same time, handing over significant responsibility for training to the employers (Flury 2004, p. 33). While the SRC did not see another option than to agree to the integration of health care training into the regular education system (Bender 2011), it tried to limit the damage to their own organization and staff by trying to transfer their training section to the BBT, but this initiative failed. The nursing association supported the reform because it hoped to attain their aim of professionalizing nursing training (Kaufmann 2010). In terms of transition process, first, to anticipate the integration of health care training in the VET system, various cantonsi.e., the respective regional state agenciesdeveloped pilot projects and created the first forms of the future health care apprenticeship (Interview-CH3). Public funding destined to counter the lack of apprenticeship places was made available to finance these local projects. In these projects, the employers, such as hospitals, were involved at the local level, while the SRC was also involved and issued diploma (Interview-CH3). In a second phase, based on these local projects, a national health care apprenticeship was developed. In 2000, the relevant state agency, BBT, created a project structure for integrating this sector into the national VET framework (Project-Group-OdASanté 2004, p. 4). The BBT fulfilled the overall steering role in this process as responsible agency for implementing the VET reform. It guided the major reform process that was to transfer the associational logic to a sector previously dominated by a state logic (Flury 2004). However, representatives of the intercantonal health office conference and the intercantonal education office conference were also involved as members of the project's steering group (BBT 2000). Once the new health care apprenticeship had been created, the state-led project steering group had the task of creating a new organization that would fulfill employer associations' traditional role in the Swiss VET system. This resulted in a new intermediary organization, namely OdASanté. While the project steering group was composed of both public and private actors, the Swiss Health Directors Conference, that is, the political coordination body of the cantons in health policy, was chairing the group (Project-Group-OdASanté 2004, p. 5). The state actors' lead position in the transition process is also illustrated by the availability of public start-up funds for OdASanté (which otherwise is fully funded by membership fees) in this development phase (Project-Group-OdASanté 2004, p. 3). While previously the SRC on behalf of the cantons had been in charge, during this transformation process other stakeholders, especially representatives of the employers in the health sector, discovered VET as an important policy field and became more strongly involved (Flury 2004, p. 32). In this context, we can observe that the state agency fulfilling its task to align the new sectors into a unified national framework acted as a bricoleur in bringing together these actors and in combining the pre-reform state logic of the newly integrated sectors with the associational logic otherwise dominant in apprenticeship training. In line with this process of state-led bricolage, we find that OdASanté differs from the average OdA in traditional VET sectors in all three core organizational dimensions (goal direction, boundary maintenance, activity system). In the following, our application of the 12 indicators (see Section 3.1) establishes that OdASanté displays organizational features reflecting hybridity. Regarding goal direction, our analysis shows that OdASanté's policy field is restricted to training (Indicator-1), whereas traditional employer associations represent interests in a broad policy domain: "OdASanté is a training organization, not a strong employer or occupational association" (Interview-CH1). The project team in charge for the transition process decided that this organization should include various stakeholders with diverging interests (Wettstein 2020, p. 122): employers and occupational associations (Interview-CH4), but crucially also cantonal authorities in the health domain . This hints at the continued relevance of the pre-reform state-oriented logic. The nurses' association is included in OdASanté and thus continues representing nurses' interests, especially regarding a clear distinction between the (tertiary educated) nurses and the newly established health care workers (Interview-CH3)but was a relatively weak actor in the OdA's establishment relative to the intercantonal health office conference. The SRC was excluded from governance, as there is no role foreseen for an NGO in the VET system. In contrast, employershospitals, home care nursing organizations, and nursing homeshave received a prominent role in collective governance. Their associations are included in OdASanté and are also represented in the cantonal sections of OdASanté together with cantonal VET and health offices (Interview-CH4). In addition, the goal direction of OdASanté is characterized by it organizing inclusiveness-enhancing short-track training (Indicator-3) and validation of prior learning . Moving on to boundary maintenance, we find that OdASanté is a joint venture that allows bringing together different pre-reform stakeholders as sponsoring organizations . Typically, in a joint venture, the sponsoring organizations have a broader domain of activity than the joint venture, which has a more specific task (Interview-CH4). OdASanté is not member of a national-level employer association (Indicator-6). Although employer associations have gained an important role in governance, firms are only indirect members of OdASanté through their associations (Indicator-7). However, OdASanté was included as member in prestigious federal commissionsbased on state invitationalongside the national umbrella employer and employee associations (Indicator-8). Furthermore, OdASanté is an active participant in national VET events, again based on state invitation (Indicator-9). Finally, regarding its activity system, OdASanté displays a high degree of professionalism (Indicator-10), which is illustrated, for instance, by its professional outreach activities and communication strategy. In contrast to traditional employer associations, it is not in charge of a collective labor agreement (Indicator-11) but OdASanté has developed a generally binding VET fund, which, as mentioned above, is a coercive measure to commit all firms to training investments backed up by state authority . In sum, the case study of the creation of this major New-Sector-OdA has provided insights into both the state-led change process and the ensuing hybridity characteristics of this organizational formespecially regarding the state as a bricoleur in the reform process and as an important pre-reform player in the integrated sectors. Next, we turn to the statistical analysis of all New-and Old-Sector-OdA, which allows us to see whether our argument around New-Sector-OdA being hybrid organizations also holds more generally. Cross-sectional comparison of New-and Old-Sector-OdA Our statistical analysis shows that New-and Old-Sector-OdA are clearly distinct regarding our indicators of hybridity. Figure 1 summarizes our statistical findings for the 12 indicators listed in Table 1, showing group averages in percentage points for the binary variables and the means of the two groups' participation and professionalism as well as the results of our tests for statistical significance. To increase readability, the indicators are organized in a way that higher values always point toward the features associated with higher levels of hybridity. Across almost all indicators, we find a strong confirmation for our expectation that New-Sector-OdA (outer circle) represent hybrid features (see also Appendix S1, Table A4). Concerning the organizational dimension of goal direction, the comparison of New-and Old-Sector-OdA shows that New-Sector-OdA limit their domain of activity more often to the policy field of VET than Old-Sector-OdA (Indicator-1). This, as we argue, is in line with what the state was envisaging when developing the National VET Act of 2002, which sets the stage for the creation of the new intermediary organizations. This leads us to the next point, membership composition. New-Sector-OdA include public authorities in their membership more often than their counterparts (Indicator-2). Public authorities were important stakeholders in the pre-reform setting, and, thus, transferred their influence into the new governance arrangement. Furthermore, our analysis suggests that New-Sector-OdA are oriented toward public goals and, hence, offer more inclusiveness-enhancing VET, but the difference is only statistically significant for short-track training (Indicator-3) and not for validation of prior learning (Indicator-4). More generally, these insights suggest that these hybrids govern new apprenticeships, for instance, by uniting the stakeholders typically found in traditional associations (i.e. employers) and stakeholders that had an important role in the previous institutional arrangement for training (i.e. public authorities). Regarding the second dimension, boundary maintenance, New-Sector-OdA are more often organized as joint ventures than as membership-based associations (Indicator-5). Their constituents are more often composed of organizational sponsors than in the case of Old-Sector-OdA. This organizational arrangement allows bringing together organizations of very different types, such as intercantonal conferences, employer associations, or employee organizations. In line with this organizational form, firms are less often direct members of New-Sector-OdA than Old-Sector-OdA (Indicator-6). This points toward the persistence of pre-reform boundaries and highlights that New-Sector-OdA are clearly distinct from the ideal-typical associations in VET. Moreover, none of the New-Sector-OdA is member of a traditional umbrella association of employers, while 43.9 percent of the Old-Sector-OdA are members of such umbrella associations (Indicator-7). Thus, New-Sector-OdA remain partly outside the existing system of interest representation in VET. Concerning the relationship with public authorities, New-Sector-OdA are more closely involved in federal commissions than Old-Sector-OdA (Indicator-8), which indicates the proximity of New-Sector-OdA to state actors. New-Sector-OdA also participate more often in national VET events than Old-Sector-OdA (Indicator-9). These results show that New-Sector-OdA have developed organizational forms that allow them to maintain pre-reform boundaries. Moreover, New-Sector-OdA are in closer contact to public authorities than Old-Sector-OdA. Thus, a mutual approximation takes place but certain boundaries also persist. Concerning their activity system, New-Sector-OdA have a slightly higher degree of professionalism than Old-Sector-OdA but this difference is not statistically significant (Indicator-10). Yet, an additional analysis shows that if professionalization items associated with their role in VET (6 items) are distinguished from items associated with general organizational development (7 items), New-Sector-OdA score significantly higher on VET-specific professionalization and lower on organization-specific professionalization. This is in line with our expectation that New-Sector-OdA have Figure 1 Comparison of New-Sector-OdA and Old-Sector-OdA (high values = more hybrid). Notes: ** p < 0.01, * p < 0.05, perfect match. The higher the score or percentage, the more salient are the hybrid features. Indicators 1-8, 11, 12 are binary variables: values represent group averages in %. Indicator 9 (count variable taking values from 0-5): average number of participations in national events, multiplied by ten to increase readability. Indicator 10: continuous variable; average score on professionalism index multiplied by ten to increase readability. a strong need to become professionalized and to create a reasonably stable activity system despite the heterogeneity of the involved logics and actors (including state actors). However, Old-Sector-OdA may be more developed as organizations due to their longer existence and broader domains of activity. Further, while 28.1 percent of the Old-Sector-OdA are direct partners in a collective labor agreement, New-Sector-OdA never engage in this activity . This shows that these organizations limit their tasks to a specific field. Finally, New-Sector-OdA are more often responsible for collective VET funds. This difference is statistically significant (Indicator-12). The results indicate that New-Sector-OdA create an activity system that enables them to fulfill their role in VET governance, but do not develop some of the typical, broader activities of traditional employer associations such as negotiating collective labor agreements. Moreover, the traditional employer associations appear more reluctant to introduce binding collective VET funds, which may be due to the associational logic and members' resistance to coercive measures. Summing up, in terms of the hybridity characteristics, the organizational form of New-Sector-OdA allows for heterogeneous organizational arrangements that include employers and, crucially, state actors. They focus on only one policy field and are set-up as joint ventures rather than traditional membership-based associations. Moreover, pre-reform stakeholders are included into the new organizations, which do not become integrated in traditional umbrella associations, do rarely include firms directly, and do not get directly involved in negotiating collective labor agreements. More generally, our indicators point toward the strong role of the state in the process of integrating new sectors, for instance, because New-Sector-OdA contribute to inclusiveness-enhancing measures, participate in national events, and are members of federal commissions. Moreover, they rely on the state's coercive authority to overcome free-rider problems, for instance through generally binding VET funds. Conclusion This paper examined a major reform around a collectively governed policy field in which the provision of collective goods by private actors is promoted through an associational institutional logic. Collectively governed VET is a key policy area in which the associational logic prevails. While the organizational form of associations and their functions are well known, their core tasks are increasingly demanding. In particular, the deindustrialization of the economy challenges established associational systems, which have their strongholds in the industry and crafts sectors . How can collective governance be extended to sectors without collective governance tradition and what does this transfer of the associational logic of traditional apprenticeships imply for collective organizing in the new sectors? We addressed these questions using the case of a Swiss VET reform. Whereas traditional dual VET systems are increasingly under pressure, the Swiss dual VET system remains surprisingly stable, not least because the state managed to extend the system of apprenticeship training to several new and rapidly growing economic sectors. Our analysis focused on the establishment of new forms of collective organizing that support this development. Using a mixed-method design, we have traced how the associational logic was transferred to the health sector through the creation of a new, hybrid organization. We then forwarded a novel way to measure hybridity at the level of organizations and compared all New-Sector-OdA to their older counterparts. We found that the new hybrid organizations provide platforms through which employers, employees, and state actors collaborate to govern apprenticeships. Thus, the successful transfer of apprenticeship to new sectors hinges on the ability to develop innovative organizational features to bridge institutional logics, for example, by creating joint ventures of stakeholders with diverging interests and limiting their policy domain to VET. While only future research can explore the long-term effects of this reform, these hybrid intermediary organizations represent a promising approach to transfer collective governance to the increasingly important service sector. Indeed, New-Sector-OdAenvisaged by policy-makers as a permanent solution from the starthave enhanced the stability of apprenticeship training in the last 20 years since their establishment. Therefore, our findings support the argument in recent organizational scholarship that hybrids have the potential to handle tensions and accommodate heterogeneity in an enduring way. To explain the establishment of the new forms of collective governance, we introduced the concept of state-led bricolage. While we acknowledge that there also more bottom-up pathways to bricolage, this paper focused on exploring state-led bricolage as a strategy to promote change. It thus contributes to ongoing discussions about possible driving forces for bricolage and the emergence of hybrids. Further research could explore in greater detail the conditions for either state-led or more bottom-up bricolage. Moving beyond studies of hybridity lacking a clear agent, we identified a state-sponsored process of bricolage. The critical institutional variety available to the bricoleur was the state-oriented logic in the VET system in the new sectors to be integrated in apprenticeship training and the associational logic in the traditional industry and crafts sectors. The state-led combination of these two logics within the newly created associations led to hybrid organizational forms fulfilling the governance tasks that are typically dealt with by traditional employer associations. Thus, we demonstrated that the state as bricoleur can combine different institutional logics to reach an instrumental policy goalhere the stabilization of the VET system. Furthermore, we found that the bricoleur successfully imprinted some of its own core policy goals on the newly created organizations (e.g., societal goals that go beyond what would typically be associated with the associational logic). Structural economic change poses a challenge to institutions that foster the provision of collective goods by private actors (Thelen 2007;Strebel et al. 2020). This is true for diverse policy fields in CMEs such as wage bargaining and policy concertation (Hall & Soskice 2001) but to some extent also for other economies (Fortwengel et al. 2019). Using the case of apprenticeship training in Switzerland, we derived general insights on how state-led bricolage has the potential to create new forms of collective organizing that take over central governance tasks. In some ways, our case is both a most likely and least likely case for state-led bricolage. On the one hand, in Switzerland, there was a large political consensus that collective governance should be extended to new economic sectors even though these sectors lacked the necessary intermediary associations. This political will cannot be taken for granted, as other countries might opt for a more direct state provision to deal with an apprenticeship crisis, which was however at no point seriously considered in Switzerland. On the other hand, in collective skill formation systems more generally, private actors have a strong sense of ownership and thus resist sustained state intervention (Culpepper 2003;Emmenegger et al. 2019). Therefore, the state can rarely resort to a strongly regulatory approach. Our analysis shows that in systems based on the associational logic, an enabling state may use bricolage to foster private actors' organizational capacity to fulfill collective governance functions before resorting to its relatively passive enabling role, leaving collective good provision again (mainly) to private actors. Bricolage thus allows states to strengthen systems of collective governance, which calls for a reassessment of the state's role in CMEs.	2021-09-27T18:54:00.147Z	2021-08-16T00:00:00.000	{ "year": 2021, "sha1": "ecd219f5757529fc79ebb38463cd55a69b819e54", "oa_license": "CCBY", "oa_url": "https://doi.org/10.1111/rego.12436", "oa_status": "HYBRID", "pdf_src": "Adhoc", "pdf_hash": "fce7394a38be83d656798afd8af177e5844ae2b3", "s2fieldsofstudy": [ "Political Science", "Education", "Economics" ], "extfieldsofstudy": [ "Political Science" ] }
254232442	pes2o/s2orc	v3-fos-license	Availability of Metribuzin-Loaded Polymeric Nanoparticles in Different Soil Systems: An Important Study on the Development of Safe Nanoherbicides Nanoformulations have been used to improve the delivery of fertilizers, pesticides, and growth regulators, with a focus on more sustainable agriculture. Nanoherbicide research has focused on efficiency gains through targeted delivery and environmental risk reduction. However, research on the behavior and safety of the application of these formulations in cropping systems is still limited. Organic matter contained in cropping systems can change the dynamics of herbicide–soil interactions in the presence of nanoformulations. The aim of this study was to use classical protocols from regulatory studies to understand the retention and mobility dynamics of a metribuzin nanoformulation, compared to a conventional formulation. We used different soil systems and soil with added fresh organic material. The batch method was used for sorption–desorption studies and soil thin layer chromatography for mobility studies, both by radiometric techniques. Sorption parameters for both formulations showed that retention is a reversible process in all soil systems (H~1.0). In deep soil with added fresh organic material, nanoformulation was more sorbed (14.61 ± 1.41%) than commercial formulation (9.72 ± 1.81%) (p < 0.05). However, even with the presence of straw as a physical barrier, metribuzin in nano and conventional formulations was mobile in the soil, indicating that the straw can act as a barrier to reduce herbicide mobility but is not impeditive to herbicide availability in the soil. Our results suggest that environmental safety depends on organic material maintenance in the soil system. The availability can be essential for weed control, associated with nanoformulation efficiency, in relation to the conventional formulation. Introduction Formulations that improve the agricultural performance of pesticides and cause less environmental impact are goals of modern agriculture. Traditional formulations are applied in high amounts per area, presenting the potential for the selection of resistant organisms and efficiency loss; furthermore, a large part can be lost to the environment [1]. Nanotechnology has been explored in several fields of knowledge, such as medicine, electronics, cosmetics, food, and agriculture [2]. In agriculture, formulations can be applied in different ways, to improve the efficiency of fertilizers, pesticides, and growth regulators, to promote sustainable agriculture, with the potential for and necessity to be explored even further [3]. Nanoformulations have the potential to minimize the impacts of pesticides on ecosystems, protect molecules, and increase efficiency in pest and disease control [4][5][6][7][8]. Studies with pesticide nanoformulations showed minimization of active losses due to degradation, volatilization, and leaching [7,[9][10][11], as well as alterations in soil retention and mobility [12][13][14], and increases in efficiency and action in plants [6,15]. succession cultivated under conventional tillage (SC-CT) presented less herbicide retention (49.13 ± 1.33% and 47.27 ± 1.2%, respectively) ( Figure 1). Desorption was pronounced in all soil systems (39.38-41.9%), in relation to the control NC soil (29.87 ± 1.11%) ( Figure 1). The sorption coefficient normalized to organic carbon (Koc) (Table S1), presented higher values for SG-MN (120.41 ± 6.42 mL g −1 ) compared to other soil systems (SC-CT, SC-NT, and NC). In the desorption values, the Koc was greatest for SG-MN (198.98 ± 33.28 mL g −1 ), differently from NC (130.29 ± 14.46 mL g −1 ) (Table S1). Figure 1. Sorption (a) and desorption (b) percentages for nanoMTZ and conventional MTZ in different soil systems (non-crop soil-NC, cultivated soil with soybean-corn succession in a no-tillage system-SC-NT, cultivated soil with soybean-corn succession in a conventional tillage system-SC-CT, and sugarcane monoculture soil-SG-MN). Bars represent the standard error of the mean (n = 2). Since the interaction between soil system and formulation is not significant, lowercase letters differ between soil systems (regardless of formulation) and ns indicates no differences between formulations by Tukey's test (p < 0.05). The Freundlich sorption coefficient (Kf) indicates the intensity of 14 C-metribuzin sorption in the soil of the different cropping systems and the value of 1/n indicates the linearity of the isotherm (Table 1 and Figure 2). The sorption and desorption isotherm slopes increase with increasing Kf values ( Figure 2). The Kf sorption values ranged from 1.49-2.66 mL g −1 for nanoMTZ and 1.33-2.73 mL g −1 for conventional MTZ (Table 1), while the Kf desorption values were between 1.39-3.17 mL g −1 and 0.26-3.78 mL g −1 for nanoMTZ Figure 1. Sorption (a) and desorption (b) percentages for nanoMTZ and conventional MTZ in different soil systems (non-crop soil-NC, cultivated soil with soybean-corn succession in a notillage system-SC-NT, cultivated soil with soybean-corn succession in a conventional tillage system-SC-CT, and sugarcane monoculture soil-SG-MN). Bars represent the standard error of the mean (n = 2). Since the interaction between soil system and formulation is not significant, lowercase letters differ between soil systems (regardless of formulation) and ns indicates no differences between formulations by Tukey's test (p < 0.05). The Freundlich sorption coefficient (Kf) indicates the intensity of 14 C-metribuzin sorption in the soil of the different cropping systems and the value of 1/n indicates the linearity of the isotherm (Table 1 and Figure 2). The sorption and desorption isotherm slopes increase with increasing Kf values ( Figure 2). The Kf sorption values ranged from 1.49-2.66 mL g −1 for nanoMTZ and 1.33-2.73 mL g −1 for conventional MTZ (Table 1), while the Kf desorption values were between 1.39-3.17 mL g −1 and 0.26-3.78 mL g −1 for nanoMTZ and conventional MTZ, respectively (Table 1). Freundlich sorption isotherm linearity ( Figure 2) for nanoMTZ and conventional MTZ showed values close to 1.0 (Table 1). These 1/n values (~1.0) represent type C curves [31], where sorption increases with increasing herbicide concentration, indicating that there is no saturation of soil sorption sites. Sorption and desorption isotherm linearity is used to calculate the hysteresis (H). Hysteresis indicates the reversibility of the sorption process and is related to the availability of the herbicide in the environment. Values of H > 0.7 indicate that the desorption process occurs at the same intensity as the sorption process [32], indicating that even after sorption, the herbicide easily returns to the soil solution, where it is available for transport, absorption, and degradation processes. Table 1. 14 C-metribuzin sorption isotherm parameters for two formulations (nanoMTZ and conventional MTZ) based on the Freundlich model, in different soil systems (non-crop soil-NC, cultivated soil with soybean-corn succession in a no-tillage system-SC-NT, cultivated soil with soybean-corn succession in a conventional tillage system-SC-CT and, sugarcane monoculture soil-SG-MN. The data indicate the parameter mean ± standard error (n = 2). Kd-sorption distribution coefficient. Kf-Freundlich equilibrium constant. 1/n-is the degree of linearity of the isotherm. R 2 (adj)-adjusted coefficient of determination. Sorption (a,c) and desorption (b,d) isotherms for nanoMTZ and MTZ, respectively, in different soil systems (non-crop soil-NC, cultivated soil with soybean-corn succession in a no-tillage system-SC-NT, cultivated soil with soybean-corn succession in a conventional tillage system-SC-CT, and sugarcane monoculture soil-SG-MN). Symbols represent Kd values (Cs/Ce) ± mean standard error (n = 2). Lines represent the curve for each soil system according to the Freundlich model. NanoMTZ and Conventional MTZ Soil Mobility The retention factor (Rf) for nanoMTZ and MTZ is shown in Supplementary Table S4. The Rf parameter aids the understanding of herbicide mobility in the soil; the mobility increases as Rf increases [33]. The Rf was between 0.45 and 0.98 for nanoMTZ and between 0.50-1.0 for conventional MTZ. Within soil systems, mobility was smaller for NC soil in both formulations (Rf = 0.45 and 0.5, for nanoMTZ and MTZ, respectively). The highest Rf was observed in DS for both formulations (Rf = 0.98-1.0) Among the other soil systems (SC-CT, SC-NT, and SG-MN) Rf ranged from 0.58 to 0.63 for nanoMTZ and 0.53-0.63 for MTZ (Table S4). In soil with organic residue addition, Rf values varied between 0.8-0.9 for nanoMTZ and 0.85-0.98 for MTZ. Within organic materials, Rf values were smaller in black oat straw and corn residue (0.8) for nanoMTZ, while for conventional MTZ, the lower mobility was obtained in corn and cassava residues (0.85-0.9) (Table S4). Mobility in different soil systems and soil with fresh organic residue addition is shown in Figure 5. Thin layer chromatography (TLC) using radiolabeled herbicides is commonly used to verify, qualitatively, the pesticide distribution in soil plates. Both formulations presented lower mobility in soil systems with high SOM content (NC, SC-CT, SC-NT, and SG-MN), compared with DS ( Figure 5). In soil with the addition of fresh organic residue, a residue line of 0.5 cm can retain the product and reduce the amount of herbicide distributed in the soil plate. However, it was not enough to reduce the mobility of metribuzin in either formulation ( Figure 6). ) and desorption (c,d) of nanoMTZ and conventional MTZ in soil amended with fresh organic residues. Formulations and residues were compared separately. Asterisks indicate differences between formulations in (a,c) (F-test, p < 0.05), and lowercase letters indicate differences between organic residues in (b,d) (Tukey's test, p < 0.05), separately. Boxes represents the 25-75% percentiles. The line inside the box indicates de mean, the bars indicates the data range and the circles outside the bars represents outliers. NanoMTZ and Conventional MTZ Soil Mobility The retention factor (Rf) for nanoMTZ and MTZ is shown in Supplementary Table S4. The Rf parameter aids the understanding of herbicide mobility in the soil; the mobility increases as Rf increases [33]. The Rf was between 0.45 and 0.98 for nanoMTZ and between 0.50-1.0 for conventional MTZ. Within soil systems, mobility was smaller for NC soil in both formulations (Rf = 0.45 and 0.5, for nanoMTZ and MTZ, respectively). The highest Rf was observed in DS for both formulations (Rf = 0.98-1.0) Among the other soil systems (SC-CT, SC-NT, and SG-MN) Rf ranged from 0.58 to 0.63 for nanoMTZ and 0.53-0.63 for MTZ (Table S4). In soil with organic residue addition, Rf values varied between 0.8-0.9 for nanoMTZ and 0.85-0.98 for MTZ. Within organic materials, Rf values were smaller in black oat straw and corn residue (0.8) for nanoMTZ, while for conventional MTZ, the lower mobility was obtained in corn and cassava residues (0.85-0.9) (Table S4). Mobility in different soil systems and soil with fresh organic residue addition is shown in Figure 5. Thin layer chromatography (TLC) using radiolabeled herbicides is commonly used to verify, qualitatively, the pesticide distribution in soil plates. Both formulations presented lower mobility in soil systems with high SOM content (NC, SC-CT, SC-NT, and SG-MN), compared with DS ( Figure 5). In soil with the addition of fresh organic residue, a residue line of 0.5 cm can retain the product and reduce the amount of herbicide distributed in the soil plate. However, it was not enough to reduce the mobility of metribuzin in either formulation ( Figure 6). Discussion Sorption is the most important process that defines pesticide behavior in the environment. This process can be described as an interaction between the compound (herbicide) and the sorbent (soil), which controls the bioavailability to plants and environmental fate [34]. The comprehension of this process will aid the proper application of pesticides [35]. Herein, we found low sorption for conventional MTZ and nanoMTZ. For herbicides, low retention can present two behavior scenarios, (i) greater availability of uptake by plants and other organisms responsible for degradation, and (ii) potential for the other fate process (transport to other environmental compartments/persistence determination). For nanoherbicides, information on the sorption of nanoparticles has been little explored, being an important environmental and plant control indicator to be studied. The MTZ herbicide is reported to be moderately mobile [36][37][38], as shown by the values described herein (Table 1), and by Mendes et al. [33], Guimarães et al. [39], and Rigi et al. [40]. Here, we showed that the sorption of this herbicide is mainly affected by SOM in different soils, but easily returns to the soil. Metribuzin presented a greater desorption process than sorption rates, becoming available for absorption and transport processes. This was also found by Mielke et al. [41], although availability is dependent on different biochar rates. Mendes et al. [33] indicate deep profile transport for metribuzin (distributed in 30 cm of soil profile), even with the application of organic material on top of the soil. This herbicide was found in the ground and surface water in different countries, presenting environmental risks [42][43][44]. However, when applied in a polymeric nanoformulation, metribuzin safety improved, based on reduced soil mobility in an irrigation field condition and lack of effect in the soil enzymatic activity [14,23]. Here, both formulations presented similar low retentions in different soil systems. MTZ has high solubility (Sw = 10.700 g L −1 ), low lipophilicity (log Kow = 1.75), and is a weak base herbicide (molecular form in pH above pKa = 1), all of which contribute to low sorption in the soil [45,46]. The pH and competition for the ionic sorption sites (CEC, presence of other ions) are not important for the retention of herbicides, such as metribuzin, in the molecular form, as demonstrated by the lack of correlation results ( Figure 3). In addition, protonation is not expected to be as important in metribuzin adsorption in the soil [46]. When nanoencapsulated, the formulation has high hydrophilicity due to the surfactant, which provides water affinity to the nanoparticles [47,48], and negative charge (−31 mV), which provides repulsion energy between soil colloids and nanoparticles. On the other hand, high clay content in the soil can offer more sites for the retention of nanoMTZ in the soil. Despite this, nanoMTZ showed similar low sorption in the soil compared to conventional MTZ. Nanoformulations based on PCL can improve the efficacy of the herbicide when applied as a pre-emergent [8,14], and soil availability is important to deliver the herbicide to the plants. Dyianat and Saedian [14] found higher retention for nanoMTZ than conventional MTZ (40 mm of rain). Nevertheless, Pereira et al. [13] indicated that PCL-atrazine nanoformulation was leached more in soil up to 8 cm in depth (~30%) than conventional atrazine (~20%), with 70 mm of simulated rain. The same formulation based on PCL was tested for MTZ, and conventional MTZ formulation in extreme rain conditions (200 mm in 48 h) presented a high concentration up to 20 cm in depth (~87%), with a great presence in the seed bank in the soil. On the other hand, conventional MTZ is a concern for water contamination, as indicated previously. Therefore, nanoformulation could be a safer alternative than conventional MTZ applications in the field. Pesticide nanoparticles have not previously been studied in soils of different cropping systems, mainly herbicides applied in the soil, such as MTZ. Here, we observed the increasing sorption potential with the amount of SOM, for both conventional and nanoformulations, in established cropping systems. Humic substances, from the organic matter degradation portion, have high reactivity, a large surface area, and variable composition, interacting with neutral or ionizable molecules [49,50], such as MTZ formulations. Even stable SOM compounds, which increase hydrophobicity in the soil, can sorb triazines and soluble herbicides such as metribuzin [41,51,52]. Therefore, despite the low sorption, soils with consolidated SOM can retain more metribuzin in the soil, in both formulations, reducing the risk of loss. This retention is reversible and can contribute to the absorption process, through the availability of herbicide in the soil. When the organic matter resource was fresh, the retention in the soil was different, being higher for nanoMTZ than conventional MTZ. Furthermore, desorption processes occurred more slowly for nanoMTZ than conventional MTZ in soil systems with added fresh organic matter. With the addition of organic residues in the soil, the interaction between pesticides and soil can be enhanced, mainly by the presence of dissolved organic matter, and consequently change the dynamics of pesticide and nanomaterial dispersion in the environment [53,54]. Dissolved organic matter (DOM) is more present in the soil with fresh materials; and is reduced during the humification process [55]. Nanoparticles can interact with DOM, which can coat nanomaterial surfaces and modify solubility and other properties [56][57][58]. However, it can increase the leaching of pesticides as carriers through soil solution [35]. Another aspect that can increase the effect of nanoformulation sorption is the porous entrapment. Some research has indicated the possibility of pesticide entrapment in biochar and porous soil organic matter [59][60][61]; structures such as nanoparticles in relation to active ingredient alone, can be retained in porous forms in the presence of organic material in the soil. This effect can result in reversible sorption and slow desorption. In addition to soil retention, short-distance mobility studies can aid understanding the availability of the herbicide to reach the seed bank and the risk of loss in the soil. The Helling and Turner [62] Rf classification considers five classes, with increasing mobility from classes 1 to 5. NanoMTZ and conventional MTZ for soil systems were considered slightly mobile according to class 3 (0.35-0.64), similar to other triazine herbicides. On the other hand, both formulations were affected by the presence of fresh organic matter and the Rf values were considered in mobile class 4 (0.65-0.89), according to less organic matter and clay content in the deep soil tested. In deep soil (less SOM) the Rf classification was high mobility (5 class, 0.9-1.0). These results show that fresh residues in poor soil (in relation to the soil systems) are not enough to reduce mobility in the soil. However, all straw types break the herbicide run on soil plates. Other authors also found organic matter as a barrier to triazine herbicides in the soil plates [33,63]. These results can help us understand the effect of organic matter under metribuzin in the soil, independently of the formulation. Even in soils poor in SOM and clay, fresh organic matter played an essential role in metribuzin containment. The combination between SOM retention and a fresh residue barrier to avoid losses can increase the safe application of nanoMTZ, associated with weed control efficiency for this PCL formulation [16,23]. The effect of fresh organic material, under nanoMTZ retention, can indicate nanoformulation interference in the retention in soil systems with recent organic matter input. However, the safe application depends on the amount of SOM accumulated after a consolidated system with continuous organic material input. In a system with organic matter addition, this management of the nanoformulation can present slow availability, prolonging the action of the active ingredient in the soil, and reducing losses in the environment. Maintenance of organic matter added in a consolidated soil system can promote the safe application of herbicides in general, mainly nanoherbicides such as nanoMTZ, due to enhanced retention of molecules in the soil with increased organic matter content. Furthermore, field studies can help us to understand the metribuzin dynamic associated with nanoparticles and consider environmental conditions, as well as the characteristics of soils tested in the laboratory. Preparation and Characterization of the Nanoformulation Metribuzin nanoformulation was prepared by the nanoprecipitation method, as published previously [23]. The organic phase consisted of poly-ε-caprolactone (PCL) (100 mg), dissolved in acetone (30 mL) under stirring (55 • C), and mixed under magnetic stirring, after complete PCL dissolution, with Myritol 318 (200 mg), Span 60 (40 mg), and metribuzin (MTZ) (10 mg). The aqueous phase (30 mL) was prepared with the addition of Tween 80 (60 mg), under magnetic stirring. The nanoparticles were formed with a mixture of the organic and aqueous phases, slowly, maintaining the magnetic stirring. The final volume was stirred for 20 min and concentrated in a rotary evaporator until 10 mL (0.4 mg mL −1 ). For the radiolabeled work solution, the 14 C-metribuzin (95% purity and 2.3 Bq mg −1 of specific activity) was added (1.66 × 10 6 Bq) together with non-radiolabeled MTZ in the organic phase, for the same final concentration (0.4 mg mL −1 and 319.6 Bq µL −1 ). The colloidal stability and encapsulation efficiency were indicated previously by Takeshita et al. [23]. Metribuzin nanoparticles were characterized by Takeshita et al. [23] and showed a hydrodynamic size of 289 ± 3 nm (using the DLS technique), zeta-potential of −31.6 ± 0.3, polydispersity index of 0.09 ± 0.02, and spherical morphology by Atomic Force Microscopy (AFM). For the conventional metribuzin comparison we used the Sencor ® 480 (480 g L −1 ), and for the soybean, the recommended dose of 480 g active ingredient (a.i) ha −1 . For the radiolabeled work solution, the 14 C-metribuzin was added together with non-radiolabeled MTZ to the solution, in a specific radioactivity amount for each assay, described later in each corresponding section. All nanoMTZ and conventional MTZ applications used the same dose recommendation. Soil Collection and Preparation The soil samples were collected from field areas, in different crop and non-crop systems, in the northern region of Paraná State, Brazil. All samples were collected after the removal of the surface vegetal residues until 0.2 m in depth. The soil samples corresponded to non-crop soil (NC) (22 • 58 17.7 S, 50 • 28 46.1 W) from an orchard of more than 15 years, cultivated soil with soybean-corn succession in a no-tillage system (SC-NT) (22 • 58 13.7 S, 50 • 28 24.2 W) for more than 5 years, cultivated soil with soybean-corn succession in a conventional tillage system (SC-CT) (22 • Table S5, and for information on fresh organic materials, see Supplementary Table S6. Sorption-Desorption Assay in Different Soil Systems For isotherm determination, the assay was completely randomized, in a 4 × 5 factorial design, with 4 soil (NC, SC-NT, SC-CT, and SG-MN) and 5 MTZ doses ( 1 4 D, 1 2 D, D, 2 D, and 4 D), with 2 replicates for each formulation (nanoMTZ and MTZ). All doses were based on the soybean recommendation (480 g a.i. ha −1 ) as the D dose. An additional treatment was used, only for apparent portion calculation, using DS soil, with the objective of elucidating the importance of OM present in the soil profile in the retention of the herbicide. Two replicates were used without soil, only with soil solution, to verify the absence of retention of the formulations in the Teflon tubes. The sorption-desorption assay was carried out based on the OECD guideline [64]. In a previous test of soil solution + soil centrifugation, we decided to use the proportion 1:2 (adsorbent: solution, m/v), considering the high clay content in all soils; at an equilibrium time of 24 h, as indicated by Takeshita et al. [23] for MTZ studies. The work solution was formed by radiolabeled nanoMTZ or conventional and non-radiolabeled MTZ, both formulations added to a CaCl 2 (0.01 M) solution, considering the amount of active ingredient per soil weight in a hectare, and the radioactivity necessary. In a Teflon tube (50 mL), soil content (10 g) was added to the soil solution (19 mL) without herbicide and shaken on the horizontal table (TE 140, Tecnal, Brazil), at 180 rpm for 12 h. Subsequently, 1 mL of work solution with herbicide was added, with shaking at 180 rpm, for 24 h. The final proportion used was 10 g of soil in 20 mL of the soil solution (500 Bq in each tube). For sorption data, all samples were centrifuged at 4500 rpm, for 15 min at 10 • C. Three technical replicates were collected from the supernatant and analyzed using a liquid scintillation counter (Tri-Carb 2910 TR LSA, PerkinElmer, Waltham, MA, USA), for 5 min. The sorption total was considered based on the initial radioactivity applied. After collecting the sorption aliquots, the supernatant was discarded, and the tubes were reweighed. Again, 20 mL of soil solution was added to the flasks, and the procedure was carried out in the same way as described for sorption evaluation. The total product desorbed was calculated in relation to the amount of product sorbed and the amount of product present in the supernatant. The sorption coefficient (Kd, mL g −1 ) was calculated using Equation (1): where Cs is the herbicide concentration (mg g −1 ) in the soil after equilibrium and Ce is the herbicide concentration (mg mL −1 ) in the soil solution after equilibrium. The sorption coefficient normalized by OC (Koc, mL g −1 ) was calculated using Equation (2): where Kd is the sorption coefficient (mL g −1 ) and OC is the organic carbon present in the soil (%). The Freundlich sorption coefficient (Kf, mL g −1 ) and 1/n (curve inclination) was calculated using Equation (3): where Cs is the herbicide concentration (mg g −1 ) in the soil after equilibrium and Ce is the herbicide concentration (mg mL −1 ) in the soil solution after equilibrium. Sorption-Desorption Assay as a Function of the Type and Amount of Fresh Organic Material in the Deep Soil Straw from different cover crops and organic residues was used in this study (forage turnip, black oat, sugarcane, corn waste, and cassava residue). For complete straw information, see Table S6 (Supplementary Material). The organic materials used were ground, homogenized, and sieved at 2 mm. The fraction used was smaller than 2 mm. These tests were carried out to observe the effect of fresh organic material on metribuzin retention, in a different way to the different crop system soils experiment. For sorption determination, the assay was completely randomized, with 5 organic residues and unamended soil, with 2 replicates, for each formulation (nanoMTZ and conventional MTZ). Two more replicates were used without soil, only with soil solution, to verify the absence of retention of the formulations in the Teflon tubes. All experimental units consisted of 10 g (soil + fresh organic material) and 20 mL of soil solution (1:2, m/v), as in the sorption study. The study was conducted in the same way as the retention study in different cropping systems, as described in the previous section. Soil Thin Layer Chromatography Assay For mobility information, soil thin layer chromatography was carried out (Soil-TLC), according to the US Environmental Protection Agency OPPTS 835.1210 [65]. The soil samples from different soil systems (NC, SC-NT, SC-CT, and SG-MC) and deep soil + fresh organic material strips in a standard soil were used in the soil-TCL plates (9 cm wide, 15 cm long, and 2 cm thick). The plates were made in duplicate, and 14 C-metribuzin from MTZ commercial formulation and nanoMTZ formulation was applied (833.33 Bq) in each plate, to visualize the herbicide mobility. The plates were gently placed in chromatographic chambers containing 100 mL of deionized water until the elution limit line (10 cm above application points) was complete. Subsequently, all plates were dried at room temperature for 48 h. Autoradiograph images were obtained using phosphorescent films, read in radio scanner equipment (Cyclone Plus Phosphor Imager, Model C431200, PerkinElmer Inc., Shelton, CT, USA). Rf (retention factor) values were calculated based on the distance traveled by the herbicide on the plate, using Equation (4): where Dh is the distance from baseline (application point) traveled by herbicide on the plate and Ds is the distance from baseline (application point) traveled by the solvent (water) on the plate (in this case, 10 cm). Statistical Analysis The sorption and desorption percentage data of soil systems, and sorption-desorption coefficient, adjusted or not for organic carbon, were submitted to an analysis of variance (ANOVA-two way), as well the sorption-desorption data of nanoMTZ and conventional MTZ with the addition of different fresh organic materials. When variables were significant, the data were submitted to Tukey's test or F's test (p < 0.05). Freundlich isotherms were calculated and plotted in Origin 2020 software (Version 9.7.0.185, OriginLab Corporation, Northampton, MA, USA), as were sorption-desorption figures (soil system and straw retention data). Pearson's correlation was obtained in the RStudio (RStudio 4.1.0 version, Boston USA), as were all statistical tests. Conclusions Our sorption-desorption and mobility experiments suggest low retention and potential mobility of nanoMTZ and conventional MTZ, as expected for the herbicide metribuzin. In consolidated crop systems, under conventional or no-tillage management, nanoformulation sorption was similar to conventional MTZ, with reversible availability. Sorption and desorption processes were correlated with soil organic matter and organic carbon content. All these soil systems presented similar characteristics, and the SOM was more influential in the retention process. Furthermore, the clay content is important to nanoformulation retention in the soil. Fresh organic residue addition in the deep soil (poorer soil systems) indicated better nanoMTZ sorption than conventional MTZ. However, TLC quality results showed that fresh residues in poor soil (in relation to the soil systems) are not enough to reduce mobility in the soil, where the organic residues only acted as a physical barrier on the ground. These results suggest that fresh input of organic matter is important to maintain the herbicide in the seed bank zone and reduce leaching risk (slow return to the soil solution), besides contributing to system maintenance. For safe applications of nanoMTZ in the soil system, continuous organic material addition is essential. Furthermore, we used radiometric techniques through classical protocols, with a focus on cultivation systems, in which nanoMTZ can be inserted early. We reinforce here the importance of testing conventional active ingredients in comparison to new nanoformulations, for better understanding. Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/plants11233366/s1, Table S1: Sorption-desorption parameters for nanoMTZ and MTZ in different soil systems. Data indicate mean ± standard error of the mean. As interaction within soil systems and formulation was not significant, lowercase letters indicate differences between soil systems and ns indicates no significance within formulation, by Tukey's test (p < 0.05); Table S2: Correlation matrices for nanoMTZ and MTZ between sorption-desorption processes and soil characteristics of different soil systems. The values correspond to Pearson's Correlation Factor (r). Numbers with ns represent non-significant correlation and those with * are significant, at 5% of significance (p < 0.05); Table S3: Sorption-desorption parameters for nanoMTZ and MTZ in soil with addition of different organic residues. Data indicate mean ± standard error of the mean. As interaction within soil systems and formulation was not significant, lowercase letters indicate differences between organic residues (regarding formulation) and uppercase letters indicate differences between formulation (regarding organic residues), by Tukey's test (p < 0.05); Table S4: Retention factor (Rf) to nanoMTZ and conventional MTZ soil mobility, obtained through soil thin layer chromatography (soil-TLC). Values represent the mean ± standard error; Table S5: Soil physicochemical properties from different crop and non-crop systems. The soil samples correspond to non-crop soil (NC), cultivated soil with soybean-corn succession in a no-tillage system (SC-NT), cultivated soil with soybean-corn succession in a conventional tillage system (SC-CT), sugarcane monoculture soil (SG-MN), and deep soil (DS) to fresh organic materials addition; Table S6: Physical-chemical properties of organic residues added to soil.	2022-12-04T23:06:18.391Z	2022-12-01T00:00:00.000	{ "year": 2022, "sha1": "ec5e5b563fb156337ea216413841c5297fed5df4", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/2223-7747/11/23/3366/pdf?version=1670203434", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "63771aa68ad8d3c8b6cf6ae41b4a0fbfeecac916", "s2fieldsofstudy": [ "Environmental Science", "Agricultural and Food Sciences", "Materials Science", "Chemistry" ], "extfieldsofstudy": [ "Medicine" ] }
172137869	pes2o/s2orc	v3-fos-license	Multifunctional Platforms Based on Graphene Oxide and Natural Products Background and objectives: In the last few years, graphene oxide has attracted much attention in biomedical applications due to its unique physico-chemical properties and can be used as a carrier for both hydrophilic and/or hydrophobic biomolecules. The purpose of this paper was to synthesize graphene oxide and to obtain multifunctional platforms based on graphene oxide as a nanocarrier loaded with few biologically active substances with anticancer, antimicrobial or anti-inflammatory properties such as gallic acid, caffeic acid, limonene and nutmeg and cembra pine essential oils. Materials and Methods: Graphene oxide was obtained according to the method developed by Hummers and further loaded with biologically active agents. The obtained platforms were characterized using FTIR, HPLC, TGA, SEM, TEM and Raman spectroscopy. Results: Gallic acid released 80% within 10 days but all the other biologically active agents did not release because their affinity for the graphene oxide support was higher than that of the phosphate buffer solution. SEM characterization showed the formation of nanosheets and a slight increase in the degree of agglomeration of the particles. The ratio I2D/IG for all samples was between 0.18 for GO-cembra pine and 0.27 for GO-limonene, indicating that the GO materials were in the form of multilayers. The individual GO sheets were found to have less than 20 µm, the thickness of GO was estimated to be ~4 nm and an interlayer spacing of about 2.12 Å. Raman spectroscopy indicated that the bioactive substances were adsorbed on the surface and no degradation occurred during loading. Conclusions: These findings encourage this research to further explore, both in vitro and in vivo, the biological activities of bioactive agents for their use in medicine. Introduction Carbon is an extraordinary element which can exist in different allotropic forms. It is found in large quantities in nature as coal or as natural graphite and in fewer quantities as diamond. The crystal structure of diamond has a cubic arrangement of the atoms compared to graphite which has a special kind of structure containing flat sheets of carbon atoms bonded into hexagonal structures. The later, has remarkable properties, with good thermal and electrical conductivity [1] and starting from these hexagonal structures new engineered materials can be obtained such as graphene and carbon nanotubes. Carbon nanomaterials contain many different allotropic forms of carbon, the most studied and used ones being carbon nanotubes and graphene. The remarkable properties of graphite ( Figure 1a) functional groups: epoxy, hydroxyl, carboxyl, carbonyl, etc.) allows it to incorporate hydrophilic and/or hydrophobic biomolecules onto the structure in order to improve its stability in a particular environment and to have minimal toxicity for biomedical applications [13,[27][28][29][30][31]. As the side effects of synthetic substances on the body are well known, lately there has been a great interest in using natural substances and in recent years great emphasis has been placed on the development of nanomaterials loaded with natural extracts [33]. There is an urgent need for controlled target delivery for cancer therapy, and synthesis under ecological conditions is important for the biomedical use of controlled release systems in the human body [34]. The antimicrobial, anticancer and antifungal effects of natural compounds extracted from different plants against various microorganisms have been demonstrated over time by researchers, which encourages us to further explore the biological activities of bioactive substances for their use in medicine [35][36][37][38][39]. Graphene oxide has an easily modified surface with a variety of biocompatible polymers such as chitosan, polyethylene glycol, poly-L-lysine and polyvinyl alcohol. Graphene oxide contains a large amount of marginal hydrophilic groups. These moieties make graphene oxide more attractive for researchers with applications in controlled release, parasitology, tissue engineering, antibacterial activity, cancer treatment, imaging, sensors and diagnostics. The most important applications of graphene oxide are shown in Figure 2. In order to use graphene oxide in clinical trials, it is essential to study its toxicity and biocompatibility through in vitro and in vivo experiments using specific cell lines and animal models [27,40]. Despite advances in technology in biomedical science, cancer still remains one of the greatest challenges for humanity. In this paper, we wanted to obtain a nanostructured platform for controlled release system using graphene oxide as a nanocarrier for several active agents with anticancer properties as well as antimicrobial and anti-inflammatory properties. The active compounds used were gallic acid, caffeic acid, limonene and two essential oils: nutmeg and cembra pine oil. As the side effects of synthetic substances on the body are well known, lately there has been a great interest in using natural substances and in recent years great emphasis has been placed on the development of nanomaterials loaded with natural extracts [33]. There is an urgent need for controlled target delivery for cancer therapy, and synthesis under ecological conditions is important for the biomedical use of controlled release systems in the human body [34]. The antimicrobial, anticancer and antifungal effects of natural compounds extracted from different plants against various microorganisms have been demonstrated over time by researchers, which encourages us to further explore the biological activities of bioactive substances for their use in medicine [35][36][37][38][39]. Graphene oxide has an easily modified surface with a variety of biocompatible polymers such as chitosan, polyethylene glycol, poly-L-lysine and polyvinyl alcohol. Graphene oxide contains a large amount of marginal hydrophilic groups. These moieties make graphene oxide more attractive for researchers with applications in controlled release, parasitology, tissue engineering, antibacterial activity, cancer treatment, imaging, sensors and diagnostics. The most important applications of graphene oxide are shown in Figure 2. In order to use graphene oxide in clinical trials, it is essential to study its toxicity and biocompatibility through in vitro and in vivo experiments using specific cell lines and animal models [27,40]. Despite advances in technology in biomedical science, cancer still remains one of the greatest challenges for humanity. In this paper, we wanted to obtain a nanostructured platform for controlled release system using graphene oxide as a nanocarrier for several active agents with anticancer properties as well as antimicrobial and anti-inflammatory properties. The active compounds used were gallic acid, caffeic acid, limonene and two essential oils: nutmeg and cembra pine oil. [19]. Synthesis of Graphene Oxide Graphene oxide was synthesized from graphite powder using a modified Hummers method presented in Scheme 1 [41]. In this method, a solution containing concentrated H2SO4 (60 mL), K2S2O8 (10 g) and P2O5 (10 g) was heated at 80 °C. Then, 20 g of graphite was added under stirring condition. The solution was left to cool down to room temperature. Then, a large amount of deionized water was added, and the solution was filtered, and the filtrate was rewashed until reached neutral pH. The washed product was dried for 24 h at 80 °C. Concentrated H2SO4 (460 mL) solution and 20 g of peroxidized graphite were added under stirring condition in a Berzelius flask in an ice bath (temperature below 5 °C). Sixty gram of KMnO4 was then added slowly into the solution and stirred until the solution became dark green. The solution was then transferred to a beaker, where it was stirred for 2 h at 35 °C. About 920 mL of distilled water was added to the mixture and stirred for 15 min, after which 2.8 L of distilled water and 50 mL of 30% H2O2 were added to eliminate the excess of KMnO4. The supernatant was decanted, and the residuals were then rewashed with 5% HCl and distilled H2O until a neutral pH was obtained. The washed GO suspension was dried at 60 °C for 24 h. Synthesis of Graphene Oxide Graphene oxide was synthesized from graphite powder using a modified Hummers method presented in Scheme 1 [41]. In this method, a solution containing concentrated H 2 SO 4 (60 mL), K 2 S 2 O 8 (10 g) and P 2 O 5 (10 g) was heated at 80 • C. Then, 20 g of graphite was added under stirring condition. The solution was left to cool down to room temperature. Then, a large amount of deionized water was added, and the solution was filtered, and the filtrate was rewashed until reached neutral pH. The washed product was dried for 24 h at 80 • C. Concentrated H 2 SO 4 (460 mL) solution and 20 g of peroxidized graphite were added under stirring condition in a Berzelius flask in an ice bath (temperature below 5 • C). Sixty gram of KMnO 4 was then added slowly into the solution and stirred until the solution became dark green. The solution was then transferred to a beaker, where it was stirred for 2 h at 35 • C. About 920 mL of distilled water was added to the mixture and stirred for 15 min, after which 2.8 L of distilled water and 50 mL of 30% H 2 O 2 were added to eliminate the excess of KMnO 4 . The supernatant was decanted, and the residuals were then rewashed with 5% HCl and distilled H 2 O until a neutral pH was obtained. The washed GO suspension was dried at 60 • C for 24 h. Phase I: Peroxidation of graphite Phase II: Oxidation of peroxidized graphite Scheme 1. Synthesis of graphene oxide. Loading of Biological Active Agent (BAA) In order to obtain graphene oxide-gallic acid (GOGA), a 1% solution of gallic acid was prepared. From this solution, three other solutions were made in a 10 mL volumetric flask, adding 0.1 g graphene oxide in each solution. The mixtures were stirred for 16 h and the GOGA nanomaterial was collected via filtration and dried in a vacuum oven at 40 °C for 24 h. The same method was used to load graphene oxide with caffeic acid, limonene, cembra pine and nutmeg essential oils (according to Table 1). The controlled release of the biologically active agents loaded onto the GO platforms was studied using a buffer solution of pH 7.4 (similar to the pH of blood). Scheme 1. Synthesis of graphene oxide. Loading of Biological Active Agent (BAA) In order to obtain graphene oxide-gallic acid (GOGA), a 1% solution of gallic acid was prepared. From this solution, three other solutions were made in a 10 mL volumetric flask, adding 0.1 g graphene oxide in each solution. The mixtures were stirred for 16 h and the GOGA nanomaterial was collected via filtration and dried in a vacuum oven at 40 • C for 24 h. The same method was used to load graphene oxide with caffeic acid, limonene, cembra pine and nutmeg essential oils (according to Table 1). The controlled release of the biologically active agents loaded onto the GO platforms was studied using a buffer solution of pH 7.4 (similar to the pH of blood). Table 1. Biologically active agents used in the development of the graphene oxide-based platforms. Active Substance Biological Activity Ref. HPLC analysis was conducted on Agilent 1260 Infinity liquid chromatograph (Agilent, Santa Clara, CA, USA) equipped with a DAD detector. The analytical column used was Eclipse Pluse C 18 (4.6 mm × 100 mm, 3.5 µm). Two-gradient elution system was used: mobile phase A contained water and 0.1% formic acid and mobile phase B was acetonitrile with 0.1% formic acid. The mobile phase composition was 50:50. Elution was performed at a solvent flow rate of 1 mL/min. The sample injection volume was 15 µL and peak detection was at 220 nm. The retention time was 0.85 min. The synthesized products were characterized by FTIR using a Nicolet iS50FT-IR (Nicolet, MA, USA) spectrometer equipped with a DTGS detector which provides information with a high sensitivity in the range of 4000 cm −1 and 100 cm −1 at a resolution of 4 cm −1 . Raman spectroscopy analyses were performed using a Horiba equipment (Labram HR Evolution, Pailaiseau, France) with an excitation wavelength of 514 nm and a 50X objective with a 10 s acquisition time. Thermogravimetric analysis was performed using an STA TG/DSC Netzsch Jupiter 449 • C equipment (Selb, Germany), with the temperature ranging between 25 and 900 • C in a dynamic atmosphere of 50 mL/min air with a heating rate of 10 K/min in an alumina crucible (Al 2 O 3 ). Scanning electron microscopy images were obtained using a high-resolution electron microscope equipped with a field emission electron source (FEI Inspect F50, Eindhoven, Nederlands) with a resolution of 1.2 nm at 30 kV and 3 nm at 1 kV (BSE). The transmission electron micrographs were obtained by using a Tecnai G 2 F30 S-TWIN high-resolution transmission electron microscope (HRTEM, ThermoFisher, Eindhoven, Nederlands) equipped with STEM with HAADF detector, EDX, PEELS, energy filter, GIF operated at 300 kV. The sample preparation was done as follows: a small amount of GO was placed into water in a centrifuge tube and sonicated for 15 min. After that, 10 microliters of dispersed sample were placed onto 400 mesh holey carbon-coated Cu grid and left to dry prior to the transmission electron microscopy (TEM) analysis. Zeta Potential was obtained using dynamic light scattering technique (DLS, DelsaMax Pro, Backman Coulter, Brea, CA, USA). The nanoparticle suspensions in ultrapure water were homogenized at room temperature using an ultrasonication probe for a period of 10 min. Analysis of Functional Groups of Graphene Oxide The Boehm method was used to analyze the content of functional groups in GO. Graphene oxide (0.1 g) was immersed into 100 mL of NaOH, NaHCO 3 and 0.1M Na 2 CO 3 separately. The base mixtures containing GO were magnetically stirred for 24 h and then filtered to separate GO sheets from solution. From the obtained extracts, 5 mL was titrated with 0.05 M HCl solution using phenolphthalein as an indicator. For proper interpretation of the results, blank samples were also performed without GO, having the same concentrations [56,57]. The concentration of various types of functional groups was calculated assuming that NaHCO 3 neutralized only the carboxylic groups, Na 2 CO 3 neutralizes the carboxylic groups and lactones and NaOH neutralizes all the carboxylic, phenolic and lactonic groups [58]. Results This report presents the analyses of the five nanostructured platforms obtained by loading graphene oxide with biologically active agents: gallic acid, caffeic acid, limonene, nutmeg and cembra pine essential oils. The content of carboxyl group was 0.45 meq/g, of phenolic groups 0.57 meq/g and of lactones 0.15 meq/g, the total number of functional groups was 1.17 meq/g. The content of phenolic groups was higher as they play an important role in the structure of the graphene oxide [58]. Zeta potential, an indicator of the stability of the sample, is around −37.36 mV, which shows a good stability of the suspension [59]. Figure 3 presents Fourier Transform Infrared Spectroscopy (FTIR) spectra for graphene oxide, gallic acid, caffeic acid and gallic and caffeic acid loaded onto GO nanocarrier. mesh holey carbon-coated Cu grid and left to dry prior to the transmission electron microscopy (TEM) analysis. Zeta Potential was obtained using dynamic light scattering technique (DLS, DelsaMax Pro, Backman Coulter, Brea, CA, USA). The nanoparticle suspensions in ultrapure water were homogenized at room temperature using an ultrasonication probe for a period of 10 min. Analysis of Functional Groups of Graphene Oxide The Boehm method was used to analyze the content of functional groups in GO. Graphene oxide (0.1 g) was immersed into 100 mL of NaOH, NaHCO3 and 0.1M Na2CO3 separately. The base mixtures containing GO were magnetically stirred for 24 h and then filtered to separate GO sheets from solution. From the obtained extracts, 5 mL was titrated with 0.05 M HCl solution using phenolphthalein as an indicator. For proper interpretation of the results, blank samples were also performed without GO, having the same concentrations [56,57]. The concentration of various types of functional groups was calculated assuming that NaHCO3 neutralized only the carboxylic groups, Na2CO3 neutralizes the carboxylic groups and lactones and NaOH neutralizes all the carboxylic, phenolic and lactonic groups [58]. Results This report presents the analyses of the five nanostructured platforms obtained by loading graphene oxide with biologically active agents: gallic acid, caffeic acid, limonene, nutmeg and cembra pine essential oils. The content of carboxyl group was 0.45 meq/g, of phenolic groups 0.57 meq/g and of lactones 0.15 meq/g, the total number of functional groups was 1.17 meq/g. The content of phenolic groups was higher as they play an important role in the structure of the graphene oxide [58]. Zeta potential, an indicator of the stability of the sample, is around −37.36 mV, which shows a good stability of the suspension [59]. Figure 3 presents Fourier Transform Infrared Spectroscopy (FTIR) spectra for graphene oxide, gallic acid, caffeic acid and gallic and caffeic acid loaded onto GO nanocarrier. Raman spectroscopy was used to analyze the disorder and defects in the structure of graphene oxide and functionalized graphene oxide ( Figure 6). Figure 5 presents FTIR spectra for graphene oxide, nutmeg and cembra pine essential oils and graphene oxide loaded with nutmeg and cembra pine essential oils. Raman spectroscopy was used to analyze the disorder and defects in the structure of graphene oxide and functionalized graphene oxide ( Figure 6). Raman spectroscopy was used to analyze the disorder and defects in the structure of graphene oxide and functionalized graphene oxide ( Figure 6). The release behavior of gallic acid loaded onto graphene oxide is presented in Figure 8. The release behavior of gallic acid loaded onto graphene oxide is presented in Figure 8. The release behavior of gallic acid loaded onto graphene oxide is presented in Figure 8. Discussion In several papers, these biological substances were studied for their benefits in the human body. Gallic acid has many biological applications such as antibacterial, antimutagenic, antiviral and Discussion In several papers, these biological substances were studied for their benefits in the human body. Gallic acid has many biological applications such as antibacterial, antimutagenic, antiviral and Discussion In several papers, these biological substances were studied for their benefits in the human body. Gallic acid has many biological applications such as antibacterial, antimutagenic, antiviral and antitumoral [43]. Based on the related research reports, caffeic acid has a variety of pharmacological effects, including anti-mutagenesis, anticancer, antibacterial, cardiovascular, anti-leukemic and immunomodulatory [46]. D-limonene is widely used in cosmetics, in the food industry, but also in pharmaceutical and medical applications. It has antioxidant, antimicrobial, anti-tumor, and antidiabetic properties [48,49]. Essential oils have traditionally been used for respiratory infections and are currently used as drugs for various diseases such as cardiovascular disease, diabetes, Alzheimer's and cancer. Furthermore, studies have demonstrated the synergistic effect of ingredients in essential oils against various human pathogens [51,52]. Fourier Transform Infrared Spectroscopy (FTIR) The resulting graphene oxide is water dispersible, the brownish aspect of the suspension is maintained even after 1 day, with only limited deposition after 1 day (~1g/L). As shown in Figure 3a, FTIR spectra of GO shows the characteristics peaks of graphene oxide. The peak from 1721 cm −1 is assigned to C=O stretching vibration present in the carbonyl and carboxyl groups of GO. The presence of the spectral band from 1600 cm −1 is attributed to C=C stretching vibration and the signals at 1224 cm −1 and 1029 cm −1 are corresponding to COC (epoxy) and COH (alcohol) stretching vibration groups. The broad peak at 3122 cm −1 is corresponding to the associated OH (hydroxyl) group from GO but also due to the adsorbed water [28,31]. Figure 3b shows characteristic bands of pure gallic acid, the signal at 1664 cm −1 indicates the presence of the phenolic group, and the signal at 1608 cm −1 is assigned to stretching vibration of C=C bonds of the aromatic ring. The stretching vibration of the carboxyl groups occurs at 1217 cm −1 and the signal at 1021 cm −1 is attributed to C-O stretching vibration of the carboxyl group. The peak at 734 cm −1 corresponds to the δ CC benzene ring vibrations. The signal at 3269 cm −1 corresponds to the OH stretching vibration. After the loading of graphene oxide with gallic acid, the characteristic bands of the components can be highlighted. The infrared spectrum of graphene oxide loaded with gallic acid (GOGA) shows the characteristic signals of both graphene oxide and gallic acid, suggesting the successful loading. Certainly, some of these bands being overlapped, instead of individual peaks a composed peak can be observed, the intensity is proportional with the sum of the individual peaks. As the peaks from 1715 cm −1 and 1580 cm −1 are characteristic to GO, in Figure 3c these peaks can be visualized but the relative intensity is lower. The signal at around 3154 cm −1 corresponds to the OH group from carboxylic and phenolic groups. By subtracting between the GOGA nanocomposite spectrum and the pure graphene oxide spectrum a visible shifting and broadening of the band around 1850 cm −1 can be observed, but also an increase in the relative intensity of the peak, also highlighting the interaction of the acid gallic with the graphene oxide, which act as a promising nanocarrier for this BAA. The same can be observed for the peak from 1030 cm −1 [43,60]. In Figure 3d, the stretching vibration of the CH groups can be highlighted at a wavelength of 2979 cm −1 . The strong intensity band from 1600 cm −1 is attributed to the stretching vibration of C=C bonds present in the carboxyl groups and the signal at 1448 cm −1 is attributed to C-C stretching vibration of the aromatic ring. The signals at 1214 cm −1 and 1071 cm −1 are corresponding to the stretching vibration of the C-OH bonds attached to the aromatic ring. The broad band between 3200 cm −1 and 3500 cm −1 corresponds to the vibration of the associated hydroxyl groups [61]. The two spectra of GO and GOCA are similar, the latter showing characteristics bands/signals of caffeic acid overlapped on the graphene oxide support. As shown in Figure 3e, differences are seen in the same areas of the spectrum, 1580 cm −1 and 1041 cm −1 indicating interactions between CA and GO. Figure 4b shows the characteristic bands of limonene. The bands from 1647 and 1677 cm −1 are assigned to the stretching vibration of C=C bonds of the aromatic ring in the vinyl group. The signals between 3010 cm −1 and 3081 cm −1 are assigned to the stretching vibration of the C-H bonds. The signals below 3000 cm −1 are generally associated with unsaturated groups. The two peaks at 1377 cm −1 and 1437 cm −1 are assigned to the bending CH 3 /CH 2 bonds, while the intense signal at 885 cm −1 may correspond to the CH=CH bending. Figure 4c shows the characteristic peaks of limonene, but these are shifted and broadening (especially the band from 1583 cm −1 ) due to the interaction between the support and the limonene. Also, a difference in relative intensity between the two peaks from 1583 cm −1 and 1721 cm −1 can be observed because of the condition of the characteristics peaks of limonene and GO [62]. Based on the literature data [63] the main chemical compounds from nutmeg essential oil are sabinene, α-pinene and myristicin. Figure 5b shows similar signals to those identified by other researchers, such as the peaks at 941 cm −1 (corresponding to the CH stretching vibration) and at 1197 cm −1 (corresponding to the CO stretching vibration). The bending vibration of CH 3 /CH 2 groups occurs at 1433 cm −1 . In addition, the signal at 3511 cm −1 corresponds to the associated hydroxyl group, and the signal at 2919 cm −1 is attributed to the stretching vibration of the C-H bonds. In Figure 5c, the nanocomposite contains characteristic signals of essential nutmeg oil which demonstrate the loading of certain compounds on graphene oxide nanocarrier [64,65]. The FTIR spectra of essential oils are similar when it comes to the loading of active substances onto the graphene nanocarrier, resulting in the successful formation of the complex [66]. Raman Spectroscopy The Raman spectrum ( Figure 6) for the graphene oxide is similar to the spectra presented in previous works [67][68][69]. The intensity ratio corresponding to D band and G band can be used as a measure of the disorder in graphene oxide this band being accentuated by the functional groups of oxygen in the material [70]. An increase of ration of the band's intensity indicates an increase of the disorder from the material. The D-band from graphene oxide is observed at 1356 cm −1 , while the G band is noticed at 1592 cm −1 [70,71]. Comparing the spectra, structural changes occurred during the loading of GO with the active compounds. The G band, specific for the sp 2 carbon forms, is observed at 1589 cm −1 for graphene oxide loaded with GA. The D band, specific for the sp 3 carbon forms can be seen at~1350 cm −1 . The 2D band appears in the 2800 cm −1 and 3200 cm −1 range. Raman spectra did not show important differences in the D and G band between GO and GO loaded with bioactive compounds spectra, [43,72,73] thereby the graphene structure has not been altered, the bioactive compounds being adsorbed on the surface and no additional disorder was induced. The intensity ratio (I D /I G ) for GO (0.97) is slightly larger than GO loaded with bioactive substances starting from 0.87 (GO-limonene) to 0.92 (GO-gallic acid) which suggests that the active compounds can repair the defects of graphene during functionalization [73]. In addition, from the Raman spectroscopy, the ratio between the 2D and G band is a method for determining the number of layers presented in the samples. The shape, position and intensity of the 2D band show the difference between single and multilayer graphene oxide. For single layer GO, the 2D band occurs as a single peak and the ratio will be seen to be equal to 2. An increase in the number of layers reduces the intensities of the 2D peaks. In the spectra of the analyzed materials, the ratio I 2D /I G for all samples is between 0.18 for GO-cembra pine and 0.27 for GO-limonene, these results indicating that GO materials are in the form of multilayers [74,75]. Thermogravimetric Analysis of Graphene Oxide Loaded with Active Compounds All samples show a loss of mass at the beginning around 12%-13%, except for samples with gallic acid and limonene which have a mass loss of about 16%. This may be due to the fact that graphene oxide can easily absorb water and therefore when heated will lose it, together with other volatile compounds that are eliminated in this phase (RT-145 • C). For graphene oxide (Figure 7a and Figure S1) the mass loss is 13.06%. Oxidation of the sample occurs suddenly (almost explosively), vigorously at~155 • C, when the sample loses 60.05% by mass (in the range 145-165 • C) as indicated by the very sharp exothermic effect with the maximum at 155.3 • C. This process can be attributed to the loss of labile oxygen groups like epoxy or hydroxyl. Pyrolysis is specific to graphene oxide even if the atmosphere is inert because it is made with the oxygen contained in it. Then there are two slower oxidation steps, 165-460 • C, with a mass loss of 6.27% and 460-660 • C with a mass loss of 14.75%. These oxidations are accompanied by weak exothermic effects with a maximum at 189, 517.7, 604.4 and 631.6 • C respectively. The effect from 189 • C and the slow mass loss between 165-460 • C represent the removal of some slightly more stable groups like carbonyl or quinine [76,77]. After 500 • C, the burning of the carbon skeleton starts. For all samples, the analysis indicates the existence of a residual mass, in this case of 5.91% [78][79][80]. Thermal analysis for gallic acid sample (Figure 7b and Figure S2) indicates a more significant mass loss in the range of 30-145 • C (16.23%), similar to what has been observed so far, but more intense, mass loss being important even from 30 • C, the process is accompanied by a weak endotherm effect. This indicates a larger quantity of absorbed water in the sample due to the presence of gallic acid. Between 145 and 215 • C, an exothermic process takes place at a maximum of 186.6 • C, the associated mass loss being 20.33%. This process is less exothermic, and the corresponding mass loss is only one third from that of GO sample, indicating a chemical interaction between gallic acid and GO, not only a physical absorption. At least some of the labile oxygen groups are interacting with gallic acid molecules by condensation or hydrogen bonding and cannot be removed easily anymore. The sample then slowly loses mass up to 365 • C (6.92%) due to more stable oxygen groups found in GO. The simple gallic acid should melt at 256-260 • C when it is losing one molecule of water. However, there is no noticeable thermal effect in this region, confirming the fact that gallic acid is chemically bonded by GO and therefore its stability is enhanced [81]. In the range of 365 and 560 • C complete oxidation takes place, the mass loss being of 50.18%, the process is accompanied by a strong exothermic effect, divided into two, with a maximum at 419.5 and 464.5 • C, respectively. At this stage, there are at least two chained oxidation processes partially overlapped, most probably oxidation of the gallic acid residue and the oxidation of the carbon skeleton of GO [82]. Figure 7c ( Figure S3) indicates that caffeic acid loaded into graphene oxide shows a slow weight loss (12.29%) between room temperature (RT) and 180 • C, like graphene oxide sample, the process is accompanied by a weak endothermic effect. Between 180 and 240 • C, an exothermic process (oxidation) takes place at a maximum of 221.3 • C, the associated weight loss being 22.4%. The sample then slowly losses mass up to 460 • C (11.27%), without any noticeable effect on the DSC curve. In the range of 460 and 640 • C complete oxidation takes place, the mass loss being 48.52%, the process is accompanied by a strong, asymmetric, exothermic effect, with the maximum at 554.1 • C. Overall the analysis is similar to the previous sample GOGA, the stabilization effect from the caffeic acid-GO bonds being slightly higher. The thermal analysis for limonene sample (Figure 7d and Figure S4) indicates a mass loss of 16.01% in the range of 30 and 180 • C, similar to that observed in the GO samples, the process is accompanied by a weak endothermic effect. As limonene is an aliphatic hydrocarbon, it cannot generate strong bonds with GO, and therefore this sample behaves almost like a simple GO. Between 180 and 225 • C, a violent exothermic process took place with a maximum of 213.5 • C, the associated mass loss being of 65.03%. The sample is then relatively stable up to~400 • C when the oxidation process of carbon skeleton starts. Up to 580 • C, 15.19% of the mass is lost, the process is accompanied by an exothermic split effect with maximums at 509.5 and 542.8 • C, respectively, indicating the existence of at least two partially overlapped oxidation processes. Thermal analysis for essential nutmeg oil sample (Figure 7e and Figure S5) shows a slow weight loss (12.53%) between RT and 175 • C, due to the elimination of water adsorbed in the sample, as well as other volatile molecules, similar to that seen with simple graphene oxide and the rest of the samples, the process is accompanied by a weak endothermic effect. Between 175 and 240 • C, there is an exothermic effect with a maximum at 220.1 • C, the associated mass loss being 23.25%. The main component of nutmeg oil is terpinen-4-ol which can condensate and generate hydrogen bonds with GO, therefore the thermal behavior of this sample being similar to those with gallic and caffeic acids. The sample slowly loses the mass up to 490 • C (12.61%). Between 490 and 640 • C, complete oxidation takes place, the mass loss being 45.97%, the process is accompanied by a strong, broad, exothermic effect, with the maximum at 590.7 • C. The sample loaded with cembra pine essential oil has a slow weight loss (12.35%) between RT and 180 • C, the process is accompanied by a weak endothermic effect, as shown in Figure 7f ( Figure S6). Between 180 and 250 • C there is an exothermic process (oxidation) with a maximum of 231.5 • C, the associated mass loss being 23.54%. The main components of cembra pine essential oil are α-terpineol and some cyclic terpene alcohols. Therefore, the sample will behave similar to gallic acid-GO and caffeic acid-GO samples, as the bonds formed between essential oil and GO are by condensation and hydrogen bonds. The sample slowly loses mass up to 490 • C (12.06%) without any noticeable effect on the DSC curve. In the range of 490 and 640 • C complete oxidation takes place, with 45.84% mass loss, the process is accompanied by a strong exothermic effect, with the maximum at 588.6 • C. As a conclusion, all samples present a first mass loss up to 145-180 • C due to the elimination of adsorbed water and some volatile molecules. A second mass loss occurs up to 165 • C for GO sample or up to 215-250 • C for the other samples and represents the loss of labile oxygen groups. The GO and limonene-GO sample undergo an energic oxidation process in this region, with a mass loss of at least 60%, indicating little chemical modification by limonene. Nevertheless, the limonene load on GO has the effect of increasing the thermal stability, the violent oxidation process taking place at 155 • C in the case of simple GO sample, but at 213 • C in the case of the limonene-GO sample. We can also observe that the limonene sample has no mass loss till 400 • C while GO sample presents a slow mass loss after 165 • C, up to 460 • C. Therefore, we can state that while limonene stabilizes the GO up to 213 • C, it will also generate the elimination of all GO oxygen groups in only one process. The gallic acid, caffeic acid, nutmeg and cembra pine essential oil samples contain molecules with -OH moiety and therefore can form hydrogen bonds and can condensate with oxygen groups from GO. The second mass loss process for these samples only represent around 20%-23%, and the exothermic effect is broader and less intense. It can be attributed to a slow oxidation process rather than an explosive one. The last mass loss for all samples is the burning of the carbon skeleton of GO and oxidation of carbonaceous residue from the acids and oils loaded on GO. The processes take place over 400 • C usually and can be overlapped partially or totally giving a broad, intense, sometimes split exothermic effect. For the samples GO and limonene GO where large quantities of carbon were oxidized at a lower temperature, the thermal effect generated by burning of carbon skeleton is less intense. Release Behavior of Active Compounds In comparison to gallic acid release behavior, the release rate of the other biologically active agents (caffeic acid, limonene, nutmeg and cembra pine essential oils) loaded onto graphene oxide is immaterial due to very low solubility and the content of the active substance released in the solution is below the limit of detection of used method [43,83]. The delivery of the gallic acid is quite fast on the first day (a burst delivery) followed by a strong decrease in the delivery for the next days. The cumulative release of gallic acid is about 80% after 10 days and practically is not changed in the next 20 days. These results have potential for GOGA platforms to be used as a drug delivery system having the ability to assure sustained release for over 10 days, as shown in Figure 8 [43]. Scanning Electron Microscopy (SEM) The structure of the graphene oxide appeared in very thin agglomerates sheets, as can be seen at 1000X magnification in Figure 9. At higher magnifications, it can observe the waved and folded shape of the graphene oxide, the results being similar to those obtained and presented in other studies in the literature [41,84,85]. The agglomerates of graphene oxide visualized at 1000X exhibit a size up to a hundred micrometers and, at increasing magnifications the lamellar morphology of graphene oxide can be observed. Due to the high degree of functionalization, the characteristic sheets are not planar. As GO has a lamellar layer structure it is also possible to measure the width of the GO sample using SEM analysis. The individual GO sheets were found to have less than 20 µm being in good agreement with the literature [86]. The morphology of graphene oxide loaded with active compounds (Figure 10) does not show important changes, having the same general aspect as the graphene oxide sample because of the relatively low content of the loaded agents as well as due to the low gluing capacity of these agents. However, a slight increase in the degree of agglomeration of the particles due to the presence of the active compounds onto the graphene oxide support can be observed, giving them a slightly irregular and enlarged form. Transmission Electron Microscopy (TEM) TEM images showed GO flakes quite flat because of the suspension in a liquid. However, on the TEM grid were still found agglomeration of layers of different sizes, the thickness of the graphene oxide flakes being estimated to be~4 nm. These findings are in accordance with the literature data [86,87]. The sheet with higher transparency indicates much thinner films of a few layers of GO due to a better exfoliation [88]. In order to explore the interlayer spacing of GO sheets, TEM analyses showed ( Figure 11b) an interlayer spacing of investigated area about 2.12 Å (0.212 nm) indicating the presence of oxygen-containing functional groups (lower 2θ in GO than graphite) and are in agreement with the literature. Moreover, the transversal HRTEM image, Figure 11c, reveals a graphene oxide sheet of~4.2 nm in thickness with a highly disordered surface arrangement of the atoms while the inner distribution of the atoms is much better because the surface oxidation is much higher [86,89]. Even if the arrangement of the atoms is not regular in cross-section view, the number of layers can be estimated to be between 10 and 20 layers. Conclusions The aim of the present work was to obtain graphene oxide-based drug delivery systems by loading caffeic acid, gallic acid, limonene, nutmeg and cembra pin essential oils starting from the corresponding solution. The SEM characterization showed the morphology of the graphene oxide support, which, compared to the loaded samples, a slight increase in the degree of agglomeration of the particles was observed due to the presence of the active compounds onto the graphene oxide samples. Raman spectra showed that no degradation occurred during loading. The ratio between I 2D /I G for all samples was between 0.18 for GO-cembra pine and 0.27 for GO-limonene; these results indicating that GO materials are in the form of multilayers. The individual GO sheets were found to have less than 20 µm. The thickness of GO flakes was estimated to be~4 nm with an interlayer spacing of about 2.12 Å. According to the solubility of loaded biologically active agents, the release behavior of the selected biologically active agents was as follows: gallic acid release within 10 days can reach 80% but all the other biologically active agents are practically not released because their affinity for the graphene oxide support is higher than that to phosphate buffer solution. Further analyses are necessary to make biological assessments, both in vitro and in vivo. This results can be exploited accordingly, the systems which are not able to release the biologically active agents being suitable for certain applications (such as preventing infections, cancer spreading inside the grafts containing graphene oxide, etc.) while the systems releasing biologically active agents can be exploited in a classical way, as a platform for delivery of biologically active agents (especially in the treatment of cancer and severe infections). All these results encourage research towards the next level by carrying out in vitro and in vivo studies to demonstrate the ability of these platforms in the treatment of cancer and severe infections and in development of drug delivery systems and optical biosensing. Author Contributions: This article was written through the contribution of all authors. All authors have given approval to the final version of the manuscript. Funding: This research was funded by the Academy of Romanian Scientists	2019-06-02T13:02:17.732Z	2019-05-30T00:00:00.000	{ "year": 2019, "sha1": "cece3e7807865f52a70f19284e5ad6dc40c435ab", "oa_license": "CCBY", "oa_url": "https://www.mdpi.com/1648-9144/55/6/230/pdf", "oa_status": "GOLD", "pdf_src": "PubMedCentral", "pdf_hash": "cece3e7807865f52a70f19284e5ad6dc40c435ab", "s2fieldsofstudy": [ "Medicine", "Materials Science", "Chemistry" ], "extfieldsofstudy": [ "Medicine", "Chemistry" ] }

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.